CN115205716A - Method, device and system for estimating oil content of olive fruits and storage medium - Google Patents
Method, device and system for estimating oil content of olive fruits and storage medium Download PDFInfo
- Publication number
- CN115205716A CN115205716A CN202210961085.4A CN202210961085A CN115205716A CN 115205716 A CN115205716 A CN 115205716A CN 202210961085 A CN202210961085 A CN 202210961085A CN 115205716 A CN115205716 A CN 115205716A
- Authority
- CN
- China
- Prior art keywords
- fruit
- content
- olive
- peel
- pigment
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/17—Terrestrial scenes taken from planes or by drones
-
- C—CHEMISTRY; METALLURGY
- C11—ANIMAL OR VEGETABLE OILS, FATS, FATTY SUBSTANCES OR WAXES; FATTY ACIDS THEREFROM; DETERGENTS; CANDLES
- C11B—PRODUCING, e.g. BY PRESSING RAW MATERIALS OR BY EXTRACTION FROM WASTE MATERIALS, REFINING OR PRESERVING FATS, FATTY SUBSTANCES, e.g. LANOLIN, FATTY OILS OR WAXES; ESSENTIAL OILS; PERFUMES
- C11B1/00—Production of fats or fatty oils from raw materials
- C11B1/02—Pretreatment
- C11B1/04—Pretreatment of vegetable raw material
-
- C—CHEMISTRY; METALLURGY
- C11—ANIMAL OR VEGETABLE OILS, FATS, FATTY SUBSTANCES OR WAXES; FATTY ACIDS THEREFROM; DETERGENTS; CANDLES
- C11B—PRODUCING, e.g. BY PRESSING RAW MATERIALS OR BY EXTRACTION FROM WASTE MATERIALS, REFINING OR PRESERVING FATS, FATTY SUBSTANCES, e.g. LANOLIN, FATTY OILS OR WAXES; ESSENTIAL OILS; PERFUMES
- C11B1/00—Production of fats or fatty oils from raw materials
- C11B1/10—Production of fats or fatty oils from raw materials by extracting
- C11B1/108—Production of fats or fatty oils from raw materials by extracting after-treatment, e.g. of miscellae
-
- C—CHEMISTRY; METALLURGY
- C11—ANIMAL OR VEGETABLE OILS, FATS, FATTY SUBSTANCES OR WAXES; FATTY ACIDS THEREFROM; DETERGENTS; CANDLES
- C11B—PRODUCING, e.g. BY PRESSING RAW MATERIALS OR BY EXTRACTION FROM WASTE MATERIALS, REFINING OR PRESERVING FATS, FATTY SUBSTANCES, e.g. LANOLIN, FATTY OILS OR WAXES; ESSENTIAL OILS; PERFUMES
- C11B3/00—Refining fats or fatty oils
- C11B3/008—Refining fats or fatty oils by filtration, e.g. including ultra filtration, dialysis
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Oil, Petroleum & Natural Gas (AREA)
- Wood Science & Technology (AREA)
- Organic Chemistry (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Microbiology (AREA)
- Remote Sensing (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a method, a device and a system for estimating the oil content of olive fruits and a storage medium, wherein the method comprises the following steps: based on a fruit image shot by an unmanned aerial vehicle, performing correlation analysis on the peel pigment content and the oil content of a corresponding fruit in the fruit image, screening out a pigment index which is remarkably related to the oil content in the olive, and establishing a dynamic mapping model; training a fruit peel pigment content prediction model based on the fruit image and the fruit peel pigment of the corresponding fruit as real values, and predicting the fruit image acquired in real time to obtain the fruit peel pigment content of the fruit; obtaining the internal oil content of the fruit according to the pigment content of the pericarp of the fruit. The method takes the olive peel pigment content as a bridge, establishes a quantitative detection model of fruits from external phenotypic characteristics to the peel pigment content and then to the internal oil content, clarifies a characterization mechanism of mapping the internal oil content to the external phenotypic characteristics of the olive fruits, realizes the intelligent picking of the olive fruits, and improves the oil yield of the fruits.
Description
Technical Field
The invention relates to the technical field of intelligent forestry and unmanned aerial vehicles, in particular to a method, a device and a system for estimating the oil content of olive fruits and a storage medium.
Background
The olive is a famous oil tree in the world and is one of main tree species for eating woody oil in China, and has extremely high nutritive value and health care function. The olive oil has unsaturated fatty acid content close to 90%, fatty acid balance mode close to that required by human body, and also contains fat-soluble vitamins such as A, D, E, K, etc., trace elements and other nutrients necessary for human body, and can reduce cholesterol and blood lipid, so it is known as the highest grade natural vegetable oil. However, the oil yield of olive fruits in China is not high at present, and the problems in the aspect of good-breed seedling culture of olive are solved, the main reason of analysis is that most of the olive fruits are planted in mountain high and steep slope regions, the harvest can be judged only by artificial experience, certain subjectivity exists, and the highest oil yield of the harvested fruits cannot be guaranteed. Therefore, research on an estimation method of the oil content of the olive fruits is a real demand which needs to be solved urgently.
The method has the advantages that the intelligent harvesting of the olive fruits is realized, the oil yield of the olive fruits is improved, one key problem lies in that the mapping relation between the external phenotypic characteristics and the internal oil content of the olive fruits is revealed, and the change of the internal oil content is reflected through the change of the external phenotypic characteristics of the olive fruits; in order to clarify the mapping relationship between the external phenotypic characteristics and the internal oil content of olive fruits, the pigment content of the pericarp affecting the external phenotypic characteristics of the fruits must be considered, and how to link the laboratory-determined pigment content of the olive pericarp with the external phenotypic characteristics and the internal oil content is a hot issue of attention by researchers in the field. The combined olive and olive oil extraction-related researches show that the researches on the maturity detection and the mechanized picking mode of olive fruits are in progress; however, the research for establishing the mapping relation between the external phenotypic characteristics and the internal oil content of the fruits by taking the pigment content of the peels as a bridge is rarely related, and particularly, the related research for estimating the internal oil content from the external phenotypic characteristics of the olive fruits by combining the unmanned aerial vehicle technology is rarely reported.
Disclosure of Invention
The present invention has been made to solve the above-mentioned problems occurring in the prior art. Therefore, a method, a device and a system for estimating the oil content of the olive fruits and a storage medium are needed, the pigment content of the peel of the olive is taken as a bridge, a quantitative detection model of the fruit from the external phenotypic characteristic to the pigment content of the peel and then to the internal oil content is established, the characterization mechanism of mapping the internal oil content of the olive fruits to the external phenotypic characteristic is clarified, the intelligent picking of the olive fruits is realized, and the oil yield of the olive fruits is improved.
According to a first aspect of the present invention, there is provided a method for estimating the oil content of olive fruits, the method comprising: based on a fruit image shot by an unmanned aerial vehicle, performing correlation analysis on the peel pigment content and the oil content of a corresponding fruit in the fruit image, screening out a pigment index which is remarkably related to the oil content in the olive, and establishing a dynamic mapping model; using a fruit image shot by the unmanned aerial vehicle as a training image, inputting a fruit peel pigment of a corresponding fruit in the fruit image as a true value into a fruit peel pigment content prediction model for training, and obtaining an olive fruit peel pigment content prediction model; predicting the fruit image collected in real time by using the olive fruit peel pigment content prediction model to obtain the fruit peel pigment content; and based on the dynamic mapping model, obtaining the internal oil content of the fruit according to the pigment content of the pericarp of the fruit.
According to a second aspect of the present invention, there is provided an apparatus for estimating an oil content of olive fruit, the apparatus comprising: the dynamic mapping model establishing module is configured to perform correlation analysis on the peel pigment content and the oil content of corresponding fruits in the fruit images based on the fruit images shot by the unmanned aerial vehicle, screen out pigment indexes which are obviously related to the oil content in the olives and establish a dynamic mapping model; the prediction model training module is configured to utilize a fruit image shot by the unmanned aerial vehicle as a training image, and a fruit peel pigment of a corresponding fruit in the fruit image is used as a true value to be input into the fruit peel pigment content prediction model for training so as to obtain an olive fruit peel pigment content prediction model; the fruit peel pigment content prediction module is configured to predict a fruit image acquired in real time by using the olive fruit peel pigment content prediction model to obtain the fruit peel pigment content of the fruit; and the internal oil content estimation module is configured to obtain the internal oil content of the fruit according to the pigment content of the peel of the fruit based on the dynamic mapping model.
According to a third aspect of the present invention, there is provided a system for estimating the oil content of olive fruits, the system comprising: a memory for storing a computer program; a processor for executing the computer program to implement the method as described above.
According to a fourth aspect of the present invention, there is provided a non-transitory computer readable storage medium having stored thereon instructions which, when executed by a processor, perform the method as described above.
According to the method, the device and the system for estimating the oil content of the olive fruits and the storage medium, which are provided by the various schemes of the invention, at least the following technical effects are achieved:
the method takes the pigment content of the olive fruit peel as a bridge, and respectively establishes a prediction model of the pigment content of the fruit peel and the phenotypic characteristics of the olive fruit and a dynamic mapping model of the pigment content of the fruit peel and the internal oil content. By extracting the phenotype characteristics of the olive images in the natural state of aerial photography by the unmanned aerial vehicle, the content of peel pigment of fruits in each image is predicted. And finally, obtaining the oil content of the fruit through a dynamic mapping model of the peel pigment and the internal oil content.
Drawings
In the drawings, which are not necessarily drawn to scale, like reference numerals may describe similar components in different views. Like reference numerals having alphabetic suffixes or different alphabetic suffixes may represent different instances of similar components. The drawings illustrate various embodiments, by way of example and not by way of limitation, and together with the description and claims, serve to explain the inventive embodiments. The same reference numbers will be used throughout the drawings to refer to the same or like parts, where appropriate. Such embodiments are illustrative, and are not intended to be exhaustive or exclusive embodiments of the present apparatus or method.
Fig. 1 shows a flow chart of a method for estimating the oil content of olive fruits according to an embodiment of the present invention.
FIG. 2 shows a flow chart of dynamic mapping model establishment according to an embodiment of the invention.
FIG. 3 shows a flow chart of training of a predictive model according to an embodiment of the invention.
FIG. 4 is a block diagram of a model for predicting the pigment content of fruit peel in accordance with an embodiment of the present invention.
Figure 5 shows a flow chart of drone altitude determination according to an embodiment of the invention,
fig. 6 is a block diagram illustrating an apparatus for estimating oil content of olive fruit according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following detailed description of embodiments of the invention is provided in connection with the accompanying drawings and the detailed description of embodiments of the invention, but is not intended to limit the invention. The order in which the various steps described herein are described as examples should not be construed as a limitation if there is no requirement for a context relationship between each other, and one skilled in the art would know that sequential adjustments may be made without destroying the logical relationship between each other, rendering the overall process impractical.
The embodiment of the invention provides a method for estimating the oil content of olive fruits, please refer to fig. 1, which is a flowchart of the method for estimating the oil content of olive fruits according to the embodiment of the invention. The method starts in step S100, correlation analysis is carried out on the peel pigment content and the oil content of corresponding fruits in the fruit images based on the fruit images shot by the unmanned aerial vehicle, pigment indexes which are obviously related to the oil content in the olives are screened out, and a dynamic mapping model is established. It should be noted that the peel pigment content and the oil content of the corresponding fruit in the fruit image described herein are relatively accurate data obtained by performing experimental tests on the olive fruits photographed by the unmanned aerial vehicle after picking, wherein the peel pigment content includes, but is not limited to, chlorophyll content and anthocyanin content.
In some embodiments, please refer to fig. 2, which is a flowchart illustrating the establishment of a dynamic mapping model according to an embodiment of the present invention. The step S100 is to perform correlation analysis on the peel pigment content and the oil content of the corresponding fruit in the fruit image based on the fruit image shot by the unmanned aerial vehicle, screen out a pigment index that is significantly correlated with the oil content in the olive and establish a dynamic mapping model, and specifically includes:
and S101, analyzing the relationship among the chlorophyll content, the anthocyanin content, the chlorophyll content/anthocyanin content and the internal oil content according to a scatter diagram corresponding to the chlorophyll content, the anthocyanin content, the chlorophyll content/anthocyanin content and the internal oil content of the fruit in the fruit image, and carrying out significance test to obtain the correlation between the internal oil content and different pigment indexes. Merely by way of example, the chlorophyll content, anthocyanin content, and the relationship between chlorophyll/anthocyanin content and internal oil content can be analyzed by the R language and tested for significance to obtain correlations between internal oil content and different pigment indicators.
And S102, respectively establishing relation curves of chlorophyll content and internal oil content, anthocyanin content and internal oil content, and chlorophyll content/anthocyanin content and internal oil content according to the correlation between the internal oil content and different pigment indexes, so as to obtain a dynamic mapping model of the olive pericarp pigment content and the internal oil content. The relationship between the chlorophyll content and the anthocyanin content and the internal oil content is shown in the graph, which shows the relationship between the chlorophyll content and the anthocyanin content and the internal oil content when the olive fruit contains both chlorophyll and anthocyanin.
The following example will specifically describe how the chlorophyll content, anthocyanin content, and internal oil content of a fruit can be accurately measured experimentally.
In some embodiments, the chlorophyll content of the fruit is calculated by:
picking olive fruits shot by an unmanned aerial vehicle, and obtaining fresh olive peel;
adding fresh olive peel to liquid nitrogen, grinding the frozen olive peel, mixing with 95% ethanol, extracting in dark for 48h, centrifuging the extract at 4 deg.C for 5 min, collecting the supernatant and measuring its absorbance at 649nm and 665nm using a spectrophotometer;
calculating the chlorophyll content of the fruit according to the following formula (2):
wherein ChI is chlorophyll content, OD 649 The absorbance, OD, of the extract at 649nm 665 The absorbance of the extract at 665nm was used.
In some embodiments, the anthocyanin content of the fruit is calculated by:
picking olive fruits shot by an unmanned aerial vehicle, and obtaining fresh olive peel;
extracting the finely ground olive pericarp with 60% ethanol at 40 deg.C, then filtering, and concentrating the filtrate under reduced pressure; two solutions were extracted from the concentrate and diluted with buffer a and buffer B, which were acidic solutions at pH =1.0 and pH =4.5, respectively, and finally the absorbance at 520nm was measured with a spectrophotometer. For example only, buffer A may be selected to be 25 v/v 0.2 mol.L -1 KCl/0.2 mol·L -1 HCl, pH =1.0, buffer B can be selected as 100 -1 CH 3 COONa/1 mol·L -1 HCl/H 2 O,pH=4.5;
The anthocyanin content of the fruit was calculated according to the following formula (3):
wherein A is 0 And A 1 Is the absorbance of anthocyanins at pH =1.0 and pH =4.5, V is the concentrate volume, N is the dilution factor, M is the molecular weight of anthocyanins, e is the standard extinction factor, M is the peel mass.
In some embodiments, the internal oil content of the fruit is calculated by:
picking olive fruits shot by an unmanned aerial vehicle;
putting olive fruits into a 65 ℃ oven to be dried until the mass of the olive fruits does not change, uniformly grinding the dried fruits in a grinder, putting a ground sample into an extraction filter paper cylinder, putting absorbent cotton into the extraction filter paper cylinder, sealing the filter paper cylinder, putting the filter paper cylinder into a Soxhlet extractor, weighing an extraction bottom bottle, adding petroleum ether into the extraction bottom bottle, connecting and pressing the petroleum ether with an extraction column, extracting for 6 to 8 hours, recovering the petroleum ether, taking down the extraction bottom bottle, putting the extraction bottom bottle into a 105 ℃ oven to be dried for 10 to 20 minutes until the mass of the extraction bottom bottle does not change, cooling to room temperature, and recording the mass of the extracted oil and the extraction bottom bottle;
the internal oil content of the fruit was calculated according to the following formula (4):
whereinMThe content of the oil in the olive is shown,m 1 is the weight of the dried olive fruits,m 2 in order to extract the weight of the bottom flask,m 3 to extract the weight of the bottom flask and olive oil.
And S200, taking the fruit image shot by the unmanned aerial vehicle as a training image, taking the peel pigment of the corresponding fruit in the fruit image as a true value, inputting a peel pigment content prediction model to train the fruit image, and obtaining an olive peel pigment content prediction model.
In some embodiments, please refer to fig. 3, which is a flowchart illustrating a training process of a prediction model according to an embodiment of the present invention. The step S200 of using the fruit image shot by the unmanned aerial vehicle as a training image, inputting the pericarp pigment of the corresponding fruit in the fruit image as a true value into a pericarp pigment content prediction model to train the fruit image, and obtaining an olive pericarp pigment content prediction model, specifically includes:
step S201, expanding the fruit image shot by the unmanned aerial vehicle by turning, adjusting the image brightness and randomly adding noise, and performing the steps of: 1: and 1, dividing the expanded image into a training set, a verification set and a test set in a proportion.
And S202, training a fruit peel pigment content prediction model based on the training set and the fruit peel pigment content of the corresponding fruit in the fruit image, determining an optimal prediction model according to the difference value of the prediction output of the trained fruit peel pigment content prediction model to the test set and the fruit peel pigment data of the real test, and obtaining the olive fruit peel pigment content prediction model.
In the training process, parameters such as learning rate, batch processing amount, dropot proportion and the like are set, the prediction result of the model and the real determination result are reduced as much as possible, the optimal prediction model is determined according to the difference value of the prediction output of the trained fruit peel pigment content prediction model to the test set and the fruit peel pigment data of the real test, and the olive fruit peel pigment content prediction model is obtained. The best prediction model is the prediction model with the minimum difference between the prediction output of the trained fruit peel pigment content prediction model to the test set and the fruit peel pigment data of the real test, and the model is used as the olive fruit peel pigment content prediction model.
It should be noted that the peel pigment content prediction model described herein is an existing convolutional neural network model, such as SSD, YOLO, faster R-CNN, efficientDet, etc., and the embodiment is not limited in this respect. For example only, please refer to fig. 4, which is a block diagram of a fruit peel pigment content prediction model according to an embodiment of the present invention. In the training process, key phenotype characteristics of olive fruits are extracted from fruit images shot by an unmanned aerial vehicle through operations such as convolution, pooling and the like, the content of chlorophyll and anthocyanin which are predicted is output according to the key phenotype characteristics, parameters are adjusted according to the content of peel pigment of corresponding fruits in the fruit images, weight parameters in the training process are stored, different weight parameters are used for configuring peel pigment content prediction models, then a test set is input for testing, a prediction model with the minimum difference value between the prediction output and the peel pigment data which are actually tested is selected as an olive peel pigment content prediction model, an accurate prediction model can be obtained, the olive peel pigment content corresponding to the fruits in the fruit images can be intelligently identified based on the fruit images shot by the unmanned aerial vehicle, and preparation is made for the subsequent internal oil content prediction.
Illustratively, the key phenotypic characteristic may be a color characteristic of the fruit, that is, different pericarp pigment contents are corresponded according to different color characteristics of the fruit. Of course, the key phenotypic characteristic may also be other characteristics, such as fruit size characteristics (including transverse diameter, longitudinal diameter, etc.), which may be one characteristic, or a combination of characteristics, such as fruit color characteristics plus size characteristics, etc. The present embodiment is not particularly limited herein.
And S203, verifying the olive peel pigment content prediction model by using the verification set so as to determine whether the olive peel pigment content prediction model can be accurately measured. The verification method can judge whether the prediction effect of the model is within a preset error range through the error between the prediction output of the model and the actually tested peel pigment data. If the overall error is too large, the process returns to step S201, a new image data set is added, and step S202 is executed again, so that the error is within the acceptable range, and an accurate fruit peel pigment content prediction model is obtained.
It should be noted that, when actually executing step S100 and step S200, step S100 or step S200 may not be executed sequentially, that is, step S100 or step S200 may be executed first, or both steps may be executed simultaneously. The order of the steps in this embodiment is only an example, and does not limit the present invention.
And step S300, predicting the fruit image collected in real time by using the olive peel pigment content prediction model to obtain the peel pigment content of the fruit.
For example, the olive peel pigment content prediction model is mounted on the unmanned aerial vehicle, and for example, the olive peel pigment content prediction model may be configured on an intelligent control terminal of the unmanned aerial vehicle. In the process of aerial photography by the unmanned aerial vehicle, the fruit images collected in real time can be input into the olive peel pigment content prediction model for prediction, the model extracts key phenotype characteristics of olive fruits through operations such as convolution, pooling and the like, and predicted chlorophyll and anthocyanin contents are output according to the key phenotype characteristics.
And step S400, based on the dynamic mapping model, obtaining the internal oil content of the fruit according to the pigment content of the peel of the fruit.
For example, the dynamic mapping model is mounted on the unmanned aerial vehicle, based on the content of the peel pigment obtained in step S300, the content of chlorophyll and anthocyanin output by the prediction model is input into the dynamic mapping model of the content of the peel pigment and the internal oil content, and the internal oil content is obtained comprehensively according to the relationship curve of chlorophyll and the internal oil content and the relationship curve of anthocyanin and the internal oil content.
In some embodiments, the method further includes determining the flying altitude of the drone, please refer to fig. 5, which is a flowchart of determining the flying altitude of the drone according to an embodiment of the present invention, specifically including:
step S1001, shooting the olive image of each inflorescence in the natural state of different varieties according to the unmanned aerial vehicle, and taking the average value of the number of circumscribed circle pixels of the fruit connected with each inflorescence as the average pixel number of the olive fruit type index.
Step S1002, calculating the fruit type index of each fruit according to the actual transverse diameter and longitudinal diameter of each fruit, and taking the average value of the fruit type index of each inflorescence fruit as the average fruit type index of the olive fruit;
step S1003, calculating the flying height of the unmanned aerial vehicle according to the following formula (1):
whereinHIs the flight height of the image acquisition of the unmanned aerial vehicle,F r to take a photographThe focal length of the lens of the image head,GSDfor the purpose of the ground resolution,S w is the size of the picture element of the camera,Wto be the width of the image,V t the average fruit type index of the fruit grafted on each inflorescence,V px is the average pixel number of the fruit index.
Fig. 6 shows a structure diagram of an oil content estimation device for olive fruits according to an embodiment of the present invention. The apparatus 600 comprises:
the dynamic mapping model establishing module 601 is configured to perform correlation analysis on the peel pigment content and the oil content of a corresponding fruit in a fruit image based on the fruit image shot by the unmanned aerial vehicle, screen out a pigment index which is obviously related to the oil content in the olive and establish a dynamic mapping model;
a prediction model training module 602 configured to input a fruit peel pigment content prediction model as a true value to train a fruit image shot by the unmanned aerial vehicle, wherein the fruit peel pigment of a corresponding fruit in the fruit image is used as the true value, and the fruit peel pigment content prediction model is obtained;
the fruit peel pigment content prediction module 603 is configured to predict a fruit image acquired in real time by using the olive fruit peel pigment content prediction model to obtain the fruit peel pigment content of the fruit;
and an internal oil content estimation module 604 configured to obtain the internal oil content of the fruit according to the pigment content of the peel of the fruit based on the dynamic mapping model.
In some embodiments, the dynamic mapping model building module 601 is further configured to analyze the relationship between the chlorophyll content, anthocyanin content, and chlorophyll content/anthocyanin content and the internal oil content according to a scatter plot of the chlorophyll content, anthocyanin content, chlorophyll content/anthocyanin content, and internal oil content of the corresponding fruit in the fruit image and perform a significance check to obtain the correlation between the internal oil content and different pigment indexes; and respectively establishing a relation curve of the chlorophyll content and the internal oil content, the anthocyanin content and the internal oil content, and a relation curve of the chlorophyll content/anthocyanin content and the internal oil content according to the correlation between the internal oil content and different pigment indexes so as to obtain a dynamic mapping model of the olive pericarp pigment content and the internal oil content.
In some embodiments, the predictive model training module 602 is further configured to: the fruit images shot by the unmanned aerial vehicle are expanded by turning, adjusting the image brightness and randomly adding noise, and the method comprises the following steps of: 1:1, dividing the expanded image into a training set, a verification set and a test set in proportion; training a fruit peel pigment content prediction model based on the training set and the fruit peel pigment content of the corresponding fruit in the fruit image, determining an optimal prediction model according to the difference value between the prediction output of the trained fruit peel pigment content prediction model to the test set and the fruit peel pigment data of the real test, and obtaining an olive fruit peel pigment content prediction model; and verifying the olive peel pigment content prediction model by using the verification set.
In some embodiments, the apparatus further comprises a fly-height determination module configured to:
shooting the olive image of each inflorescence in a natural state of different varieties according to an unmanned aerial vehicle, and taking the average value of the number of circumscribed circle pixels of the fruit connected with each inflorescence as the average pixel number of the olive fruit type index;
calculating the fruit type index of each fruit according to the actual transverse diameter and longitudinal diameter of each fruit, and taking the average value of the fruit type indexes of the fruits of each inflorescence as the average fruit type index of the olive fruits;
calculating the flying height of the unmanned aerial vehicle according to the following formula (1):
whereinHIs the flight height of the image acquisition of the unmanned aerial vehicle,F r is the focal length of the lens of the camera,GSDfor the purpose of the ground resolution,S w is the size of the picture element of the camera,Wis the width of the image to be displayed,V t is each one ofAverage fruit type index of fruit grafted by an inflorescence,V px is the average pixel number of the fruit index.
It should be noted that the modules described in the embodiments of the present invention may be implemented by software, or may be implemented by hardware, and the described modules may also be disposed in a processor. Wherein the names of the modules do not in some cases constitute a limitation of the module itself.
The device for estimating the oil content of the olive fruits, which is provided by the embodiment of the invention, belongs to the same technical concept as the method explained in the foregoing, and the technical effects are basically consistent, so that the details are not repeated.
The embodiment of the invention also provides a system for estimating the oil content of the olive fruits, which comprises:
a memory for storing a computer program;
a processor for executing the computer program to implement the method of any of the embodiments of the invention.
Embodiments of the present invention also provide a non-transitory computer readable medium storing instructions that, when executed by a processor, perform a method according to any of the embodiments of the present invention.
Moreover, although exemplary embodiments have been described herein, the scope thereof includes any and all embodiments based on the present invention with equivalent elements, modifications, omissions, combinations (e.g., of various embodiments across), adaptations or alterations. The elements of the claims are to be interpreted broadly based on the language employed in the claims and not limited to examples described in the present specification or during the prosecution of the application, which examples are to be construed as non-exclusive. It is intended, therefore, that the specification and examples be considered as exemplary only, with a true scope and spirit being indicated by the following claims and their full scope of equivalents.
The above description is intended to be illustrative and not restrictive. For example, the above-described examples (or one or more versions thereof) may be used in combination with each other. For example, other embodiments may be used by those of ordinary skill in the art upon reading the above description. In addition, in the above-described embodiments, various features may be grouped together to streamline the disclosure. This should not be interpreted as an intention that features of an unclaimed invention be essential to any of the claims. Rather, inventive subject matter may lie in less than all features of a particular inventive embodiment. Thus, the following claims are hereby incorporated into the detailed description as examples or embodiments, with each claim standing on its own as a separate embodiment, and it is contemplated that these embodiments may be combined with each other in various combinations or permutations. The scope of the invention should be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.
Claims (10)
1. A method for estimating the oil content of olive fruits, which is characterized by comprising the following steps:
based on a fruit image shot by an unmanned aerial vehicle, performing correlation analysis on the peel pigment content and the oil content of a corresponding fruit in the fruit image, screening out a pigment index which is remarkably related to the oil content in the olive, and establishing a dynamic mapping model;
using a fruit image shot by the unmanned aerial vehicle as a training image, inputting a fruit peel pigment of a corresponding fruit in the fruit image as a true value into a fruit peel pigment content prediction model for training, and obtaining an olive fruit peel pigment content prediction model;
predicting the fruit image collected in real time by using the olive peel pigment content prediction model to obtain the peel pigment content of the fruit;
and based on the dynamic mapping model, obtaining the internal oil content of the fruit according to the pigment content of the pericarp of the fruit.
2. The method according to claim 1, wherein the pigment indexes comprise chlorophyll content and anthocyanin content, and the correlation analysis is performed on the peel pigment content and the oil content of the corresponding fruit in the fruit image based on the fruit image shot by the unmanned aerial vehicle, so that the pigment indexes which are significantly related to the oil content in the olive are screened out, and a dynamic mapping model is established, and the method comprises the following steps:
analyzing the relationship among the chlorophyll content, the anthocyanin content, the chlorophyll content/anthocyanin content and the internal oil content according to a scatter diagram corresponding to the chlorophyll content, the anthocyanin content, the chlorophyll content/anthocyanin content and the internal oil content of the fruit in the fruit image, and carrying out significance test to obtain the correlation between the internal oil content and different pigment indexes;
and respectively establishing a relation curve of the chlorophyll content and the internal oil content, the anthocyanin content and the internal oil content, and a relation curve of the chlorophyll content/anthocyanin content and the internal oil content according to the correlation between the internal oil content and different pigment indexes so as to obtain a dynamic mapping model of the olive pericarp pigment content and the internal oil content.
3. The method according to claim 1, wherein the using the fruit image shot by the unmanned aerial vehicle as a training image, inputting the peel pigment of the corresponding fruit in the fruit image as a true value into a peel pigment content prediction model for training, and obtaining an olive peel pigment content prediction model comprises:
the fruit images shot by the unmanned aerial vehicle are expanded by turning, adjusting the image brightness and randomly adding noise, and the method comprises the following steps of: 1:1, dividing the expanded images into a training set, a verification set and a test set in proportion;
training a fruit peel pigment content prediction model based on the training set and the fruit peel pigment content of the corresponding fruit in the fruit image, determining an optimal prediction model according to the difference value between the prediction output of the trained fruit peel pigment content prediction model to the test set and the fruit peel pigment data of the real test, and obtaining an olive fruit peel pigment content prediction model;
and verifying the olive peel pigment content prediction model by using the verification set.
4. The method of claim 1, further comprising:
shooting the olive image of each inflorescence in a natural state of different varieties according to an unmanned aerial vehicle, and taking the average value of the number of circumscribed circle pixels of the fruit connected with each inflorescence as the average pixel number of the olive fruit type index;
calculating the fruit type index of each fruit according to the actual transverse diameter and longitudinal diameter of each fruit, and taking the average value of the fruit type index of each inflorescence fruit as the average fruit type index of the olive fruit;
calculating the flight height of the unmanned aerial vehicle according to the following formula (1):
whereinHIs the flight height of the image acquisition of the unmanned aerial vehicle,F r is the focal length of the lens of the camera,GSDfor the purpose of the ground resolution,S w is the size of the picture element of the camera,Wto be the width of the image,V t the average fruit type index of the fruit grafted on each inflorescence,V px is the average pixel number of the fruit index.
5. The method according to claim 2, wherein the chlorophyll content of the fruit is calculated by:
picking olive fruits shot by an unmanned aerial vehicle, and obtaining fresh olive peel;
adding fresh olive pericarp to liquid nitrogen, then finely grinding the frozen olive pericarp, mixing with 95% ethanol, extracting in the dark, finally centrifuging the extract at 4 deg.C, collecting the supernatant and measuring its absorbance at 649nm and 665nm using a spectrophotometer;
calculating the chlorophyll content of the fruit according to the following formula (2):
wherein ChI is chlorophyll content, OD 649 Is the absorption of the extract at 649nmLight intensity, OD 665 The absorbance of the extract at 665nm was used.
6. The method according to claim 2, characterized in that the anthocyanin content of the fruit is calculated by:
picking olive fruits shot by an unmanned aerial vehicle, and obtaining fresh olive peel;
extracting the finely ground olive pericarp with 60% ethanol at 40 deg.C, then filtering, and concentrating the filtrate under reduced pressure; extracting two solutions from the concentrate, diluting with buffer a and buffer B, which are acidic solutions of pH =1.0 and pH =4.5, respectively, and finally measuring the absorbance at 520nm with a spectrophotometer;
the anthocyanin content of the fruit was calculated according to the following formula (3):
wherein A is 0 And A 1 Is the absorbance of anthocyanins at pH =1.0 and pH =4.5, V is the concentrate volume, N is the dilution factor, M is the molecular weight of anthocyanins, e is the standard extinction factor, M is the peel mass.
7. The method according to claim 2, wherein the internal oil content of the fruit is calculated by:
picking olive fruits shot by an unmanned aerial vehicle;
putting olive fruits into a 65 ℃ oven to be dried until the mass of the olive fruits does not change, uniformly grinding the dried fruits in a grinder, putting a ground sample into an extraction filter paper cylinder, putting absorbent cotton into the extraction filter paper cylinder, sealing the filter paper cylinder, putting the filter paper cylinder into a Soxhlet extractor, weighing an extraction bottom bottle, adding petroleum ether into the extraction bottom bottle, connecting and pressing the petroleum ether with an extraction column, extracting for 6 to 8 hours, recovering the petroleum ether, taking down the extraction bottom bottle, putting the extraction bottom bottle into a 105 ℃ oven to be dried for 10 to 20 minutes until the mass of the extraction bottom bottle does not change, cooling to room temperature, and recording the mass of the extracted oil and the extraction bottom bottle;
the internal oil content of the fruit was calculated according to the following formula (4):
whereinMThe content of the oil in the olive is shown,m 1 is the weight of the dried olive fruits,m 2 in order to extract the weight of the bottom flask,m 3 to extract the weight of the bottom flask and olive oil.
8. An apparatus for estimating the oil content of olive fruits, comprising:
the dynamic mapping model establishing module is configured to perform correlation analysis on the peel pigment content and the oil content of corresponding fruits in the fruit images based on the fruit images shot by the unmanned aerial vehicle, screen out pigment indexes which are obviously related to the oil content in the olives and establish a dynamic mapping model;
the prediction model training module is configured to input a fruit peel pigment content prediction model as a true value to train the fruit peel pigment content prediction model by using a fruit image shot by the unmanned aerial vehicle as a training image, and obtain an olive fruit peel pigment content prediction model;
the fruit peel pigment content prediction module is configured to predict a fruit image acquired in real time by using the olive fruit peel pigment content prediction model to obtain the fruit peel pigment content of the fruit;
and the internal oil content estimation module is configured to obtain the internal oil content of the fruit according to the pigment content of the peel of the fruit based on the dynamic mapping model.
9. A system for estimating the oil content of olive fruits is characterized in that: the system comprises:
a memory for storing a computer program;
a processor for executing the computer program to implement the method of any one of claims 1 to 4.
10. A non-transitory computer-readable storage medium having instructions stored thereon, which when executed by a processor, perform the method of any one of claims 1-4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210961085.4A CN115205716B (en) | 2022-08-11 | 2022-08-11 | Method, device and system for estimating oil content of olive fruits and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210961085.4A CN115205716B (en) | 2022-08-11 | 2022-08-11 | Method, device and system for estimating oil content of olive fruits and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115205716A true CN115205716A (en) | 2022-10-18 |
CN115205716B CN115205716B (en) | 2023-04-07 |
Family
ID=83586232
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210961085.4A Active CN115205716B (en) | 2022-08-11 | 2022-08-11 | Method, device and system for estimating oil content of olive fruits and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115205716B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117660106A (en) * | 2023-12-29 | 2024-03-08 | 山东兴泉油脂有限公司 | Preparation monitoring method and system of nutritional healthy corn oil |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5764819A (en) * | 1991-10-18 | 1998-06-09 | Dekalb Genetics Corporation | Methods for classifying plants for evaluation and breeding programs by use of remote sensing and image analysis technology |
JP2005315877A (en) * | 2004-03-31 | 2005-11-10 | Keita Sugiyama | Method of determining quantitatively pigment content in vegetable using image analysis |
US20060062487A1 (en) * | 2002-10-15 | 2006-03-23 | Makoto Ouchi | Panorama synthesis processing of a plurality of image data |
CN110736709A (en) * | 2019-10-26 | 2020-01-31 | 苏州大学 | blueberry maturity nondestructive testing method based on deep convolutional neural network |
CN112254713A (en) * | 2020-10-20 | 2021-01-22 | 中国石油大学(华东) | Unmanned aerial vehicle oblique photography parameter determination method for tall and large dense building group |
CN113192026A (en) * | 2021-04-28 | 2021-07-30 | 北京林业大学 | Walnut maturity detection and prediction method based on visible light image |
CN114112992A (en) * | 2022-01-20 | 2022-03-01 | 武汉工程大学 | Detection method and device for blue pigment of cream and storage medium |
-
2022
- 2022-08-11 CN CN202210961085.4A patent/CN115205716B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5764819A (en) * | 1991-10-18 | 1998-06-09 | Dekalb Genetics Corporation | Methods for classifying plants for evaluation and breeding programs by use of remote sensing and image analysis technology |
US20060062487A1 (en) * | 2002-10-15 | 2006-03-23 | Makoto Ouchi | Panorama synthesis processing of a plurality of image data |
JP2005315877A (en) * | 2004-03-31 | 2005-11-10 | Keita Sugiyama | Method of determining quantitatively pigment content in vegetable using image analysis |
CN110736709A (en) * | 2019-10-26 | 2020-01-31 | 苏州大学 | blueberry maturity nondestructive testing method based on deep convolutional neural network |
CN112254713A (en) * | 2020-10-20 | 2021-01-22 | 中国石油大学(华东) | Unmanned aerial vehicle oblique photography parameter determination method for tall and large dense building group |
CN113192026A (en) * | 2021-04-28 | 2021-07-30 | 北京林业大学 | Walnut maturity detection and prediction method based on visible light image |
CN114112992A (en) * | 2022-01-20 | 2022-03-01 | 武汉工程大学 | Detection method and device for blue pigment of cream and storage medium |
Non-Patent Citations (4)
Title |
---|
ELISA FIERINI; LISA BANELLI; PATRIZIA PINELLI; ANNALISA ROMANI; DAMIANO REMORINI; GIOVANNI AGATI: "In situ assessment of quality-related compounds in fruits by using fluorescence sensors", 《IEEE》 * |
孔维宝;李万武;邢文黎;董妙音;韩锐;赵维;白万明;白小勇;张继;: "武都主栽油橄榄品种的果实品质研究", 中国粮油学报 * |
彭邵锋;陈永忠;张日清;杨小胡;王湘南;陆佳;: "油茶果形果色分类及经济性状", 中南林业科技大学学报 * |
柴阿丽;李宝聚;王倩;石延霞;黄海洋;: "基于计算机视觉技术的番茄叶片叶绿素含量的检测", 园艺学报 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117660106A (en) * | 2023-12-29 | 2024-03-08 | 山东兴泉油脂有限公司 | Preparation monitoring method and system of nutritional healthy corn oil |
Also Published As
Publication number | Publication date |
---|---|
CN115205716B (en) | 2023-04-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Barbedo et al. | Identifying multiple plant diseases using digital image processing | |
CN108444928A (en) | A method of identifying cereal seed frostbite situation using embryo chromatogram characteristic wave band | |
CN108335300A (en) | A kind of food hyperspectral information analysis system and method based on CNN | |
CN109827957B (en) | Rice leaf SPAD value estimation method and system based on computer vision | |
CN115205716B (en) | Method, device and system for estimating oil content of olive fruits and storage medium | |
CN110736709A (en) | blueberry maturity nondestructive testing method based on deep convolutional neural network | |
CN105021617A (en) | Hyperspectral imaging-based measuring equipment of chlorophyll content of whole rice plant and method thereof | |
CN103528967B (en) | Overdone indigo fruit fruit identification method based on high spectrum image | |
CN107679569A (en) | Raman spectrum substance automatic identifying method based on adaptive hypergraph algorithm | |
CN108038499A (en) | A kind of seeds sorting technique and system based on deep learning | |
CN112183448B (en) | Method for dividing pod-removed soybean image based on three-level classification and multi-scale FCN | |
CN110516668A (en) | A kind of honey adulteration detection method and device based on high light spectrum image-forming technology | |
CN116129260A (en) | Forage grass image recognition method based on deep learning | |
Deshpande et al. | Automated detection of Diabetic Retinopathy using VGG-16 architecture | |
CN116026795A (en) | Rice grain quality character nondestructive prediction method based on reflection and transmission spectrum | |
CN116129096A (en) | Digital measurement system and method for potato root length | |
Srivastava et al. | Machine-vision based handheld embedded system to extract quality parameters of citrus cultivars | |
Ratha et al. | Papaya fruit maturity estimation using wavelet and ConvNET | |
Hortinela et al. | Classification of Cacao Beans Based on their External Physical Features Using Convolutional Neural Network | |
CN111398211A (en) | Information distinguishing and processing method of rhizoma atractylodis granules | |
CN117589767A (en) | Tobacco leaf harvesting time determining method, electronic equipment and storage medium | |
Zhang et al. | UAV mission height effects on wheat lodging ratio detection | |
CN104568639B (en) | A kind of determination method and apparatus of sugar degree | |
CN113256733A (en) | Camera spectral sensitivity reconstruction method based on confidence voting convolutional neural network | |
Qian et al. | Development and application of crop monitoring system for detecting chlorophyll content of tomato seedlings |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |