CN201508328U - Machine vision-based real-time detecting and grading device for pork appearance quality - Google Patents
Machine vision-based real-time detecting and grading device for pork appearance quality Download PDFInfo
- Publication number
- CN201508328U CN201508328U CN2009201194969U CN200920119496U CN201508328U CN 201508328 U CN201508328 U CN 201508328U CN 2009201194969 U CN2009201194969 U CN 2009201194969U CN 200920119496 U CN200920119496 U CN 200920119496U CN 201508328 U CN201508328 U CN 201508328U
- Authority
- CN
- China
- Prior art keywords
- pork
- quality
- camera
- muscle
- image
- 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.)
- Expired - Lifetime
Links
- 235000015277 pork Nutrition 0.000 title claims abstract description 69
- 230000003287 optical effect Effects 0.000 claims description 10
- 230000002459 sustained effect Effects 0.000 claims description 3
- 210000003205 muscle Anatomy 0.000 abstract description 54
- 238000012545 processing Methods 0.000 abstract description 16
- 238000000605 extraction Methods 0.000 abstract description 10
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000011156 evaluation Methods 0.000 abstract description 6
- 235000013305 food Nutrition 0.000 abstract description 4
- 238000007689 inspection Methods 0.000 abstract description 3
- 230000008676 import Effects 0.000 abstract 1
- 238000000034 method Methods 0.000 description 29
- 235000013372 meat Nutrition 0.000 description 25
- 239000004579 marble Substances 0.000 description 15
- 235000015278 beef Nutrition 0.000 description 12
- 238000010586 diagram Methods 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 5
- 238000011160 research Methods 0.000 description 4
- 210000000577 adipose tissue Anatomy 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 235000013594 poultry meat Nutrition 0.000 description 3
- 238000001303 quality assessment method Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 230000001953 sensory effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 208000003464 asthenopia Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 210000000845 cartilage Anatomy 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 239000000796 flavoring agent Substances 0.000 description 1
- 235000019634 flavors Nutrition 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000008595 infiltration Effects 0.000 description 1
- 238000001764 infiltration Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 108010048734 sclerotin Proteins 0.000 description 1
- 238000003307 slaughter Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
Images
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The utility model discloses a machine vision-based real-time detecting and grading device for pork appearance quality, which comprises a pork image acquisition hardware part and a grading software part, wherein the image acquisition hardware part comprises a camera, a lens, a polarizer, a light source, a computer and the like. The pork image is collected, and the grading software is utilized to carry out the real-time image processing and feature extraction; the characteristics information which reflects the appearance quality of pork is obtained; and the quality of pork is evaluated by re-using a prediction model, the grading evaluation of color, marbling and tenderness and comprehensive quality for eye muscle pork is completed, and the grading evaluation of muscle color, lean-to-fat ratio and comprehensive quality for streaky pork is completed. The utility model can achieve that the quality detection of pork in China has objectivity, accuracy and efficiency, can be suitable for supermarkets and local or import and export food inspection departments to evaluate the pork quality, and standardizes pork markets, thus truly pricing by quality.
Description
Technical field
The utility model relates to a kind of real-time detection grading plant based on machine vision, and particularly a kind of pork exterior quality based on machine vision detects grading plant in real time.
Background technology
Meat is one of common food of people, and according to statistics, 2000-2006 China Pork total production is progressively ascendant trend, and 4,031 ten thousand tons of 5,197 ten thousand tons of being increased to 2006 from 2000 increase by 5.8% every year on average, accounts for more than 60% of poultry meat total production.The occupation rate in market is very big at home to this shows pork, and people are increasing to the demand of pork.
Though China is meat big producing country, not meat industry big country, the export volume of meat is the trend that reduces year by year in recent years.The main cause that causes this phenomenon to produce is that mixed levels such as the meat of China mixes on market are more serious, does not realize fixing the price according to the quality as other developed country.Poultry meat classification standard is instructing meat production and is playing an important role in price.China has formulated classification standard to the ox trunk at present, beef is carried out classification from quality-class and two aspects of output level, wherein ox carcase quality grade be with manual observation after slaughtering, cooling off between ox trunk the 12nd～13 chest rib the terminal cartilage sclerotin of the vertebra apophysis of the marble grain grade at eye muscle tangent plane place, muscle color, fatty color and ox degree as judgment basis.Aspect the pork quality ranking, professor Chen Runsheng of Northeast Agricultural University has made pork colour and marble grain standards of grading figure.This figure adopts international five point scale as standards of grading.
Mainly be to carry out classification by visual inspection with according to the classification standard collection of illustrative plates for meat appearance quality detection and classification at present by grading person through training.Because grading person need carry out the comparison work between standard diagram and the actual sample for a long time under low temperature environment when classification, make classification person be easy to generate visual fatigue, thereby classification results is exerted an influence, the inevasible subjective factor that has of result of simultaneously artificial classification, but also need the result of classification be revised, thereby this subjectivity, poor efficiency, the needs in the more and more incompatible current market of hierarchical approaches that accuracy is little, and the emerging machine vision technique that grows up is quick with it, harmless, objective, efficiently, characteristic of accurate has replaced grading person's work gradually.
Domesticly carry out machine vision technique to detect the research of poultry meat quality mainly be at beef, Sun Yong sea for example, Zhao Xiwei, Xianyu builds the river and equals to be published in 2004 " agricultural mechanical journal " the 35th the 1st phase of volume " based on chilled beef freshness evaluation method of computer vision ", Ren Fazheng, Zheng Limin, Wang Guiqin, Liao Shuhua, Zhu Hong equals to be published in 2002 " meat research " application MATLAB image processing techniques in the first phase and passes judgment on beef marbling, the long English of the outstanding Ji of Chen Kun was published in " agricultural mechanical journal " the 38th the 5th phase of volume " based on beef marbling dividing method of image operation " in 2007, Tukon, Wang Fuchang was published in " applied research of computer vision in the beef marbling classification " of " grain and oil processing and food machinery " the 10th phase, Zhao Jiewen in 2004, Zou Xiaobo, Liu Muhua, the grand patent of declaring of yellow star " the computer vision detection and classification method and the device of beef carcase quality " in 2004.Image processing algorithm that above-mentioned document and patent the inside are introduced and pick-up unit all are special flavor evaluations at beef, but because pork and beef are in color, aspects such as texture all exist than big-difference, show that mainly fresh pig muscle is pale red and different parts meat color distortion is bigger, and the pork texture is than the beef exquisiteness, be prone to infiltration phenomenon etc. in the open when simultaneously pork is placed and cause the meat quality evaluation with respect to coming difficulty the beef, therefore early stage, the scholar detected the various image processing algorithms that classification proposes at beef and the device of building no longer is adapted to pork research.
Summary of the invention
In view of the deficiency that above-mentioned prior art exists, the purpose of this utility model is to provide a kind of pork exterior quality based on machine vision to detect grading plant in real time.Gather the pork image, carry out Flame Image Process, extract the characteristic information that characterizes pork exterior quality index according to existing grade scale, foundation detects the index quantification hierarchy model based on each individual event of image feature information, computes weighted according to each single index then and obtains pork integrated quality grade.
The technical solution adopted in the utility model is as follows:
Comprise that antinose-dive leg, four light source brackets, four light sources, computing machine, cameras articulate crossbeam, camera, optical lens, polariscope, cabinet exterior, objective table; Around four light source brackets are installed on the antinose-dive leg, four light source brackets can around the antinose-dive leg adjusted to sustained height and fixing, four light sources are separately fixed on separately the light source bracket, camera articulates crossbeam and is installed in the antinose-dive leg top, camera is fixed on the center that camera articulates crossbeam, and polariscope is spun on the optical frames optical lens, and objective table places the tank floor center position, whole box body forms closed chamber by wallboards of box body, and camera is connected by the USB line with computing machine.
Pork exterior quality based on machine vision detects stage division in real time:
Gather the pork image, carry out realtime graphic processing, feature extraction, utilize grading software that the exterior quality of pork is detected and classification then, its concrete steps are as follows:
1) set up pork exterior quality grading software system:
At first when the pork exterior quality grade scale of formulating according to China agricultural industry and consumer's actual purchase to the requirement of the index of quality, determine to detect index, set up pork exterior quality appraisement system; Then each exterior quality index of pork is carried out sensory evaluation scores; Pork after the scoring is gathered the pork image, image is handled, extract each characteristic index to be detected; Utilize the characteristics of image index and corresponding subjective appreciation score value foundation quantitative classification model separately extracted again; Set up pork integrated quality quantitative classification model according to each single index weight in integrated quality of surveying at last;
2) detect in real time:
When detecting in real time, carry out Flame Image Process, feature extraction behind the collection pork image, utilize single quality hierarchy model that single index is predicted then and obtain its corresponding grade, obtain the pork integrated level according to the integrated level quantitative model at last.
Described pork comprises that the exterior quality of eye muscle meat and streaky pork detects and classification, the characteristics of image index of extracting at eye muscle meat comprises muscle color, marble grain and texture, and the characteristics of image index of extracting at streaky pork comprises muscle color, muscle fat area ratio and muscle fat homogeneity.
The characteristics of image index of described extraction, when at be eye muscle meat the time, its concrete operations step comprises: remove noise and background segment, fat and musculature are cut apart, longissimus dorsi muscle extracted region, muscle color, marble grain and tender degree image characteristics extraction; When at be streaky pork the time, its concrete operations step comprises: remove noise, background segment, fat and cut apart and mark, calculating muscle color, muscle fat area ratio, muscle fat area homogeneity image feature information with muscle region.
The described detection index of determining according to national agricultural industry criteria is applicable to eye muscle meat, comprises muscle color, marble grain and tender degree; The index of during according to consumer's actual purchase the requirement of the index of quality being determined is applicable to streaky pork; Comprise muscle color, lean to fat ratio, girth of a garment homogeneity.
The beneficial effects of the utility model are:
Can fast and effeciently finish the detection and the classification of single index of quality such as muscle color, marble grain and tender degree and integrated quality to eye muscle meat; Finish the detection and the classification of single indexes such as muscle color, lean to fat ratio, girth of a garment homogeneity and integrated quality for streaky pork.Utilize the utility model that China meat quality is detected and possess objectivity, accuracy and high efficiency, and go for supermarket, place or gateway Food Inspection department meat quality is evaluated, standard pork market, thus real the realization fixed the price according to the quality.
Description of drawings
Fig. 1 apparatus structure synoptic diagram of the present utility model.
Fig. 2 whole software structure block diagram of the present utility model.
Fig. 3 eye muscle meat image processing process process flow diagram.
Fig. 4 longissimus dorsi muscle extracted region process flow diagram flow chart.
Fig. 5 streaky pork image processing process process flow diagram.
Among the figure: 1. antinose-dive leg; 2. light source bracket; 3. light source; 4. computing machine; 5. camera articulates crossbeam; 6. camera; 7. optical lens; 8. polariscope; 9. cabinet exterior; 10. objective table.
The utility model is described in further detail below in conjunction with drawings and embodiments.
The utility model is made up of image acquisition hardware part and exterior quality evaluation software section.
As shown in Figure 1, the utility model comprises that antinose-dive leg 1, four light source brackets 2, four light sources (cold-cathode tube) 3, computing machine 4, cameras articulate crossbeam 5, camera (industrial digital camera) 6, optical lens 7, polariscope 8, cabinet exterior 9, objective table 10; Around four light source brackets 2 are installed on the antinose-dive leg 1, four light source brackets 2 can around antinose-dive leg 1 adjusted to sustained height and fixing, four light sources 3 are separately fixed on separately the light source bracket 2, camera articulates crossbeam 5 and is installed in antinose-dive leg 1 top, camera 6 is fixed on the center that camera articulates crossbeam 5, polariscope 8 is spun on optical frames optical lens 7, objective table 10 places the tank floor center position, whole box body forms closed chamber by wallboards of box body 9, and camera 6 is connected by the USB line with computing machine 4.
Embodiment
Grading software partly comprises four modules such as image input device control module, image processing module, image characteristics extraction module, pork grade output module.
It is 2/3 times of tank floor width that described adjusting highly makes its distance apart from objective table, its calculate principle and method as follows: according to the Brewster principle: on nonmetallic interface, when reflection ray and refracted ray meet at right angles, reflection ray will be a linearly polarized light, the incident angle of this moment just is called Brewster angle, in the present embodiment: light is from air (medium 1, can regard vacuum as, refractive index n 1=1) injects moisture (medium 2, its refractive index n 2=1.33), tangent value tan θ=n2/n1=1.33 of Brewster angle θ then, thereby 53 ° of light source incident angle θ ≈, for this reason, four cold-cathode tubes are fixed on apart from illumination casing bottom width 2/3 place, thereby by polariscope can filtering pork surface the moisture imaging, thereby effectively reduce because the deviation that the moisture reflective phenomenon brings to subsequent characteristics extraction work.
As shown in Figure 2, pork exterior quality evaluation software section comprises four modules such as image input device control module, image processing module, image characteristics extraction module, pork grade output module, wherein the function of image input device control module is to gather the pork image in real time, the function of image processing module is the image pre-service and cuts apart, the function of image characteristics extraction module is for extracting yellowish pink, marble grain and textural characteristics, and the function of pork grade output module is yellowish pink, marble grain and texture grade and integrated level output.
With above-mentioned hardware components eye muscle meat and the streaky pork of buying carried out image acquisition and various processing, set up pork exterior quality assessment system.
Described process flow diagram to pig eye muscle meat image processing process as shown in Figure 3, concrete processing procedure is as follows:
(1) at first utilizes maximum variance adaptive threshold method that black background is removed from pork eye muscle image, then adopt median filtering method to carry out denoising;
(2) utilize nuclear fuzzy C-means clustering (KFCM) in RGB and hsv color space, to cut apart muscle and adipose tissue;
(3) with improved watershed algorithm longissimus dorsi muscle is extracted from the eye muscle image; Utilize the longissimus dorsi muscle image to extract the image texture characteristic of reflection pork tenderness information then;
(4) utilizing (2) image that obtains of step and (3) to go on foot the image that obtains performs mathematical calculations and extracts musculature and adipose tissue;
Described with improving the method that watershed algorithm extracts the longissimus dorsi muscle zone, may further comprise the steps:
(1) image after cutting apart through KFCM being carried out the cavity fills;
(2) utilize to improve watershed algorithm eliminate around being connected between musculature and the longissimus dorsi muscle, fill through the cavity, proceed condition expansion after improving limit corrosion, as shown in Figure 4;
(3) extract the longest zone of the back of the body, by the area of each connected region of search after (2) step handled, the zone of finding out maximum area is the position at longissimus dorsi muscle place in image.
Described to streaky pork Flame Image Process flow process as shown in Figure 5, concrete processing procedure is as follows:
(1) employing utilizes maximum variance adaptive threshold method to remove background, then median filtering method denoising;
(2) utilize KFCM to cut apart muscle and adipose tissue;
(3) carry out the morphology opening operation and eliminate tiny coupling part between fat and the muscle region;
(4) muscle in the image and fat region are carried out zone marker;
The described pork exterior quality assessment system of setting up, set up like this:
(1), eye muscle meat is extracted the feature (mean value and the standard deviation that comprise R, G, B, H, S, I) of reflection muscle color characteristic information, the feature of reflection marble grain characteristic information (comprising that large scale marble grain, medium size marble grain and small size marble grain and institute marbled number, area account for the ratio of the eye muscle total area), reflects that the feature of tender degree information (comprises 5 textural characteristics such as energy, entropy, correlativity, contrast, reciprocal difference square and maximum probability that utilize gray level co-occurrence matrixes to extract through after the above-mentioned Flame Image Process; 5 textural characteristics such as the long stroke that utilizes the length of stroke method to extract increases the weight of, short stroke increases the weight of, the homogeneity of the homogeneity of gray level, length of stroke, stroke number percent; Utilize 5 textural characteristics such as contrast that the grey scale difference statistic law calculates, angle second moment, entropy, mean value, unfavourable balance square); The feature (mean value and the standard deviation that comprise R, G, B, H, S, I), muscle fat area that streaky pork is extracted muscle color characteristic information than, reflection girth of a garment homogeneity characteristic information feature: all fat, muscle zonule area variance;
(2) utilize the principle component regression method to choose eye muscle meat color respectively, marble grain, the major component component of textural characteristics and streaky pork muscle color, muscle fat area ratio, the inhomogeneity major component component of the girth of a garment, the major component component of then PCA being selected is respectively as neural network, support vector machine, the input of multiple linear regression analysis, set up the eye muscle yellowish pink respectively, relational model between marble grain and tender degree and the sensory evaluation scores, streaky pork muscle color, muscle fat area ratio, relational model between girth of a garment homogeneity and the streaky pork grading, and the models of three kinds of methods foundation of comparison, select the forecast model of optimization model as each single index, the integrated level evaluation model that respectively single index is separately computed weighted and obtains eye muscle meat and streaky pork again, and then finish the foundation of pork exterior quality assessment system.
The method that utilization is introduced above realizes the establishment of pork exterior quality grading software on the VC++6.0 development platform, can fast and effeciently finish real-time collection, processing, feature extraction and the pattern-recognition of pork image.When this device is used for supermarket or inspection and quarantine department detection, eye muscle meat or streaky pork are positioned on the objective table of tank floor, utilize grading software can obtain the grade of each single index then, each single index is input to go in the unified model to compute weighted then and can obtains integrated level.Can realize the ranking of yellowish pink, marble grain and tender degree and integrated quality to eye muscle meat; Can realize the ranking of yellowish pink and integrated quality to streaky pork.
Claims (1)
1. the pork exterior quality based on machine vision detects grading plant in real time, it is characterized in that: comprise that antinose-dive leg (1), four light source brackets (2), four light sources (3), computing machine (4), camera articulate crossbeam (5), camera (6), optical lens (7), polariscope (8), cabinet exterior (9), objective table (10); Around four light source brackets (2) are installed on the antinose-dive leg (1), four light source brackets (2) can around antinose-dive leg (1) adjusted to sustained height and fixing, four light sources (3) are separately fixed on separately the light source bracket (2), camera articulates crossbeam (5) and is installed in antinose-dive leg (1) top, camera (6) is fixed on the center that camera articulates crossbeam (5), polariscope (8) is spun on optical frames optical lens (7), objective table (10) places the tank floor center position, whole box body forms closed chamber by wallboards of box body (9), and camera (6) is connected by the USB line with computing machine (4).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009201194969U CN201508328U (en) | 2009-05-07 | 2009-05-07 | Machine vision-based real-time detecting and grading device for pork appearance quality |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009201194969U CN201508328U (en) | 2009-05-07 | 2009-05-07 | Machine vision-based real-time detecting and grading device for pork appearance quality |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201508328U true CN201508328U (en) | 2010-06-16 |
Family
ID=42469455
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009201194969U Expired - Lifetime CN201508328U (en) | 2009-05-07 | 2009-05-07 | Machine vision-based real-time detecting and grading device for pork appearance quality |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201508328U (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855640A (en) * | 2012-08-10 | 2013-01-02 | 上海电机学院 | Fruit grading system based on neural network |
| WO2017133178A1 (en) * | 2016-02-02 | 2017-08-10 | 意力(广州)电子科技有限公司 | Automatic optical inspection program-based touch panel automatic inspection apparatus |
| CN109459391A (en) * | 2019-01-07 | 2019-03-12 | 塔里木大学 | A kind of jujube Quality Detection, jujube Polarization Detection model generating method and device |
| CN110189148A (en) * | 2019-05-29 | 2019-08-30 | 广州影子科技有限公司 | Retroactive method, retrospective device and traceability system |
| CN112683899A (en) * | 2020-04-29 | 2021-04-20 | 海南远生渔业有限公司 | Aquatic product quality detection method |
| CN113706483A (en) * | 2021-08-16 | 2021-11-26 | 佛山职业技术学院 | Detection method and detection system for pork quality |
| US11497221B2 (en) | 2019-07-19 | 2022-11-15 | Walmart Apollo, Llc | Systems and methods for managing meat cut quality |
-
2009
- 2009-05-07 CN CN2009201194969U patent/CN201508328U/en not_active Expired - Lifetime
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855640A (en) * | 2012-08-10 | 2013-01-02 | 上海电机学院 | Fruit grading system based on neural network |
| WO2017133178A1 (en) * | 2016-02-02 | 2017-08-10 | 意力(广州)电子科技有限公司 | Automatic optical inspection program-based touch panel automatic inspection apparatus |
| CN109459391A (en) * | 2019-01-07 | 2019-03-12 | 塔里木大学 | A kind of jujube Quality Detection, jujube Polarization Detection model generating method and device |
| CN110189148A (en) * | 2019-05-29 | 2019-08-30 | 广州影子科技有限公司 | Retroactive method, retrospective device and traceability system |
| US11497221B2 (en) | 2019-07-19 | 2022-11-15 | Walmart Apollo, Llc | Systems and methods for managing meat cut quality |
| US11864562B2 (en) | 2019-07-19 | 2024-01-09 | Walmart Apollo, Llc | Systems and methods for managing meat cut quality |
| CN112683899A (en) * | 2020-04-29 | 2021-04-20 | 海南远生渔业有限公司 | Aquatic product quality detection method |
| CN113706483A (en) * | 2021-08-16 | 2021-11-26 | 佛山职业技术学院 | Detection method and detection system for pork quality |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101561402B (en) | Machine vision-based real-time detection and grading method and grading device for pork appearance quality | |
| CN201508328U (en) | Machine vision-based real-time detecting and grading device for pork appearance quality | |
| CN102147402A (en) | Machine vision technology based method for rapidly detecting egg freshness | |
| KR100948406B1 (en) | Chicken carcass quality grade automatic decision and weight measuring system | |
| CN110321815A (en) | A kind of crack on road recognition methods based on deep learning | |
| CN102854148A (en) | Detection and grading system for tenderness of fresh beef based on multispectral imagery | |
| CN105651776A (en) | Device and method for automatically grading beef carcass meat yield based on computer vision | |
| CN103983190B (en) | Visual inspection classification device and method for appearance dimensions of group origin duck eggs | |
| CN105354831B (en) | A kind of more defect inspection methods based on image block variance-weighted eigenvalue | |
| CN108318494B (en) | The red online vision detection and classification devices and methods therefor for proposing fruit powder | |
| CN204287031U (en) | A kind of online the cannot-harm-detection device of fish freshness based on high light spectrum image-forming technology | |
| CN102156129A (en) | Beef quality intelligent grading system and method based on machine vision | |
| CN108855988A (en) | Walnut kernel stage division and walnut kernel grading plant based on machine vision | |
| CN108663367A (en) | A kind of egg quality lossless detection method based on egg unit weight | |
| CN107564000A (en) | Hericium erinaceus Non-Destructive Testing stage division based on computer vision | |
| CN111768401A (en) | Rapid grading method for freshness of iced pomfret based on deep learning | |
| CN104616204B (en) | A kind of polynary, fine, intelligent trunk meat stage division on automatic pig slaughtering line | |
| CN110675400A (en) | Rapid and intelligent mutton quality index detection method based on mobile phone APP | |
| CN103149163A (en) | Multispectral image textural feature-based beef tenderness detection device and method thereof | |
| CN100376888C (en) | Method and device for computer vision detection and classification of beef carcase quality | |
| Zhao et al. | Distribution of beef tenderness grading based on texture features by hyperspectral image analysis | |
| CN108460380A (en) | A kind of bamboo cane method for sorting colors and system based on domain color | |
| Thong et al. | Mango classification system uses image processing technology and artificial intelligence | |
| CN114898405A (en) | Portable broiler chicken abnormity monitoring system based on edge calculation | |
| CN104573662A (en) | Cloud judgment method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| AV01 | Patent right actively abandoned |
Granted publication date: 20100616 Effective date of abandoning: 20090507 |