CN116740765A - Livestock sorting method, livestock sorting system and related devices - Google Patents
Livestock sorting method, livestock sorting system and related devices Download PDFInfo
- Publication number
- CN116740765A CN116740765A CN202310732516.4A CN202310732516A CN116740765A CN 116740765 A CN116740765 A CN 116740765A CN 202310732516 A CN202310732516 A CN 202310732516A CN 116740765 A CN116740765 A CN 116740765A
- Authority
- CN
- China
- Prior art keywords
- livestock
- target
- information
- image
- sorting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 244000144972 livestock Species 0.000 title claims abstract description 661
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000011161 development Methods 0.000 claims abstract description 72
- 238000001514 detection method Methods 0.000 claims abstract description 63
- 238000004590 computer program Methods 0.000 claims abstract description 44
- 230000036541 health Effects 0.000 claims abstract description 39
- 238000003860 storage Methods 0.000 claims abstract description 20
- 241001465754 Metazoa Species 0.000 claims description 117
- 238000012545 processing Methods 0.000 claims description 45
- 230000000295 complement effect Effects 0.000 claims description 44
- 238000005303 weighing Methods 0.000 claims description 33
- 210000001624 hip Anatomy 0.000 claims description 24
- 230000001815 facial effect Effects 0.000 claims description 22
- 238000001914 filtration Methods 0.000 claims description 21
- 230000003862 health status Effects 0.000 claims description 20
- 210000000689 upper leg Anatomy 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 13
- 210000002414 leg Anatomy 0.000 claims description 12
- 230000018109 developmental process Effects 0.000 description 68
- 241000282887 Suidae Species 0.000 description 36
- 238000012549 training Methods 0.000 description 32
- 241000282898 Sus scrofa Species 0.000 description 24
- 239000011159 matrix material Substances 0.000 description 15
- 230000008569 process Effects 0.000 description 15
- 201000010099 disease Diseases 0.000 description 14
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 14
- 238000004458 analytical method Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 238000013136 deep learning model Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 9
- 238000007726 management method Methods 0.000 description 9
- 241000283690 Bos taurus Species 0.000 description 7
- 210000001217 buttock Anatomy 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 7
- 210000003491 skin Anatomy 0.000 description 7
- 241001494479 Pecora Species 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 6
- 210000001508 eye Anatomy 0.000 description 6
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 238000012795 verification Methods 0.000 description 5
- 230000012173 estrus Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000001502 supplementing effect Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000009395 breeding Methods 0.000 description 3
- 230000001488 breeding effect Effects 0.000 description 3
- 125000004122 cyclic group Chemical group 0.000 description 3
- 230000001419 dependent effect Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 210000005069 ears Anatomy 0.000 description 3
- 230000001788 irregular Effects 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 210000001331 nose Anatomy 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 206010063409 Acarodermatitis Diseases 0.000 description 2
- 241000272517 Anseriformes Species 0.000 description 2
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 description 2
- 241000283074 Equus asinus Species 0.000 description 2
- 241001331845 Equus asinus x caballus Species 0.000 description 2
- 241000287828 Gallus gallus Species 0.000 description 2
- 241000282412 Homo Species 0.000 description 2
- 241000447727 Scabies Species 0.000 description 2
- 208000035472 Zoonoses Diseases 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000005452 bending Methods 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 210000000038 chest Anatomy 0.000 description 2
- 235000013330 chicken meat Nutrition 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 235000013365 dairy product Nutrition 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000013523 data management Methods 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 230000006806 disease prevention Effects 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 235000013305 food Nutrition 0.000 description 2
- 230000008014 freezing Effects 0.000 description 2
- 238000007710 freezing Methods 0.000 description 2
- 230000005802 health problem Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 210000003205 muscle Anatomy 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 208000005687 scabies Diseases 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 206010048282 zoonosis Diseases 0.000 description 2
- 241000272525 Anas platyrhynchos Species 0.000 description 1
- 241000272814 Anser sp. Species 0.000 description 1
- 241000283707 Capra Species 0.000 description 1
- 241000283086 Equidae Species 0.000 description 1
- 241000283073 Equus caballus Species 0.000 description 1
- 208000024949 Swine Erysipelas Diseases 0.000 description 1
- 208000002474 Tinea Diseases 0.000 description 1
- 241000893966 Trichophyton verrucosum Species 0.000 description 1
- 235000015278 beef Nutrition 0.000 description 1
- 230000037396 body weight Effects 0.000 description 1
- 244000309466 calf Species 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 244000078703 ectoparasite Species 0.000 description 1
- 235000013601 eggs Nutrition 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 210000003608 fece Anatomy 0.000 description 1
- 210000003194 forelimb Anatomy 0.000 description 1
- 230000037308 hair color Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 239000010985 leather Substances 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 210000003141 lower extremity Anatomy 0.000 description 1
- 235000013372 meat Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003387 muscular Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000004549 pulsed laser deposition Methods 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 210000001991 scapula Anatomy 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000003307 slaughter Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 230000004936 stimulating effect Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
The application provides a livestock sorting method, a livestock sorting system, an electronic device, a computer readable storage medium and a computer program product, wherein the method comprises the following steps: when a target livestock enters a detection area, acquiring a livestock image of the target livestock; acquiring livestock information of the target livestock according to the livestock image; acquiring sorting information corresponding to the target livestock based on the livestock information; the target livestock are sorted to a sorting area corresponding to the sorting information. The application can solve the problems that the livestock sorting quality is poor because the livestock is only sorted according to the weight of the livestock and the influence of the age, sex, development condition, health condition and other factors of the livestock cannot be comprehensively considered.
Description
Technical Field
The present application relates to the field of livestock sorting technology, and in particular to a method for sorting livestock, a system for sorting livestock, an electronic device, a computer readable storage medium and a computer program product.
Background
With the social development, livestock breeding is changed from bulk breeding to large-scale, intensive, automatic and intelligent breeding. The livestock sorting system is an automatic system for sorting and sorting livestock (such as cattle, pigs, sheep and the like) and aims to improve the sorting efficiency of the livestock and reduce manual operations. However, the existing livestock sorting system is only a single control system based on electromechanical technology and/or pneumatic technology, and sorting and feeding control management of livestock is performed by the single index according to the weight of the livestock. Because the livestock are sorted according to the weight of the livestock, the influence of the age, sex, development condition, health condition and other factors of the livestock on the livestock cannot be comprehensively considered, and the problem of poor livestock sorting quality exists.
Based on this, the present application provides a method of sorting livestock, a system of sorting livestock, an electronic device, a computer readable storage medium and a computer program product to improve the prior art.
Disclosure of Invention
The application aims to provide an animal sorting method, an animal sorting system, electronic equipment, a computer readable storage medium and a computer program product, which can solve the problem that the sorting quality of animals is poor when sorting is performed only according to the weight of the animals.
The application adopts the following technical scheme:
in a first aspect, the present application provides a method of livestock sorting, the method comprising:
when a target livestock enters a detection area, acquiring a livestock image of the target livestock;
acquiring livestock information of the target livestock according to the livestock image;
acquiring sorting information corresponding to the target livestock based on the livestock information;
the target livestock are sorted to a sorting area corresponding to the sorting information.
The beneficial effect of this technical scheme lies in: detecting whether the target livestock enters the designated detection area, acquiring a livestock image of the target livestock once the target livestock is detected to enter the detection area, and extracting livestock information of the target livestock, such as variety, livestock characteristics, health condition, and the like, by performing image processing and analysis on the acquired livestock image. Based on the livestock information of the target livestock, sorting information corresponding to the target livestock is acquired, the difference in the livestock information may cause the difference in the sorting information corresponding to the target livestock, the target livestock is sorted to a sorting area corresponding to the sorting information according to the acquired sorting information, for example, a gate of the sorting area is opened, and the target livestock is guided into the sorting area. On the one hand, the automatic and intelligent sorting of the target livestock is realized, manual intervention is not needed, the working efficiency is improved, the labor cost is reduced, and the risk of human errors is reduced. On the other hand, by acquiring the animal image of the target animal to obtain the animal information of the target animal, and sorting the target animal based on the animal information, it is helpful to ensure that the target animal is properly sorted. On the other hand, the target livestock is sorted according to the livestock information of the target livestock, the livestock information not only comprises the weight information of the target livestock, and the problem that the sorting quality of the livestock is poor due to the fact that the livestock is sorted according to the weight of the livestock and the influence of factors such as age, sex, development condition and health condition of the livestock cannot be comprehensively considered is solved.
In conclusion, through steps of image processing, information analysis, automatic sorting and the like, accurate sorting of target livestock is achieved, so that sorting efficiency is improved, and sorting errors are reduced.
In some optional embodiments, the acquiring the livestock information of the target livestock according to the livestock image includes:
and performing image processing on the livestock image to identify and obtain the health state information of the target livestock.
The beneficial effect of this technical scheme lies in: the obtained livestock image is subjected to image processing, such as image enhancement, edge detection, feature extraction and other technologies, so that features and information related to the livestock health state in the livestock image can be extracted. Based on the processing results of the livestock images, algorithms such as machine learning, computer vision, and the like can be utilized to identify health status information of the target livestock. For example, the health condition of livestock is judged by analyzing the characteristics of the livestock such as the body state, expression, skin color, skin surface state, eye state and the like, such as whether the livestock suffers from diseases, is in estrus or not, and the like. And according to the health state information of the target livestock, acquiring sorting information corresponding to the target livestock, for example, sorting the livestock with diseases to a specific area for further processing, or sorting the livestock in estrus to an area related to reproduction. Based on the sorting information, it is determined that the target livestock should be sorted to the corresponding sorting area, which is achieved by predefined rules and logic, ensuring that the target livestock are properly sorted and handled. On the one hand, the health state information of the target livestock is accurately identified by processing and analyzing the livestock images, so that the diseased livestock or the livestock with health problems can be found early, and corresponding processing measures can be taken. On the other hand, customized sorting schemes are provided for each animal based on the health status information of the target animal, which helps to ensure that the animals are sorted to the proper sorting area for corresponding treatment, improving sorting efficiency and quality. In yet another aspect, data-based decision support is provided through the use of image processing and analysis techniques, which helps reduce interference with subjective judgment, providing more objective and accurate sorting decisions. On the other hand, by automatically acquiring the health state information and the corresponding sorting information of the target livestock, the automatic and intelligent livestock sorting process is realized, the production efficiency can be improved, the labor cost can be reduced, the incidence of zoonotic diseases can be reduced, and the livestock can be ensured to be correctly processed and sorted.
In some alternative embodiments, the method further comprises:
acquiring real-time weight information of the target livestock;
acquiring an animal identification of the target animal based on the animal image;
acquiring historical weight information of the target livestock according to the livestock identification;
acquiring development state information of the target livestock based on the real-time weight information and the historical weight information;
the acquiring sorting information corresponding to the target livestock according to the livestock information comprises the following steps:
and determining sorting information corresponding to the target livestock based on the health state information and the development state information.
The beneficial effect of this technical scheme lies in: weight information of the target livestock is acquired in real time by using the livestock weighing assembly. To provide accurate data of the current weight status of the target livestock. By processing and analyzing the target livestock image, the livestock identification of the target livestock, such as identification information of numbers, names, and the like, is identified and extracted. Acquiring historical weight information of the target livestock according to the livestock identification of the target livestock by inquiring a database or an information system, wherein the historical weight information records past weight information of the target livestock; the weight growth trend and speed of the target livestock are calculated by comparing the real-time weight of the target livestock with the historical weight data thereof, so that the development state information of the target livestock is determined. For example, it may be determined that the target livestock is developing normally, growing too fast or growing too slow, etc. In combination with the health status information and the development status information of the target livestock, a specific area or process flow to which the target livestock should be sorted can be determined. On the one hand, the real-time weight information and the historical weight information of the target livestock are obtained to obtain the development state information of the target livestock, so that more comprehensive analysis and judgment are performed, more accurate and personalized sorting schemes are facilitated to be formulated, and the requirements and demands of the target livestock are met to the greatest extent. On the other hand, finer sorting classification is realized according to the real-time weight information and the historical weight information of the target livestock, and reasonable division of the livestock according to the weight and the development state of the livestock is facilitated, so that the precision and the effect of feeding management are improved. In yet another aspect, the selection of a feeding environment, i.e., a sorting area, suitable for the growth and development needs of the target livestock based on the development status information of the target livestock helps to provide good feeding conditions and promote healthy growth and production benefits of the livestock. On the other hand, based on the same livestock image, the health state information of the target livestock can be determined, and the livestock identification of the target livestock can be determined, so that the development state information of the target livestock is determined, multiplexing of the livestock image is realized, the whole image processing data amount is reduced, and the sorting efficiency is improved.
In some optional embodiments, the acquiring the development status information of the target livestock based on the real-time weight information and the historical weight information includes:
acquiring a preset weight corresponding relation;
and acquiring the development state information of the target livestock based on the real-time weight information, the historical weight information and the weight correspondence.
The beneficial effect of this technical scheme lies in: and acquiring a preset weight corresponding relation, setting a group of weight corresponding relation, and converting the weight of the livestock into specific development state information, for example, establishing a relation model or a mathematical model of a weight range and a growth stage. And acquiring real-time weight information of the target livestock by means of weighing equipment or sensors and the like. And meanwhile, acquiring historical weight information of the target livestock by inquiring a database or an information system, and comparing and matching the real-time weight information and the historical weight information with development state information by utilizing a preset weight corresponding relation. The information of the development state of the target livestock, such as the growth stage, the development degree and the like, is determined through calculation, comparison or interpolation and the like. On the one hand, the development state of the target livestock is accurately estimated by using a preset weight corresponding relation. The weight is an important index of the growth and development of livestock, and by correlating the weight with the development status, the growth stage and development level of the target livestock can be more accurately judged. On the other hand, through obtaining the development state information of the target livestock, personalized management and adjustment are carried out on the livestock at different development stages so as to meet the requirements of the livestock at different development states.
In some optional embodiments, the acquiring the animal identification of the target animal based on the animal image includes:
detecting whether the livestock characteristics of the target livestock in the livestock image meet preset image conditions or not; the livestock characteristics include one or more of facial characteristics, neck characteristics, ear characteristics, shoulder characteristics, back characteristics, chest characteristics, waist characteristics, hip characteristics, thigh characteristics, leg characteristics, tail characteristics, and arm characteristics of the target livestock; the preset image condition comprises at least one of feature definition and feature integrity;
if not, inputting the livestock image into a preset image complement model to obtain a complement image corresponding to the livestock image;
and acquiring the livestock identification of the target livestock based on the complement image.
The beneficial effect of this technical scheme lies in: the method comprises the steps of analyzing livestock characteristics of target livestock in livestock images, and detecting whether the definition and/or the integrity of the livestock characteristics meet preset image conditions, such as the definition (such as resolution, contrast and the like) of facial characteristics, neck characteristics, ear characteristics, shoulder characteristics, back characteristics, chest characteristics, waist characteristics, hip characteristics, thigh characteristics, leg characteristics, tail characteristics or arm characteristics and the integrity of facial characteristics, neck characteristics, ear characteristics, shoulder characteristics, back characteristics, chest characteristics, waist characteristics, hip characteristics, thigh characteristics, leg characteristics, tail characteristics or arm characteristics. If the characteristics of the livestock do not meet the preset image conditions, inputting the livestock image into a preset image complement model. And performing image reconstruction or complementation operation according to the existing image information and part of the characteristics of the livestock, and generating a complementation image of the characteristics of the target livestock. Based on the obtained complement image, identification information of the livestock, such as identification information of numbers, names and the like, is determined through image processing and recognition technology. The identification information is used for identity confirmation of the target livestock. On the one hand, the accuracy of the livestock identification is improved by collecting the livestock images facing the advancing direction of the target livestock and according to the preset image conditions and the image complement model. Even if the livestock features are incomplete or have low definition, the complement image of the livestock features with legibility can be generated, so that the acquisition rate and accuracy of the livestock identification are improved. On the other hand, by utilizing the image processing and pattern recognition technology, the automatic and intelligent recognition and processing of the livestock images are realized, the processing efficiency is improved, the manual intervention is reduced, and the influence of human factors on the acquisition of the livestock identification is reduced. In yet another aspect, consistency and convenience of data is achieved by using livestock images as a definitive source of livestock identification. The animal identification may be directly associated with animal information, such as health status information, weight information, etc., providing more possibilities for subsequent data analysis, management and decision making.
In some optional embodiments, the acquiring real-time weight information of the target livestock includes:
acquiring weight pressure signals of the target livestock in real time;
and filtering the weight pressure signal to obtain real-time weight information of the target livestock.
The beneficial effect of this technical scheme lies in: the weight pressure signal applied by the target livestock on the livestock weighing assembly is monitored in real time through the livestock weighing assembly, and the weight pressure signal is proportional to the weight of the livestock and can be used for calculating the real-time weight of the livestock. After obtaining the weight pressure signal, the signal is subjected to a filtering process, such as a kalman filter. The filtering process can eliminate noise and interference in the measurement, thereby obtaining more stable and accurate real-time weight information. On the one hand, since irregular movement of the target livestock into the detection area can interfere with acquisition of the weight pressure signal, more accurate real-time weight information is provided by acquiring the weight pressure signal in real time and applying a filtering processing method (such as kalman filtering). The filtering process may reduce the effects of noise and interference, thereby improving the accuracy and stability of the weight estimation. On the other hand, weight pressure signals of the target livestock are obtained in real time and converted into real-time weight information, so that the weight change condition of the livestock can be known in time, and decisions such as health condition assessment, feeding adjustment and disease prevention can be made. On the other hand, the real-time weight information acquisition and filtering processing is an automatic process, so that the automatic acquisition and processing of data can be realized, the requirements of manual intervention and operation are reduced, and the working efficiency and the data consistency are improved.
In a second aspect, the present application provides an electronic device comprising a memory and at least one processor, the memory storing a computer program, the at least one processor implementing the following steps when executing the computer program:
when a target livestock enters a detection area, acquiring a livestock image of the target livestock;
acquiring livestock information of the target livestock according to the livestock image;
acquiring sorting information corresponding to the target livestock based on the livestock information;
the target livestock are sorted to a sorting area corresponding to the sorting information.
In some alternative embodiments, the at least one processor, when executing the computer program, obtains the animal information of the target animal from the animal image in the following manner:
and performing image processing on the livestock image to identify and obtain the health state information of the target livestock.
In some alternative embodiments, the at least one processor, when executing the computer program, further performs the steps of:
acquiring real-time weight information of the target livestock;
acquiring an animal identification of the target animal based on the animal image;
Acquiring historical weight information of the target livestock according to the livestock identification;
acquiring development state information of the target livestock based on the real-time weight information and the historical weight information;
the at least one processor, when executing the computer program, obtains sorting information corresponding to the target livestock according to the livestock information in the following manner:
and determining sorting information corresponding to the target livestock based on the health state information and the development state information.
In some alternative embodiments, the at least one processor, when executing the computer program, obtains the developmental state information of the target livestock based on the real-time weight information and the historical weight information in the following manner:
acquiring a preset weight corresponding relation;
and acquiring the development state information of the target livestock based on the real-time weight information, the historical weight information and the weight correspondence.
In some alternative embodiments, the at least one processor, when executing the computer program, obtains the animal identification of the target animal based on the animal image in the following manner:
detecting whether the livestock characteristics of the target livestock in the livestock image meet preset image conditions or not; the livestock characteristics include one or more of facial characteristics, neck characteristics, ear characteristics, shoulder characteristics, back characteristics, chest characteristics, waist characteristics, hip characteristics, thigh characteristics, leg characteristics, tail characteristics, and arm characteristics of the target livestock; the preset image condition comprises at least one of feature definition and feature integrity;
If not, inputting the livestock image into a preset image complement model to obtain a complement image corresponding to the livestock image;
and acquiring the livestock identification of the target livestock based on the complement image.
In some alternative embodiments, the at least one processor, when executing the computer program, obtains the real-time weight information of the target livestock in the following manner:
acquiring weight pressure signals of the target livestock in real time;
and filtering the weight pressure signal to obtain real-time weight information of the target livestock.
In a third aspect, the present application provides a livestock sorting system, the system comprising:
the detection module is used for detecting livestock information of the target livestock;
the sorting module is used for sorting the target livestock;
an electronic device comprising a memory storing a computer program and a processor configured to implement the steps of any of the methods described above when the computer program is executed.
In some alternative embodiments, the detection module includes an image acquisition assembly and a livestock weighing assembly;
The image acquisition component is used for acquiring an animal image of the target animal when the target animal enters the detection area;
the livestock weighing assembly is used for acquiring weight pressure signals of the target livestock in real time.
In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program which, when executed by at least one processor, performs the steps of any of the methods or performs the functions of any of the electronic devices described above.
In a fifth aspect, the application also provides a computer program product comprising a computer program which, when executed by at least one processor, performs the steps of the method or performs the functions of the electronic device described in any of the preceding claims.
Drawings
The application will be further described with reference to the drawings and embodiments.
Fig. 1 shows a block diagram of a livestock sorting system according to an embodiment of the present application.
Fig. 2 shows a schematic flow chart of a livestock sorting method according to an embodiment of the application.
Fig. 3 is a schematic flow chart of determining sorting information according to an embodiment of the present application.
Fig. 4 is a schematic diagram of a loss curve according to an embodiment of the present application.
Fig. 5 shows a schematic diagram of an accuracy curve according to an embodiment of the present application.
Fig. 6 shows a schematic flow chart of image complement according to an embodiment of the present application.
Fig. 7 is a schematic diagram showing image contrast of an image complement result according to an embodiment of the present application.
Fig. 8 shows a flowchart of acquiring real-time weight information according to an embodiment of the present application.
Fig. 9 shows a schematic flow chart of kalman filtering according to an embodiment of the present application.
Fig. 10 shows a block diagram of an electronic device according to an embodiment of the present application.
Fig. 11 shows a schematic structural diagram of a program product according to an embodiment of the present application.
Detailed Description
The technical scheme of the present application will be described below with reference to the drawings and the specific embodiments of the present application, and it should be noted that, on the premise of no conflict, new embodiments may be formed by any combination of the embodiments or technical features described below.
In embodiments of the present application, "at least one" means one or more, and "a plurality" means two or more. "and/or", describes an association relationship of an association object, and indicates that there may be three relationships, for example, a and/or B, and may indicate: a alone, a and B together, and B alone, wherein a, B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one (one) of a, b or c may represent: a, b, c, a and b, a and c, b and c, a and b and c, wherein a, b and c can be single or multiple. It is noted that "at least one" may also be interpreted as "one (a) or more (a)".
It is also noted that, in embodiments of the present application, words such as "exemplary" or "such as" are used to mean serving as an example, instance, or illustration. Any implementation or design described as "exemplary" or "e.g." in the examples of this application should not be construed as preferred or advantageous over other implementations or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete fashion.
The technical field and related terms of the embodiments of the present application are briefly described below.
Livestock refers to animals raised by humans for purposes of food, agricultural work, transportation, fur, and the like. They are commonly used to provide meat, dairy, egg, leather and other animal products. The livestock include cattle, sheep, pig, horse, mule, donkey, chicken, duck, goose, etc. In agriculture and animal husbandry, livestock are often considered an important resource, providing humans with food and other living necessities.
The livestock sorting system refers to a system for automatically or semi-automatically sorting and sorting livestock, which are grouped according to predetermined criteria and requirements by detecting, identifying and analyzing the livestock using a detecting device. Its advantages include high productivity, low cost, low error rate and high accuracy of management of livestock. Livestock sorting systems are widely used in the fields of animal husbandry, slaughter industry, etc., and can provide efficient, accurate and reliable livestock sorting and sorting solutions.
ResNet-50 is a convolutional neural network model, which is a variant of the ResNet (Residual Network) series. ResNet is a deep learning model architecture proposed by Microsoft Research, which aims to solve the problem of degradation in the deep neural network training process (degradation problem). ResNet-50 refers specifically to the ResNet model that contains 50 convolutional layers and fully-concatenated layers. Its main structure consists of a number of residual blocks (residual blocks), each of which contains a number of convolutional layers and a batch normalization layer, using a modified linear unit (ReLU) as an activation function. ResNet-50 is trained on image Net datasets and achieves good performance in computer vision tasks such as image classification, object detection, and image segmentation. It has a deeper network structure, and can extract higher-level image features, so it is widely used in many visual tasks.
The ImageNet dataset is a benchmark dataset for image recognition and computer vision tasks; the ImageNet database contains millions of high resolution images, covering images from various fields and categories. The images are organized into a hierarchical label structure, each image being assigned a category label.
The Logistic model, also called Logistic regression model, is a statistical model for classification problems, belonging to a generalized linear model, which is often used to predict the probability of two or more classifications. The output of the Logistic model is a probability value that indicates the probability that the sample belongs to a certain class.
(System example)
The livestock sorting system provided by the application is described in detail below.
Referring to fig. 1, fig. 1 shows a block diagram of a livestock sorting system according to an embodiment of the present application.
An embodiment of the present application provides a livestock sorting system, the system comprising:
a detection module 20, wherein the detection module 20 is used for detecting livestock information of the target livestock;
a sorting module 30, wherein the sorting module 30 is used for sorting the target livestock;
an electronic device 10.
The type of the target livestock is not limited in the embodiment of the application, and can be, for example, cattle, sheep, pigs, horses, mules, donkeys, chickens, ducks, geese and the like.
In the present embodiment, the animal information includes at least one of an animal variety, an animal age, an animal sex, an animal weight, an animal health condition, an animal appearance characteristic, and an animal identification.
The livestock breed refers to a breed to which the target livestock belongs, such as cattle (beef cattle, dairy cows), sheep (goats, sheep), pigs (piglets, boars, sows, pregnant pigs, sick pigs), and the like. The livestock age refers to an age range or specific age of the target livestock, such as young livestock, adult livestock, aged livestock, and the like. Livestock sex refers to the sex of the target livestock, such as male and female. Livestock weight refers to weight information of the target livestock, and livestock health status refers to the health status of the target livestock, including physical status, diseased or injured condition. The animal appearance features refer to the appearance features of the target animal, such as speckle, hair color, disease spots, eye droppings, horn shape, etc. The livestock identification refers to identification information of the target livestock, such as a number, a name, and the like.
In this embodiment, the detection module 20 includes an image acquisition assembly and a livestock weighing assembly;
the image acquisition component is used for acquiring an animal image of the target animal when the target animal enters the detection area;
the livestock weighing assembly is used for acquiring weight pressure signals of the target livestock in real time.
The image acquisition component can be a camera, a scanner or other equipment (mobile phone, tablet personal computer and the like) with a photographing function. The camera may comprise, for example, an optical camera and/or an infrared camera. The image acquisition assembly is not limited herein.
The livestock weighing assembly comprises a weighing platform and a weighing sensor, wherein the weighing sensor can be a pressure sensor, a piezoelectric sensor, a vibration sensor or the like. The type of load cell is not limited here. The pressure sensor comprises a piezoresistive sensor, a capacitive sensor and a piezoelectric sensor. The type of pressure sensor is not limited here.
In some embodiments, the image capturing assembly further includes a light supplementing unit, where the light supplementing unit is used to enhance illumination of the detection area, and the light supplementing unit may use an LED lamp, or may use an incandescent lamp, and the light supplementing unit is not limited herein.
In the present embodiment, the detection area refers to a specific area for performing livestock detection. Which is a delimited space for guiding the target livestock therein and for necessary detection, observation and data collection.
As an example, the detection area is an area in a passage, and the traveling direction of the target livestock is guided by a rail, a guide device, or a fence, etc., and is ensured to enter the detection area. After the target livestock enters the detection area, a limited area is formed by closing the fence or the gate and the like, and the target livestock is limited in a controllable range for detection. The livestock weighing assembly is arranged in the detection area and is positioned on the ground of the detection area, and target livestock are driven onto a weighing platform of the livestock weighing assembly for weight measurement and recording.
In this embodiment, the sorting module 30 includes a gate control assembly and a drive assembly;
the gate control assembly is used for controlling gates of all sorting areas;
the driving component is used for driving the target livestock to move.
Wherein the gate control assembly is an assembly for controlling the gate according to the sorting information. And controlling the opening and closing of the corresponding gates according to the sorting information of the target livestock so as to guide the target livestock into the sorting area corresponding to the sorting information. The shutter control assembly may be implemented electrically, electronically, pneumatically, or mechanically. According to design requirements, the gate control assembly can be automated, and the state of the gate is detected and controlled in real time through the sensor and the actuator; or may be semi-automatic, requiring manual control by an operator based on sorting information.
The driving component is a component for driving the target livestock to move. The driving assembly is used for guiding and driving the target livestock to move towards a specified direction so as to guide the target livestock to a required position for sorting, weighing or other operations. For example, the following components may be included: driving rod, sound guiding device, electric door, pneumatic door, railing, manipulator etc. The animals are driven by the driving assembly, ensuring that they move in a desired path or flow.
As one example, the sorting information of the target livestock indicates sorting the target livestock to the sorting area C, the gate control assembly opens the gate of the sorting area C, and the driving assembly emits a stimulating sound behind the target livestock forcing the target livestock along the aisle towards the gate of the sorting area C. After the target livestock enter the sorting area C, the gate control assembly closes the gate of the sorting area C, thereby completing sorting of the target livestock.
In this embodiment, the detection module 20 further includes an inlet detection assembly;
the entrance detection assembly is used for detecting whether the target livestock enters the detection area.
The entrance detection assembly may use an infrared sensor to detect, or may use a camera to detect, which is not limited herein.
When the entrance detection assembly detects by adopting the infrared sensor, the infrared sensor is arranged on two sides of the entrance of the detection area, when the target livestock enter the detection area, the target livestock shield the single Shu Gong infrared rays emitted by the infrared sensor, so that the obstacle at the entrance is determined, when the livestock weighing assembly acquires the weight pressure signal, and the infrared sensor does not detect shielding, the target livestock enter the detection area.
When the entrance detection assembly adopts a camera to detect, the camera faces the entrance of the detection area, and whether the target livestock enters the detection area is determined through analysis by acquiring a video stream.
In the embodiment of the present application, the electronic device may be configured to implement the steps of the livestock sorting method, and the livestock sorting method will be described first and then the electronic device will be described.
(method example)
Referring to fig. 2, fig. 2 shows a schematic flow chart of a livestock sorting method according to an embodiment of the present application.
The application provides a livestock sorting method, which comprises the following steps:
step S101: when a target livestock enters a detection area, acquiring a livestock image of the target livestock;
step S102: acquiring livestock information of the target livestock according to the livestock image;
Step S103: acquiring sorting information corresponding to the target livestock based on the livestock information;
step S104: and sorting the target livestock to a sorting area corresponding to the sorting information.
In the present embodiment, the livestock image may be a whole body image, a sideways image, a face image, or the like of the target livestock, and the livestock image is not limited herein.
In this embodiment, the sorting area is an area that is pre-planned according to different sorting attributes. Wherein the sorting attribute comprises at least one of sex, age, weight, health status, developmental status of the target livestock. The sorting areas can be separated by fences or electronic fences, or independent rooms can be used as one sorting area, and the implementation mode of the sorting area is not limited.
As one example, when a target animal enters a detection zone, an image acquisition device (e.g., a camera) is used to acquire an animal image of the target animal. The animal image may be a whole body image, a side image, or other suitable view angle to obtain the animal's appearance characteristics and identification information. The acquired livestock images are analyzed and processed by image processing and computer vision techniques to extract livestock information of the target livestock. For example, the target livestock is determined to be pigs, the breed is white pigs, and accordingly, the target livestock is sorted to the sorting area a, and the sorting information includes information of sorting the target livestock to the sorting area a. Thereby sorting the target livestock based on the sorting information.
Thus, whether or not the target livestock enters the specified detection area is detected, and once the target livestock is detected to enter the detection area, the livestock image of the target livestock is acquired, and the livestock information of the target livestock, such as the breed, the livestock characteristics, the health condition, and the like, can be extracted by performing image processing and analysis on the acquired livestock image. Based on the livestock information of the target livestock, sorting information corresponding to the target livestock is acquired, the difference in the livestock information may cause the difference in the sorting information corresponding to the target livestock, the target livestock is sorted to a sorting area corresponding to the sorting information according to the acquired sorting information, for example, a gate of the sorting area is opened, and the target livestock is guided into the sorting area. On the one hand, automatic sorting of target livestock is achieved, manual intervention is not needed, working efficiency is improved, labor cost is reduced, and risk of human errors is reduced. On the other hand, by acquiring the animal image of the target animal to obtain the animal information of the target animal, and sorting the target animal based on the animal information, it is helpful to ensure that the target animal is properly sorted. On the other hand, the target livestock is sorted according to the livestock information of the target livestock, the livestock information not only comprises the weight information of the target livestock, and the problem that the sorting quality of the livestock is poor due to the fact that the livestock is sorted according to the weight of the livestock and the influence of factors such as the age, the sex and the health condition of the livestock cannot be comprehensively considered is solved.
In conclusion, through steps of image processing, information analysis, automatic sorting and the like, accurate sorting of target livestock is achieved, and the method has the effects of improving sorting efficiency, reducing sorting errors and the like.
In some embodiments, the acquiring the livestock image of the target livestock when the target livestock enters the detection zone comprises:
acquiring a plurality of alternative images of the target livestock;
based on preset image evaluation conditions, quality scoring is carried out on each candidate image so as to obtain a quality score corresponding to each candidate image;
and selecting one of the plurality of candidate images as the livestock image based on the quality score corresponding to each candidate image.
In this embodiment, the preset image evaluation conditions may include a sharpness requirement, an integrity requirement, a contrast requirement, or the like, which is not limited herein.
As one example, multiple alternative images of the target livestock are captured from different angles or fields of view. The alternative images may include images from different perspectives of the side, front, back, etc. And using an image quality evaluation algorithm to score the quality of each alternative image, and comprehensively considering indexes such as definition, contrast, brightness, noise level and the like of the alternative images to evaluate the quality of the images. And obtaining a corresponding quality score for each candidate image according to the output of the quality scoring algorithm. The quality score may be a numerical value reflecting the quality level of the image, generally the higher the quality the better. And selecting one image with higher quality as a final livestock image according to the quality score corresponding to each candidate image. The selection may be based on the highest score, a score meeting a threshold requirement, or based on a pre-set scoring rule to determine the best livestock image.
Thus, a plurality of alternative images are acquired as the target livestock enters the detection zone. The candidate images may be taken from different angles and under different illumination conditions, or may be taken from the same angle and in continuous time, and the acquisition mode of the candidate images is not limited herein. Scoring is performed for each candidate image using preset image quality assessment conditions. The quality score of each candidate image is obtained taking into account factors such as image sharpness, noise level, contrast, etc. One of the plurality of candidate images is selected as a final livestock image based on the quality score of each candidate image. The criteria for selection may depend on the application requirements, e.g. selecting the highest score candidate image or the candidate image meeting a certain threshold requirement. By scoring the quality of the candidate images, high-quality images can be screened out for subsequent processing and analysis, which is helpful for improving the definition of the images, reducing noise, etc., and providing more accurate information.
In some embodiments, the acquiring the livestock information of the target livestock according to the livestock image includes:
and performing image processing on the livestock image to identify and obtain the health state information of the target livestock.
In this embodiment, for the expression of livestock diseases, the health status information of the target livestock is obtained by identifying the target deep livestock line in the livestock image.
As one example, the target livestock is live pigs, and diseases of the live pigs often appear as abnormal in eyes, ears, nose and skin of the live pigs, so that whether the target livestock is ill or not can be judged by identifying whether the eyes, ears, nose and skin of the target livestock appear as abnormal or not in an image of the livestock. So that diseased live pigs are individually sorted into a sorting area for subsequent treatment by management personnel. At the same time, the disease is prevented from being mutually transmitted among live pigs.
Thus, the obtained livestock images are subjected to image processing, such as image enhancement, edge detection, feature extraction and other technologies, so that features and information related to the health state of the livestock in the livestock images can be extracted. Based on the processing results of the livestock images, algorithms such as machine learning, computer vision, and the like can be utilized to identify health status information of the target livestock. For example, the health status of livestock is judged by analyzing the characteristics of the body state, skin color, eye state and the like of livestock, such as whether the livestock suffers from diseases, is in estrus or not, and the like. And according to the health state information of the target livestock, acquiring sorting information corresponding to the target livestock, for example, sorting the livestock with diseases to a specific area for further processing, or sorting the livestock in estrus to an area related to reproduction. Based on the sorting information, the sorting area to which the target livestock should be sorted is determined, which is implemented by predefined rules and logic, ensuring that the target livestock is correctly sorted and handled. On the one hand, the health state information of the target livestock is accurately identified by processing and analyzing the livestock images, so that the diseased livestock or the livestock with health problems can be found early, and corresponding processing measures can be taken. On the other hand, customized sorting schemes are provided for each animal based on the health status information of the target animal, which helps to ensure that the animals are sorted to the proper sorting area for corresponding treatment, improving sorting efficiency and quality. In yet another aspect, data-based decision support is provided through the use of image processing and analysis techniques, which helps reduce interference with subjective judgment, providing more objective and accurate sorting decisions. On the other hand, by automatically acquiring the health state information and the corresponding sorting information of the target livestock, an automatic livestock sorting process is realized, the production efficiency can be improved, the labor cost can be reduced, the incidence of zoonotic diseases can be reduced, and the livestock can be ensured to be correctly processed and sorted.
In some embodiments, the image processing the livestock image to identify health status information of the target livestock includes:
and inputting the livestock image into a preset state detection model to obtain the health state information of the target livestock.
The state detection model training process comprises the following steps:
acquiring a training set, wherein the training set comprises a plurality of training data, and each training data comprises a sample livestock image and labeling data of health state information corresponding to the sample livestock image;
for each training data in the training set, performing the following processing:
inputting a sample livestock image in the training data into a preset deep learning model to obtain prediction data of health state information corresponding to the sample livestock image;
updating model parameters of the deep learning model based on the prediction data and the labeling data of the health state information corresponding to the sample livestock image;
detecting whether a preset training ending condition is met; if yes, taking the trained deep learning model as the state detection model; if not, continuing to train the deep learning model by using the next training data.
Therefore, through designing, a proper amount of neuron computing nodes and a multi-layer operation hierarchical structure are established, a proper input layer and a proper output layer are selected, a preset deep learning model can be obtained, through learning and tuning of the deep learning model, a functional relation from input to output is established, although the functional relation between input and output cannot be found by 100%, the functional relation between input and output can be approximated to the actual relation as much as possible, the state detection model obtained through training can be obtained based on livestock images, the application range is wide, and the accuracy and the reliability of the computing result are high.
In some embodiments of the application, the application may be trained to obtain a state detection model.
In other embodiments of the application, the application may employ a pre-trained state detection model.
In this embodiment, the preset deep learning model may be a convolutional neural network model or a cyclic neural network model, which is not limited herein to the implementation manner of the preset deep learning model.
The training process of the state detection model is not limited, and for example, the training mode of the supervised learning can be adopted, or the training mode of the semi-supervised learning can be adopted, or the training mode of the unsupervised learning can be adopted.
The preset training ending condition is not limited, and for example, the training times can reach the preset times (the preset times are, for example, 1 time, 3 times, 10 times, 100 times, 1000 times, 10000 times, etc.), or the training data in the training set can be all trained once or a plurality of times, or the total loss value obtained in the training is not more than the preset loss value.
In some embodiments, the health status information further includes a disease type; the method further comprises the steps of:
generating notification information based on the disease type and animal information when the health status information indicates that the target animal is ill;
and sending the notification information to user equipment to prompt the target livestock to be ill.
The present embodiment is not limited to the type of disease, and may be, for example, scabies, ear scabies, ringworm, dermatoma, ectoparasite infection, and the like.
In the embodiment of the present application, the user device is not limited, and may be, for example, an intelligent terminal device such as a mobile phone, a tablet computer, a notebook computer, a desktop computer, an intelligent wearable device, or the user device may be a workstation, a controller, or a console. Examples of the manner of sending the prompt message include short message pushing, mail pushing, in-application pushing, and telephone notification.
As an example, a message is sent to the administrator a01 in a manner of pushing a message, the text content included in the message is "administrator a01 is good, the live pig C51 is suspected to suffer from skin sore disease (Swine Erysipelas), and the live pig C51 is sorted into the sorting area D, and is processed as soon as possible, and the user is thanks-! "
Referring to fig. 3, fig. 3 is a schematic flow chart of determining sorting information according to an embodiment of the present application.
In some embodiments, the method further comprises:
step S201: acquiring real-time weight information of the target livestock;
step S202: acquiring an animal identification of the target animal based on the animal image;
step S203: acquiring historical weight information of the target livestock according to the livestock identification;
step S204: acquiring development state information of the target livestock based on the real-time weight information and the historical weight information;
the acquiring sorting information corresponding to the target livestock according to the livestock information comprises the following steps:
step S205: and determining sorting information corresponding to the target livestock based on the health state information and the development state information.
The real-time weight information of the target livestock refers to weight information obtained by real-time detection of the target livestock in the detection area through the livestock weighing assembly. The historical weight information of the target livestock refers to weight information recorded during the feeding of the target livestock, including the weight of the target livestock at different time points.
In this embodiment, the developmental status information may include low, sub-low, normal, higher, high developmental level; development grade differences, good, excellent may also be included; the developmental status information is not limited herein.
As an example, real-time weight information of live pigs is obtained by the livestock weighing assembly, resulting in a real-time weight of 150 kg. By using image recognition techniques such as computer vision and image processing algorithms, the live pig animal image is analyzed and animal identification of the live pig, such as number, name, etc., is determined. Assume that the animal identification PZ123 is successfully acquired from the animal image. The historical weight information for the live pig is retrieved in a record or database using the livestock identification PZ123. The historical weight information may include weight recordings of live pigs at different time points. It is assumed that over the past several months, the historical body weight of live pigs is recorded as follows: 3 months ago: 100 kg, 2 months ago: 120 kg, 1 month ago: 140 kg. Based on the real-time weight information and the historical weight information, the development state information of the live pigs is determined to be the development level difference through calculation. And sorting the live pigs to a sorting area A by combining the health status information of the live pigs as normal so as to adjust the feed and optimize the development status of the live pigs. If the health status information of the live pigs is diseased, sorting the live pigs to a sorting area B so as to facilitate management staff to treat the disease.
Thus, weight information of the target livestock is acquired in real time by using the livestock weighing assembly. To provide accurate data of the current weight status of the target livestock. By processing and analyzing the target livestock image, the livestock identification of the target livestock, such as identification information of numbers, names, and the like, is identified and extracted. Acquiring historical weight information of the target livestock according to the livestock identification of the target livestock by inquiring a database or an information system, wherein the historical weight information records past weight information of the target livestock; the weight growth trend and speed of the target livestock are calculated by comparing the real-time weight of the target livestock with the historical weight data thereof, so that the development state information of the target livestock is determined. For example, it may be determined that the target livestock is developing normally, growing too fast or growing too slow, etc. In combination with the health status information and the development status information of the target livestock, a specific area or process flow to which the target livestock should be sorted can be determined. On the one hand, the real-time weight information and the historical weight information of the target livestock are obtained to obtain the development state information of the target livestock, so that more comprehensive analysis and judgment are performed, more accurate and personalized sorting schemes are facilitated to be formulated, and the requirements and demands of the target livestock are met to the greatest extent. On the other hand, finer sorting classification is realized according to the real-time weight information and the historical weight information of the target livestock, and reasonable division of the livestock according to the weight and the development state of the livestock is facilitated, so that the precision and the effect of feeding management are improved. In yet another aspect, the selection of a feeding environment, i.e., a sorting area, suitable for the growth and development needs of the target livestock based on the development status information of the target livestock helps to provide good feeding conditions and promote healthy growth and production benefits of the livestock. On the other hand, based on the same livestock image, the health state information of the target livestock can be determined, and the livestock identification of the target livestock can be determined, so that the development state information of the target livestock is determined, multiplexing of the livestock image is realized, the whole image processing data amount is reduced, and the sorting efficiency is improved.
In some embodiments, the acquiring the developmental state information of the target livestock based on the real-time weight information and the historical weight information includes:
acquiring a preset weight corresponding relation;
and acquiring the development state information of the target livestock based on the real-time weight information, the historical weight information and the weight correspondence.
The weight correspondence is a correspondence between real-time weight information and historical weight information and development status information of the target livestock, and may be determined by a formula, a matrix or a model, for example, and the weight correspondence is not limited herein.
As an example, a growth curve of the target livestock is drawn based on real-time weight information and historical weight information, and the growth curve is subjected to nonlinear fitting by using a Logistic model, which can be represented by the following formula:
W=A/(1+e (a-rt) )
wherein W is the weight (kg) at t days of age, A is the limiting weight parameter (kg), a is the constant scale, r is the growth rate parameter, and e is the constant.
By determining the coefficient R 2 Representing the degree of fitting of the curve, can be represented by the following formula:
where y represents the actual observed value,representing model predictive value, +.>Representing the mean of the observations.
Determining the coefficient R 2 The closer to 1 the better the fitting effect of the model is, the better the variability of the dependent variable can be explained; the closer to 0 means that the model fitting is less effective and the variability of the dependent variables cannot be well explained.
After relevant parameters of the Logistic model are determined through experimental calculation, the development state information of the target livestock can be determined based on real-time weight information and historical weight information.
Thus, a preset weight correspondence is obtained, a set of weight correspondences is set, and the weight correspondence is used for converting the weight of livestock into specific development state information, for example, a relationship model or a mathematical model of weight range and growth stage is established. And acquiring real-time weight information of the target livestock by means of weighing equipment or sensors and the like. And meanwhile, acquiring historical weight information of the target livestock by inquiring a database or an information system, and comparing and matching the real-time weight information and the historical weight information with development state information by utilizing a preset weight corresponding relation. The information of the development state of the target livestock, such as the growth stage, the development degree and the like, is determined through calculation, comparison or interpolation and the like. On the one hand, the development state of the target livestock is accurately estimated by using a preset weight corresponding relation. The weight is an important index of the growth and development of livestock, and by correlating the weight with the development status, the growth stage and development level of the target livestock can be more accurately judged. On the other hand, through obtaining the development state information of the target livestock, personalized management and adjustment are carried out on the livestock at different development stages so as to meet the requirements of the livestock at different development states.
In some embodiments, the animal identification refers to an identification determined for identifying and tracking livestock individuals for identification, traceability, and management of the livestock. Which may be determined, for example, by means of ear tags, body tags, animal character recognition, etc.
Specifically, the acquiring the livestock identification of the target livestock based on the livestock image includes:
identifying the livestock image to acquire the livestock characteristics of the target livestock; the livestock characteristics include one or more of facial characteristics, neck characteristics, ear characteristics, shoulder characteristics, back characteristics, chest characteristics, waist characteristics, hip characteristics, thigh characteristics leg characteristics, tail characteristics, and arm characteristics of the target livestock;
based on the animal characteristics, an animal identification of the target animal is determined.
Wherein facial features refer to the morphology, structure and characteristics of the livestock face. Facial features include eye features, nose features, ear features, texture features, etc., and are not limited herein. Neck characteristics refer to the shape, structure and characteristics of the livestock neck, including neck length, thickness, curvature, etc., and are not limited herein. Ear characteristics refer to the shape, structure and characteristics of the livestock ear, including the size, shape, standing or hanging ears, etc., and are not limited herein. Shoulder characteristics refer to the morphology, structure and characteristics of the livestock shoulder, including shoulder width, slope, angle of the scapula, etc., and are not limited herein. The back characteristics refer to the shape, structure and characteristics of the back of the livestock, including the straight or bending degree of the back line, the height of the back, etc., and the back characteristics are not limited herein. Chest characteristics refer to the morphology, structure and characteristics of livestock chest, including chest depth, chest width, shape of the chest, etc., and are not limited herein. The waist features refer to the shape, structure and characteristics of the livestock waist, including waist length, waist width, waist-abdomen tightness, etc., and are not limited herein. Buttock characteristics refer to the shape, structure and characteristics of the buttocks of livestock, including buttock width, buttock height, muscular development of the buttocks, etc., and are not limited herein. Thigh characteristics refer to the morphology, structure and characteristics of the livestock thigh (thigh), including the length of the femur, the development of the femur muscle, etc., and are not limited herein. The leg characteristics refer to the shape, structure and characteristics of the hind leg and calf of livestock. Tail characteristics refer to tail shape, thickness, bending degree and the like of livestock. Arm characteristics refer to the morphology, structure and characteristics of the livestock arm (forelimb upper arm), including arm length, arm thickness, arm muscle development, etc., and are not limited herein.
In this embodiment, the livestock identifier may be a combination of characters, numbers, and symbols, such as live pig a01, flowers, cow C51, sheep 105, Z10245, and the like, and the livestock identifier is not limited herein.
In some embodiments, the acquiring the animal identification of the target animal based on the animal image includes:
inputting the livestock image into a preset livestock identification model to obtain the livestock identification of the target livestock.
The livestock identification model can be obtained based on convolutional neural network model training or based on cyclic neural network model training, and the implementation mode of the livestock identification model is not limited. The convolutional neural network model may be a ResNet-50 model, among others.
As one example, pretraining is performed on an ImageNet dataset using a res net-50 based model, and the identification of live pigs is performed on the basis of pretraining. Freezing the convolution layer of the pre-trained ResNet-50 model on the ImageNet data set, and retraining the full-connection layer parameters to obtain the accuracy of the network. And after the shot pictures are classified, stored, screened and cut, 6000 pictures are used as training sets, 2000 pictures are used as verification sets, and 2000 pictures are used as test set data. And during testing, the images under the test data set are sent to the trained livestock identification model for carrying out livestock identification, and the successfully identified images are stored in the corresponding data set. Experiments show that the accuracy of the 25 convolution layers before freezing is highest. Because the pig grows fast, the necessary sorting frequency is improved, and the accuracy of identity recognition can be ensured. Thus, with 50 epochs (full traversals), iterating 750 times, training with a batch size of 8, learning rate of 0.005, using the Relu function as the activation function. The loss curve of the training result is shown in fig. 4, the accuracy curve is shown in fig. 5, in fig. 4, the abscissa represents the iteration number, the ordinate represents the loss value, "loss" is an index for measuring the error between the model predicted output and the actual label on each training batch or each epoch during training of the neural network, and represents the fitting degree of the model on the training data. "val_loss" refers to the loss value calculated during model training using a validation data set, which is a separate data set from the training data set, used to evaluate the performance of the model on unseen data. In fig. 5, the abscissa represents the iteration number, the ordinate represents the accuracy, "acc" (accuracy) represents the proportion of samples of the model that are completely matched with the actual label in the prediction output during training or verification, and the ratio is an index for measuring the performance of the classification model. "val_acc" (verification accuracy) refers to the accuracy calculated on the verification data set, and similar to val_loss, the verification accuracy is used to evaluate the performance of the model on unseen data to determine the generalization ability of the model.
Referring to fig. 6, fig. 6 shows a schematic flow chart of image completion according to an embodiment of the present application.
In some embodiments, the livestock image is an image acquired by collecting the target livestock in the traveling direction of the target livestock when the target livestock enters the detection area;
the acquiring the livestock identification of the target livestock based on the livestock image comprises the following steps:
step S301: detecting whether the livestock characteristics of the target livestock in the livestock image meet preset image conditions or not;
step S302: if not, inputting the livestock image into a preset image complement model to obtain a complement image corresponding to the livestock image;
step S303: and acquiring the livestock identification of the target livestock based on the complement image.
Wherein the livestock characteristics include one or more of facial characteristics, neck characteristics, ear characteristics, shoulder characteristics, back characteristics, chest characteristics, waist characteristics, hip characteristics, thigh characteristics, leg characteristics, tail characteristics, and arm characteristics of the target livestock; the preset image condition includes at least one of feature definition and feature integrity.
In some embodiments, the image acquisition component is disposed in a direction of travel of the target animal and is directly opposite the target animal to make facial features of the target animal more complete in the animal image, thereby determining an animal identification of the target animal based on the facial features.
In this embodiment, the image complement model may be a convolutional neural network model or a cyclic neural network model, and the implementation manner of the preset image complement model is not limited here.
In some embodiments, the image completion model employs a method of generating image completions against a network WGAN by adding masks to the input sample livestock image to combine the context and perception losses. The context loss ensures that the obtained complement image corresponding to the livestock image and the sample livestock image are the same as possible, and the perception loss ensures the authenticity of the image, so that the complement repair picture is obtained. The mask is to mark the area to be complemented on the input sample livestock image, and the area can be set manually or detected automatically by other algorithms, which represents the part to be repaired or filled. The context Loss (context Loss) is a Loss function that measures the similarity of a generated image to an input image in a complement region. By minimizing the context loss, the generated image complement model can keep the complement image consistent with the original image in the complement area as much as possible, so as to preserve the overall structure and detail of the original image. The perceived Loss (perceived Loss) is a Loss function based on perceived quality that is used to measure the difference between the generated image and the real image. Which measures the authenticity of an image by calculating the difference between the generated image and the actual image on the high-level feature representation. The perceived loss can help the model generate a realistic image, making it visually closer to a real image.
As an example, fig. 7 is an image contrast schematic diagram of an image complement result, where an original image is blocked to obtain a blocked image, and an image complement model performs image complement on the blocked image to obtain a complemented image.
In practical application, the characteristics of the blocked livestock in the livestock image are complemented by inputting the livestock image into an image complement model.
Thus, the animal characteristics of the target animal in the animal image are analyzed to detect whether the sharpness and/or integrity of the animal characteristics meet preset image conditions, such as the sharpness of facial, neck, ear, shoulder, back, chest, waist, buttocks, thigh, leg, tail or arm characteristics (e.g., resolution, contrast, etc. of the image) and the integrity of facial, neck, ear, shoulder, back, chest, waist, buttocks, thigh, leg, tail or arm characteristics. If the characteristics of the livestock do not meet the preset image conditions, inputting the livestock image into a preset image complement model. And performing image reconstruction or complementation operation according to the existing image information and part of the characteristics of the livestock, and generating a complementation image of the characteristics of the target livestock. Based on the obtained complement image, identification information of the livestock, such as identification information of numbers, names and the like, is determined through image processing and recognition technology. The identification information is used for identity confirmation of the target livestock. On the one hand, the accuracy of the livestock identification is improved by collecting the livestock images facing the advancing direction of the target livestock and according to the preset image conditions and the image complement model. Even if the livestock features are incomplete or have low definition, the complement image of the livestock features with legibility can be generated, so that the acquisition rate and accuracy of the livestock identification are improved. On the other hand, by utilizing the image processing and pattern recognition technology, the automatic and intelligent recognition and processing of the livestock images are realized, the processing efficiency is improved, the manual intervention is reduced, and the influence of human factors on the acquisition of the livestock identification is reduced. In yet another aspect, consistency and convenience of data is achieved by using livestock images as a definitive source of livestock identification. The animal identification may be directly associated with animal information, such as health status information, weight information, etc., providing more possibilities for subsequent data analysis, management and decision making.
Referring to fig. 8, fig. 8 is a schematic flow chart of acquiring real-time weight information according to an embodiment of the present application.
In some embodiments, weight collection may be disturbed due to irregular movements of the target livestock. It is therefore necessary to more accurately determine real-time weight information of the target livestock by performing a filtering operation on the weight pressure signal.
Specifically, the acquiring the real-time weight information of the target livestock includes:
step S401: acquiring weight pressure signals of the target livestock in real time;
step S402: and filtering the weight pressure signal to obtain real-time weight information of the target livestock.
In this embodiment, the weight pressure signal of the target livestock is obtained by using the livestock weighing assembly, and the livestock weighing assembly may employ a voltage type weighing sensor connected to the weighing platform of the livestock weighing assembly, and the weight pressure signal is obtained in real time by the voltage type weighing sensor when the target livestock arrives on the weighing platform.
In some embodiments, the weight pressure signal is filtered using a kalman filter algorithm, where the kalman filter equation can be represented by the following formulas (1) to (5), and the formula (1) is a state prediction equation, and the state at the kth time is estimated from the state at the k-1 time, as shown in the following formulas:
X k =AX k-1 +BU k-1 (1)
Wherein X is k A state at time k; a represents a state transition matrix; b represents an input control matrix; u represents the outside worldActing as a medicine.
Equation (2) represents the error matrix prediction, using the covariance P at time k-1 k-1 Recursion is performed to obtain covariance estimation P at the kth moment k The following formula is shown:
P k =AP k-1 A T +Q (2)
wherein P is k A covariance estimate at time k; a represents a state transition matrix; q represents the prediction noise covariance matrix.
Equation (3) represents the kalman gain calculation, as shown in the following equation:
K k =P k H T (HP k H T +R) -1 (3)
wherein K is k Is Kalman gain; h is an observation matrix; and R is a measurement noise covariance matrix.
Equation (4) represents a state correction, as shown in the following equation:
X k =X k +K k (Z k -HX k ) (4)
wherein K is k Is Kalman gain; h is an observation matrix; z is Z k Is the observed value at k.
Equation (5) represents covariance matrix update, as shown in the following equation:
P k =(I-K k H)P k (5)
wherein K is k Is Kalman gain; i represents an identity matrix; h is an observation matrix; p (P) k The covariance estimate at time k is shown.
Referring to a schematic flow chart of kalman filtering shown in fig. 9, experimental analysis is performed to determine a prediction noise covariance matrix and a measurement noise covariance matrix, so as to complete construction of a kalman filtering equation.
Thus, through the livestock weighing assembly, the weight pressure signal applied by the target livestock on the livestock weighing assembly is monitored in real time, and the weight pressure signal is proportional to the weight of the livestock and can be used for calculating the real-time weight of the livestock. After obtaining the weight pressure signal, the signal is subjected to a filtering process, such as a kalman filter. The filtering process can eliminate noise and interference in the measurement, thereby obtaining more stable and accurate real-time weight information. On the one hand, since irregular movement of the target livestock into the detection area can interfere with acquisition of the weight pressure signal, more accurate real-time weight information is provided by acquiring the weight pressure signal in real time and applying a filtering processing method (such as kalman filtering). The filtering process may reduce the effects of noise and interference, thereby improving the accuracy and stability of the weight estimation. On the other hand, weight pressure signals of the target livestock are obtained in real time and converted into real-time weight information, so that the weight change condition of the livestock can be known in time, and decisions such as health condition assessment, feeding adjustment and disease prevention can be made. On the other hand, the real-time weight information acquisition and filtering processing is an automatic process, so that the automatic acquisition and processing of data can be realized, the requirements of manual intervention and operation are reduced, and the working efficiency and the data consistency are improved.
In a specific application scenario, an embodiment of the present application provides a method for sorting livestock, the method including:
when a target livestock enters a detection area, acquiring a livestock image of the target livestock; the livestock image is an image obtained by collecting the target livestock in the advancing direction of the target livestock when the target livestock enters the detection area;
performing image processing on the livestock image to identify and obtain health state information of the target livestock;
acquiring real-time weight information of the target livestock;
detecting whether the livestock characteristics of the target livestock in the livestock image meet preset image conditions or not; the livestock characteristics include one or more of facial characteristics, neck characteristics, ear characteristics, shoulder characteristics, back characteristics, chest characteristics, waist characteristics, hip characteristics, thigh characteristics, leg characteristics, tail characteristics, and arm characteristics of the target livestock; the preset image condition comprises at least one of feature definition and facial features and feature integrity;
if not, inputting the livestock image into a preset image complement model to obtain a complement image corresponding to the livestock image;
Acquiring a livestock identification of the target livestock based on the complement image;
acquiring historical weight information of the target livestock according to the livestock identification;
acquiring a preset weight corresponding relation;
acquiring development state information of the target livestock based on the real-time weight information, the historical weight information and the weight correspondence;
determining sorting information corresponding to the target livestock based on the health state information and the development state information;
the target livestock are sorted to a sorting area corresponding to the sorting information.
In practical application, taking a target livestock as a live pig as an example, detecting whether the live pig enters a detection area, and acquiring a livestock image of the live pig through an image acquisition component facing the running direction of the live pig when the live pig enters the detection area. And performing image processing on the livestock images to identify health state information of the live pigs. If the facial skin of the live pig is abnormal, the live pig is possibly indicated to suffer from diseases, the live pig is judged to be ill by image analysis and detection, corresponding health state information is judged to be ill, real-time weight information of a target live pig is obtained through the livestock weighing assembly, weight pressure signals of the live pig are monitored and recorded in real time, and then the real-time weight information is obtained through processing of the weight pressure signals. Detecting whether the facial features and the ear features of the live pigs in the livestock images meet preset image conditions or not, wherein the requirements of the facial features, the ear feature definition and the feature integrity are met. If the facial features of the live pigs in the livestock image are detected to be incomplete, inputting the livestock image into a preset image complement model to complement the facial features of the live pigs in the livestock image, so that a facial complement image of the target live pigs is obtained. Based on the complement image, a livestock identification of the live pig, such as a code of the live pig, is obtained. And acquiring historical weight information of the target live pig according to the live pig code. A historical weight record associated with the live pig's code is obtained by retrieving the record or querying a database. And acquiring a preset weight corresponding relation, such as a matrix, a model or a mathematical formula. And acquiring the development state information of the target live pig based on the real-time weight information, the historical weight information and the weight corresponding relation. If the development state information of the live pigs obtained through calculation indicates that the development level of the live pigs is normal, sorting information corresponding to target live pigs is determined based on the health state information and the development state information. Since the health status of the live pigs is sick, the development status is normal, and the live pigs are determined to be sorted to a sorting area C according to the information, wherein the sorting area C is an area which needs to inform a manager to carry out subsequent treatment on livestock in the sorting area. According to the sorting information, the gate control assembly opens the gate of the sorting area C, and utilizes the driving assembly to manufacture sound or light to drive the live pigs into the sorting area C, and after the live pigs enter the sorting area C, the gate of the sorting area C is closed, so that the live pigs are sorted.
(electronic device)
The embodiment of the application also provides an electronic device, the specific embodiment of which is consistent with the embodiment described in the method embodiment and the achieved technical effect, and part of the contents are not repeated.
The electronic device comprises a memory and at least one processor, the memory storing a computer program, the at least one processor implementing the following steps when executing the computer program:
when a target livestock enters a detection area, acquiring a livestock image of the target livestock;
acquiring livestock information of the target livestock according to the livestock image;
acquiring sorting information corresponding to the target livestock based on the livestock information;
the target livestock are sorted to a sorting area corresponding to the sorting information.
In some alternative embodiments, the at least one processor, when executing the computer program, obtains the animal information of the target animal from the animal image in the following manner:
and performing image processing on the livestock image to identify and obtain the health state information of the target livestock.
In some alternative embodiments, the at least one processor, when executing the computer program, further performs the steps of:
Acquiring real-time weight information of the target livestock;
acquiring an animal identification of the target animal based on the animal image;
acquiring historical weight information of the target livestock according to the livestock identification;
acquiring development state information of the target livestock based on the real-time weight information and the historical weight information;
the at least one processor, when executing the computer program, obtains sorting information corresponding to the target livestock according to the livestock information in the following manner:
and determining sorting information corresponding to the target livestock based on the health state information and the development state information.
In some alternative embodiments, the at least one processor, when executing the computer program, obtains the developmental state information of the target livestock based on the real-time weight information and the historical weight information in the following manner:
acquiring a preset weight corresponding relation;
and acquiring the development state information of the target livestock based on the real-time weight information, the historical weight information and the weight correspondence.
In some alternative embodiments, the at least one processor, when executing the computer program, obtains the animal identification of the target animal based on the animal image in the following manner:
Detecting whether the livestock characteristics of the target livestock in the livestock image meet preset image conditions or not; the livestock characteristics include one or more of facial characteristics, neck characteristics, ear characteristics, shoulder characteristics, back characteristics, chest characteristics, waist characteristics, hip characteristics, thigh characteristics, leg characteristics, tail characteristics, and arm characteristics of the target livestock; the preset image condition comprises at least one of feature definition and feature integrity;
if not, inputting the livestock image into a preset image complement model to obtain a complement image corresponding to the livestock image;
and acquiring the livestock identification of the target livestock based on the complement image.
In some alternative embodiments, the at least one processor, when executing the computer program, obtains the real-time weight information of the target livestock in the following manner:
acquiring weight pressure signals of the target livestock in real time;
and filtering the weight pressure signal to obtain real-time weight information of the target livestock.
Referring to fig. 10, fig. 10 is a block diagram of an electronic device 10 according to an embodiment of the present application.
The electronic device 10 may for example comprise at least one memory 11, at least one processor 12 and a bus 13 connecting the different platform systems.
Memory 11 may include readable media in the form of volatile memory, such as Random Access Memory (RAM) 111 and/or cache memory 112, and may further include Read Only Memory (ROM) 113.
The memory 11 also stores a computer program executable by the processor 12 to cause the processor 12 to implement the steps of any of the methods described above.
Memory 11 may also include utility 114 having at least one program module 115, such program modules 115 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment.
Accordingly, the processor 12 may execute the computer programs described above, as well as may execute the utility 114.
The processor 12 may employ one or more application specific integrated circuits (ASICs, application Specific Integrated Circuit), DSPs, programmable logic devices (PLDs, programmable Logic Device), complex programmable logic devices (CPLDs, complex Programmable Logic Device), field programmable gate arrays (FPGAs, fields-Programmable Gate Array), or other electronic components.
Bus 13 may be a local bus representing one or more of several types of bus structures including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or any of a variety of bus architectures.
The electronic device 10 may also communicate with one or more external devices such as a keyboard, pointing device, bluetooth device, etc., as well as one or more devices capable of interacting with the electronic device 10 and/or with any device (e.g., router, modem, etc.) that enables the electronic device 10 to communicate with one or more other computing devices. Such communication may be via the input-output interface 14. Also, the electronic device 10 may communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet, through a network adapter 15. The network adapter 15 may communicate with other modules of the electronic device 10 via the bus 13. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with the electronic device 10 in actual applications, including, but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, data backup storage platforms, and the like.
(computer-readable storage Medium)
The embodiment of the application also provides a computer readable storage medium, and the specific embodiment of the computer readable storage medium is consistent with the embodiment recorded in the method embodiment and the achieved technical effect, and part of the contents are not repeated.
The computer readable storage medium stores a computer program which, when executed by at least one processor, performs the steps of any of the methods or performs the functions of any of the electronic devices described above.
The computer readable medium may be a computer readable signal medium or a computer readable storage medium. In embodiments of the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The computer readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, with readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable storage medium may also be any computer readable medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing. Program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the C programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN), a Wide Area Network (WAN) or a public network (Internet), or may be connected to an external computing device (e.g., through an Internet connection using an Internet service provider).
(computer program product)
The embodiment of the application also provides a computer program product, the specific embodiment of which is consistent with the embodiment described in the method embodiment and the achieved technical effect, and part of the contents are not repeated.
The present application provides a computer program product comprising a computer program which, when executed by at least one processor, performs the steps of any of the methods or performs the functions of any of the electronic devices described above.
Referring to fig. 11, fig. 11 is a schematic structural diagram of a computer program product according to an embodiment of the present application.
The computer program product is configured to implement the steps of any of the methods described above or to implement the functions of any of the electronic devices described above. The computer program product may employ a portable compact disc read only memory (CD-ROM) and comprise program code and may run on a terminal device, such as a personal computer. However, the computer program product of the present application is not limited thereto, and the computer program product may employ any combination of one or more computer readable media.
The present application has been described in terms of its purpose, performance, advancement, and novelty, and the like, and is thus adapted to the functional enhancement and use requirements highlighted by the patent statutes, but the description and drawings are not limited to the preferred embodiments of the present application, and therefore, all equivalents and modifications that are included in the construction, apparatus, features, etc. of the present application shall fall within the scope of the present application.
Claims (11)
1. A method of livestock sorting, the method comprising:
when a target livestock enters a detection area, acquiring a livestock image of the target livestock;
acquiring livestock information of the target livestock according to the livestock image;
acquiring sorting information corresponding to the target livestock based on the livestock information;
the target livestock are sorted to a sorting area corresponding to the sorting information.
2. The method of claim 1, wherein the animal information includes health status information, the acquiring the animal information of the target animal from the animal image includes:
and performing image processing on the livestock image to identify and obtain the health state information of the target livestock.
3. The method according to claim 2, wherein the method further comprises:
acquiring real-time weight information of the target livestock;
acquiring an animal identification of the target animal based on the animal image;
acquiring historical weight information of the target livestock according to the livestock identification;
acquiring development state information of the target livestock based on the real-time weight information and the historical weight information;
The acquiring sorting information corresponding to the target livestock according to the livestock information comprises the following steps:
and determining sorting information corresponding to the target livestock based on the health state information and the development state information.
4. A method according to claim 3, wherein said obtaining developmental status information of said target livestock based on said real-time weight information and said historical weight information comprises:
acquiring a preset weight corresponding relation;
and acquiring the development state information of the target livestock based on the real-time weight information, the historical weight information and the weight correspondence.
5. A method according to claim 3, wherein said acquiring an animal identification of said target animal based on said animal image comprises:
detecting whether the livestock characteristics of the target livestock in the livestock image meet preset image conditions or not; the livestock characteristics include one or more of facial characteristics, neck characteristics, ear characteristics, shoulder characteristics, back characteristics, chest characteristics, waist characteristics, hip characteristics, thigh characteristics, leg characteristics, tail characteristics, and arm characteristics of the target livestock; the preset image condition comprises at least one of feature definition and feature integrity;
If not, inputting the livestock image into a preset image complement model to obtain a complement image corresponding to the livestock image;
and acquiring the livestock identification of the target livestock based on the complement image.
6. A method according to claim 3, wherein said obtaining real-time weight information of said target livestock comprises:
acquiring weight pressure signals of the target livestock in real time;
and filtering the weight pressure signal to obtain real-time weight information of the target livestock.
7. An electronic device comprising a memory and at least one processor, the memory storing a computer program, the at least one processor implementing the following steps when executing the computer program:
when the target livestock enter the detection area, acquiring a livestock image of the target livestock;
acquiring livestock information of the target livestock according to the livestock image;
acquiring sorting information corresponding to the target livestock based on the livestock information;
the target livestock are sorted to a sorting area corresponding to the sorting information.
8. A livestock sorting system, the system comprising:
The detection module is used for detecting livestock information of the target livestock;
the sorting module is used for sorting the target livestock;
an electronic device comprising a memory storing a computer program and a processor configured to implement the steps of the method of any of claims 1-6 when the computer program is executed.
9. The system of claim 8, wherein the detection module comprises an image acquisition assembly and a livestock weighing assembly;
the image acquisition component is used for acquiring an animal image of the target animal when the target animal enters the detection area;
the livestock weighing assembly is used for acquiring weight pressure signals of the target livestock in real time.
10. A computer-readable storage medium, characterized in that it stores a computer program which, when executed by at least one processor, implements the steps of the method of any of claims 1-6 or the functions of the electronic device of claim 7.
11. A computer program product, characterized in that it comprises a computer program which, when executed by at least one processor, implements the steps of the method according to any one of claims 1-6 or the functions of the electronic device according to claim 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310732516.4A CN116740765A (en) | 2023-06-19 | 2023-06-19 | Livestock sorting method, livestock sorting system and related devices |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310732516.4A CN116740765A (en) | 2023-06-19 | 2023-06-19 | Livestock sorting method, livestock sorting system and related devices |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116740765A true CN116740765A (en) | 2023-09-12 |
Family
ID=87911169
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310732516.4A Pending CN116740765A (en) | 2023-06-19 | 2023-06-19 | Livestock sorting method, livestock sorting system and related devices |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116740765A (en) |
-
2023
- 2023-06-19 CN CN202310732516.4A patent/CN116740765A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Neethirajan | The role of sensors, big data and machine learning in modern animal farming | |
Wang et al. | ASAS-NANP SYMPOSIUM: Applications of machine learning for livestock body weight prediction from digital images | |
Cihan et al. | A review of machine learning applications in veterinary field | |
Amraei et al. | Broiler weight estimation based on machine vision and artificial neural network | |
Garcia et al. | Lameness detection challenges in automated milking systems addressed with partial least squares discriminant analysis | |
MX2015006775A (en) | Systems and methods for predicting the outcome of a state of a subject. | |
Subedi et al. | Tracking floor eggs with machine vision in cage-free hen houses | |
Huang et al. | An intelligent method for detecting poultry eating behaviour based on vocalization signals | |
Lee et al. | Prediction of feather damage in laying hens using optical flows and Markov models | |
US11278007B2 (en) | Methods and systems for using sound data to analyze health condition and welfare states in collections of farm animals | |
Majewski et al. | Multipurpose monitoring system for edible insect breeding based on machine learning | |
KR102601051B1 (en) | Livestock disease real-time surveillance technology and intelligent quarantine and sanitation system | |
CN116740765A (en) | Livestock sorting method, livestock sorting system and related devices | |
CN115830078A (en) | Live pig multi-target tracking and behavior recognition method, computer equipment and storage medium | |
Ntalampiras et al. | Automatic detection of cow/calf vocalizations in free-stall barn | |
Ali et al. | Application of artificial intelligence in monitoring of animal health and welfare | |
Lupo et al. | Feasibility of screening broiler chicken flocks for risk markers as an aid for meat inspection | |
Bastiaansen et al. | Continuous real-time cow identification by reading ear tags from live-stream video | |
Alon et al. | Machine vision-based automatic lamb identification and drinking activity in a commercial farm | |
Youfu et al. | Status Quo of Waterfowl Intelligent Farming Research Review and Development Trend Analysis | |
Owalekar et al. | An agri vigilance system based on computer vision and deep learning | |
CN118135500B (en) | Intelligent monitoring platform and monitoring method for livestock breeding | |
Yoon et al. | Deep learning framework for bovine iris segmentation | |
Cassimiro et al. | Milk weighing scale based on machine learning | |
CN111626196B (en) | Knowledge-graph-based intelligent analysis method for body structure of typical bovine animal |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |