CN103076288B - A kind of flesh of fish self-grading device based on computer vision and method - Google Patents
A kind of flesh of fish self-grading device based on computer vision and method Download PDFInfo
- Publication number
- CN103076288B CN103076288B CN201210573966.5A CN201210573966A CN103076288B CN 103076288 B CN103076288 B CN 103076288B CN 201210573966 A CN201210573966 A CN 201210573966A CN 103076288 B CN103076288 B CN 103076288B
- Authority
- CN
- China
- Prior art keywords
- image
- flesh
- fish
- equation
- yellowish pink
- 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.)
- Active
Links
Abstract
The present invention relates to a kind of flesh of fish self-grading device based on computer vision, comprising: vision collecting device, illumination system, external trigger module and computer system; Vision collecting device is connected with illumination system, computer system, is also connected with computer system by external trigger module; Method comprises: computer system receives after instruction, sends to control signal to external trigger module and make it trigger vision collecting device collection flesh of fish image information; After the flesh of fish image information of image pick-up card collection is carried out Image semantic classification by computer system, color feature extracted is carried out in the sliced meat region of extracting in flesh of fish image, then the color feature value obtained is substituted into the yellowish pink equation of the standard set up, the equation value obtained is flesh of fish grade.Structure of the present invention is simple, and classification is accurate, simple operation, detects rapidly, can be applied in production easily.
Description
Technical field
The present invention relates to a kind of apparatus and method of technical field of computer vision, specifically, be the yellowish pink self-grading device of a kind of atlantic salmon based on computer vision and automatic grading method thereof, the assessment of production central Atlantic salmon chromaticity matter can be applied to.
Background technology
Atlantic salmon (salmon salar) belongs to salmon shape order, salmon section, and salmon belongs to, and being commonly called as " salmon ", is that a kind of hereditary variability is more stable, the worldwide cultured fishes be of high nutritive value.The yellowish pink evaluation method of traditional atlantic salmon mainly relies on manual method: through the reviewer of professional training, by Roche standard color comparison card, observe and obtain the color grade of the Roche standard color comparison card the most close with sliced meat color, and it can be used as the color grade of Atlantic salmon fish meat sheet.There are some defects in the method, first, the method substantially increases production cost, and human resources occupy sizable ratio in production cost; The second, the hierarchical speed of classic method is slowly, seriously consuming time, is unfavorable for evaluating Atlantic salmon fish meat sheet color rapidly in production or in selling, so that the carrying out of batch is classified, or evaluates its yellowish pink quality in print; 3rd, classic method also also exists stability scarcely, and reviewer is due to visual fatigue, and memory for color lacks, and a variety of causes such as colour recognition ability difference, and rating result may be caused to occur error, lacks reliability.
Computer vision is one and develops new branch of science rapidly.Realize measuring by the vision mechanism of simulating people, location, the functions such as monitoring have been an important topic in current intelligent machine field.Over nearly 20 years, the research of computer vision has moved towards practical application from laboratory, from simple binary Images Processing to the image procossing of many gray scales, from general two-dimensional signal process to the vision mechanism of three-dimensional, the research of model and algorithm, makes great progress.And the raising at full speed of computer industry level and the development of the subject such as artificial intelligence and neuroid, more facilitate the practical of computer vision system.At present.Computer vision system is just widely used in the every field such as vision-based detection.
Through finding the literature search of prior art, China Patent No. is: 201210083596, patent name is: a kind of colour difference meter assessment method of pork color, its readme is: lie on red plastic plate by new for the cube meat cut cut side up, and with individual layer water white transparency preservative film parcel cube meat, make film be close to cube meat, avoid bubble as far as possible, fold, is then placed in 0 ~ 4 DEG C of color development 30 ~ 60min; Utilize colour difference meter to carry out yellowish pink to measure.The method of this patent still belongs to the category of artificial grading, is easily similar to mistake and instability that traditional yellowish pink ranking method occurs.Patent name is: (China Patent No. is: 201110376478) for a kind of fresh beef appetizer freshness quick nondestructive evaluation method and system, although this patent has got multiple index affecting meat freshness, but comparatively simple to the evaluation method of its yellowish pink, accuracy has much room for improvement.
Summary of the invention
The object of the invention is to overcome tradition and the deficiencies in the prior art, provide a kind of yellowish pink self-grading device based on computer vision, the technical scheme that the present invention is adopted for achieving the above object is:
Based on a flesh of fish self-grading device for computer vision, comprising: vision collecting device, illumination system, external trigger module and computer system; Vision collecting device is connected with illumination system, computer system, is also connected with computer system by external trigger module;
Illumination system: for providing photoenvironment and preventing the sliced meat surface reflection put into;
Vision collecting device: for gathering the image information being placed in the illumination system inside flesh of fish;
External trigger module: the control signal of receiving computer system also triggers vision collecting device and makes it gather image information;
Computer system: receive the instruction of man-machine interface and the collection of control flesh of fish image information, the image information collected is carried out the grade that image procossing obtains oppressing; The process of collection is presented in man-machine interface in real time simultaneously.
Described illumination system comprises casing, fluorescent tube and white reflection plate; Casing one side is the opening for putting into the flesh of fish, and inside surface is white reflection plate, and inside front is provided with fluorescent tube, is the glass gathering image for vision collecting device in the middle part of end face.
Described vision collecting device comprises ccd image sensor and image pick-up card; The optical axis of ccd image sensor is perpendicular to the bottom surface of illumination system, and ccd image sensor and external trigger model calling, be also connected with computer system by image pick-up card.
Based on a flesh of fish automatic grading method for computer vision, comprise the following steps:
After the man-machine interface of computer system receives instruction, send and control signal to external trigger module and make it trigger vision collecting device collection flesh of fish image information;
After the flesh of fish image information of image pick-up card collection is carried out Image semantic classification by computer system, color feature extracted is carried out in the sliced meat region of extracting in flesh of fish image, then the color feature value obtained is substituted into the yellowish pink equation of the standard set up, the equation value obtained is flesh of fish grade, and said process and flesh of fish grade is presented in real time in man-machine interface.
The instruction that the man-machine interface of described computer system receives comprises: illumination sets, and Image Acquisition, obtains yellowish pink grade.
Described Image semantic classification is specially and the flesh of fish image information of image pick-up card collection is converted to RGB image, and the median filter of recycling 3 × 3 carries out mask to whole image, does noise reduction process.
Described color feature extracted to be specially the image in qualified region by RGB color space conversion to CIE Lab color space, and the mean value extracted in RGB image under all pixel G passages, extract the mean value under all pixel L passages, a passage in CIE Lab image.
The yellowish pink equation of described Criterion comprises the following steps:
Gather 10 random areas on the coloured image of standard color comparison card 20 to 34 grades, in RGB color space, obtain each mean value of the R eigenwert of all pixels in each region of image, G eigenwert, B eigenwert;
Again by RGB color space conversion to CIE Lab color space and HSV color space, obtain the mean value of each eigenwert of L, a, b, H, S, V of all pixel regionals respectively, obtain above-mentioned 9 single channel color feature value; Then these 9 eigenwerts and respective standard colorimetric card grade are carried out multiple linear regression analysis and obtain a ternary linear equation;
Best three features of the linear relationship of the grade that selected characteristic value and standard color comparison card are demarcated as the parameter of the yellowish pink equation of standard, the yellowish pink equation of the standard that namely obtains.
The yellowish pink equation of described standard is
MCL＝-0.038G-0.031×L+0.18×a+22.365
MCL is yellowish pink grade, and G is green channel, and L is lightness passage, and a is red/green passage.
The present invention has following beneficial effect and advantage:
1. structure of the present invention is simple, and classification is accurate, simple operation, detects rapidly, can be applied in production and selling easily.
2. device of the present invention is by the yellowish pink grade of matching atlantic salmon and the funtcional relationship of the corresponding single channel color characteristic of its coloured image, selects three single channel color component G that linear relationship is best, L, a.Utilize multiple linear regression analysis method to simulate the yellowish pink equation of standard of the yellowish pink grade of atlantic salmon and G, L, a tri-color components, to achieve atlantic salmon yellowish pink automatically, classification fast, accurately.
3. result of the present invention is more stable, and reliably, cost is cheaper.
Accompanying drawing explanation
Fig. 1 is structure drawing of device of the present invention;
Fig. 2 is the connection diagram between image processing module and other each modules;
Fig. 3 is the connection diagram of control module and other modules;
Fig. 4 is structured flowchart of the present invention;
Fig. 5 is the process flow diagram of computer working of the present invention.
Embodiment
Below in conjunction with drawings and Examples, the present invention is described in further detail.
As shown in Figure 1, the present invention includes Four processes part: vision collecting device 1, illumination system 6, external trigger module 4 and computer system 5.Obtained image information for gathering the Atlantic salmon fish meat sheet being placed in illumination system 6, and is sent to computer system 5 by vision collecting device 1.
Computer system 5 is provided with control module, display module 21, image processing module 9 and yellowish pink diversity module 18.As shown in Figure 2, display module 21 is delivered on image information one road that vision collecting device 1 delivers to computer system 5, and image processing module 9 is delivered on another road; Display module 21 shows the image and colouring information that obtain in real time, and provides human-computer interaction interface; Image processing module 9 is by carrying out pre-service to the image obtained, region of interest regional partition, color space conversion, and single channel color feature extracted obtains required eigenwert; Yellowish pink diversity module 18 obtains the yellowish pink grade of atlantic salmon, as shown in Figure 4.
The control module of computer system 5 is connected with vision collecting device 1 by external trigger module 4, and the outside trigger module 4 of control module sends instruction, and the instruction that external trigger module 4 sends according to control module sends control signal to vision collecting device 1.External trigger module 4, sends control signal according to the instruction that control module sends to ccd image sensor 2.
Vision collecting device 1 comprises ccd image sensor 2 and image pick-up card 3.During installation, the optical axis of ccd image sensor is vertical with the upper surface of the sliced meat be placed in illumination system 6.The sliced meat image that ccd image sensor 2 collection is placed on platform, and by image pick-up card 3 by obtained image transfer to computer system 5.
Illumination system 6, is provided with LED lamp tube 7 above system side, system inside surface is white reflection plate 8, makes each even angle light of tested sliced meat, eliminates reflective and shade.Be glass in the middle part of end face, gather image for vision collecting device 1.
Control module, as shown in Figure 3, the setting being input as controling parameters in human-computer interaction interface of control module, sends instruction to external trigger module 4 by computer system I/O interface.Wherein, controling parameters comprises illumination setting, and Atlantic salmon fish meat sheet Image Acquisition, obtains yellowish pink grade.
Display module 21, obtains image output information from the output of image procossing modules, or parameter information, and shows on screen, simultaneously also will as human-computer interaction interface, the operation of response user.
As shown in Figure 5, in computer system, image processing module 9 comprises image collection module 10, image pre-processing module 11, region of interest regional partition module 12, these four functional modules of color feature extracted module 15, modules obtains image from previous module respectively, and image is processed, the image after process is delivered to next module; Yellowish pink diversity module 18 comprises the yellowish pink equation 19 of standard and automatic classification submodule 20, is passed the eigenwert of coming, complete the automatic classification of atlantic salmon yellowish pink by image processing module 9; Each step information shows at display module 21 in real time.
Image processing module 9 comprises image collection module 10, image pre-processing module 11, region of interest regional partition module 12, these four modules of color feature extracted module 15.Modules obtains image from previous module respectively, by processing image, the image after process is delivered to next module.Be specially: image collection module obtains original image information from image pick-up card, carry out conversion process, for follow-up Image semantic classification submodule provides required image information; Image semantic classification submodule utilizes 3 × 3 median filters to carry out noise reduction process to obtained image; Region of interest regional partition module obtains illumination and the uniform qualified sliced meat region of color by the segmentation of OTSU method; Color feature extracted module, by former RGB image is converted to CIE Lab space, obtains the color value of these 6 passages of R, G, B, L, a, b, and the color feature value of these 3 passages of G, L, a is passed to yellowish pink diversity module.
Image collection module 10 obtains original image from image pick-up card 3, and original image is RGB image, then for image pre-processing module 10 provides required image information.The picture format exported for different images capture card is different, the image obtained from image pick-up card 3 needs to carry out format conversion by image collection module, be the manageable picture format of image pre-processing module 11 (the mainly difference of two kinds of cameras, USB and 1394) image format conversion.
Image pre-processing module 11.This module carries out noise reduction process to the original image information that image collection module 10 obtains, to camera, Fuzzy Processing is carried out in the distortion of the pixel color value that illumination factor etc. causes, and the pixel making color distortion comparatively serious is consistent with the color-values of surrounding pixel point by medium filtering.
Region of interest regional partition module 12 comprises sliced meat region segmentation submodule 13 and qualified region segmentation submodule 14.Sliced meat segmentation submodule 13 gets image data information from image pre-processing module 11, under L passage, utilize Sobel operator extraction to the sliced meat marginal information be positioned on illumination system 6 measuring table, and utilize marginal information (i.e. coordinate figure) to extract sliced meat region; Qualified region segmentation submodule 14 shines and the uniform region of color for surface light in the panel region that cuts meat, be specially the L eigenwert of each pixel extracting sliced meat region, the histogram now utilizing these eigenwerts to draw can present bimodal distribution, utilize OTSU method, self-adaptation obtains suitable threshold value, and by binaryzation by qualified sliced meat extracted region out.
Color feature extracted module 15 comprises color space conversion submodule 16 and single channel color feature extracted submodule 17.Color space conversion submodule 16 by determined for qualified region segmentation submodule 14 sliced meat area image from RGB color space conversion to CIE Lab color space.Single channel color feature extracted submodule 17, extracts the G passage mean value of all pixels in RGB color space on Atlantic salmon fish meat sheet image, in CIE Lab color space, and L passage mean value and a passage mean value.
Yellowish pink diversity module 18 comprises the yellowish pink equation 19 of standard and automatic classification submodule 20, and the foundation of yellowish pink diversity module is the color characteristic of sliced meat and the linear relationship of yellowish pink grade.Utilize Roche standard color comparison card as sample collection, the yellowish pink equation of standard is by the color feature value of the Roche standard color comparison card of 20 to 34 grades and color grade matching corresponding to it; The process of establishing of the yellowish pink equation 19 of standard is sampled by 10 random areas of each level images to Roche standard color comparison card, Image semantic classification, Iamge Segmentation and color feature extracted, obtain R, G, B in each region, H, S, V, L, the mean eigenvalue of all pixels under 9 single channels such as a, b, add by the R value of all pixels and after again divided by the average strain value that namely number of pixel is all pixel R values, the mean value of all the other 8 eigenwerts above-mentioned such as G, B also so calculates; And obtain three the single channel color characteristics the highest with Roche standard color comparison card color grade correlativity through correlation analysis, i.e. G, L, a, the yellowish pink equation 19 of the standard that then simulates.Automatic classification submodule 20 is mainly by the G of the atlantic salmon yellowish pink of acquisition, and L, a tri-eigenwerts are updated to the yellowish pink equation 19 of standard, calculate the color grade belonging to sliced meat.About the foundation of the yellowish pink equation of standard, can see the dotted arrow in Fig. 5, namely the foundation of the yellowish pink equation of standard is also according to the flow process shown in Fig. 5, but time sequencing is different.Can the yellowish pink equation of Criterion at first, and then utilize this equation to carry out classification to atlantic salmon yellowish pink in yellowish pink classification afterwards.
Wherein correlation analysis and multiple linear regression analysis, refer to multiple eigenwerts of each single channel color characteristic in the region collected from 20-34 grade standard colorimetric card, the grade that the standard color comparison card corresponding with it marks carries out a unitary linear fit, now can obtain the goodness of fit R of each feature and corresponding standard color comparison card grade
2, then select best (the i.e. R of the goodness of fit
2be worth maximum) three features as the parameter of the yellowish pink equation of standard, simulate a ternary linear equation.Fit procedure is: the sampled data for each eigenwert of G, L, a sets up the vector (number of n representative sampling) of a 1 × n; Then input model Y=AX
1+ BX
2+ CX
3+ D; Each vector substitutes into model, solves A, B, C, D and obtains equation (as long as n>4, generally just can simulate equation).
Claims (4)
1., based on a flesh of fish automatic grading method for computer vision, it is characterized in that comprising the following steps:
After the man-machine interface of computer system (5) receives instruction, send and control signal to external trigger module (4) and make it trigger vision collecting device (1) to gather flesh of fish image information;
After the flesh of fish image information that image pick-up card (3) gathers is carried out Image semantic classification by computer system (5), color feature extracted is carried out in the sliced meat region of extracting in flesh of fish image, then the color feature value obtained is substituted into the yellowish pink equation of the standard set up, the equation value obtained is flesh of fish grade, and after the flesh of fish image information that image pick-up card (3) gathers is carried out Image semantic classification, extract the flesh of fish image in sliced meat region carry out color feature extracted process and the flesh of fish grade be presented in real time in man-machine interface;
The foundation of the yellowish pink equation of described standard comprises the following steps:
Gather 10 random areas on the coloured image of standard color comparison card 20 to 34 grades, in RGB color space, obtain each mean value of the R eigenwert of all pixels in each region of image, G eigenwert, B eigenwert;
Again by RGB color space conversion to CIE Lab color space and HSV color space, obtain the mean value of each eigenwert of L, a, b, H, S, V of all pixel regionals respectively, obtain R, G, B, L, a, b, H, S, V 9 single channel color feature value; Then these 9 eigenwerts and respective standard colorimetric card grade are carried out multiple linear regression analysis and obtain a ternary linear equation;
Best three features of the linear relationship of the grade that selected characteristic value and standard color comparison card are demarcated as the parameter of the yellowish pink equation of standard, the yellowish pink equation of the standard that namely obtains;
The yellowish pink equation of described standard is
MCL＝-0.038G-0.031×L+0.18×a+22.365
MCL is yellowish pink grade, and G is green channel, and L is lightness passage, and a is red/green passage.
2. a kind of flesh of fish automatic grading method based on computer vision according to claim 1, is characterized in that: the instruction that the man-machine interface of described computer system (5) receives comprises: illumination sets, and Image Acquisition, obtains yellowish pink grade.
3. a kind of flesh of fish automatic grading method based on computer vision according to claim 1, it is characterized in that: the flesh of fish image information that described Image semantic classification is specially image pick-up card (3) gathers is converted to RGB image, the median filter of recycling 3 × 3 carries out mask to whole image, does noise reduction process.
4. a kind of flesh of fish automatic grading method based on computer vision according to claim 1, it is characterized in that: described color feature extracted to be specially the image in qualified region by RGB color space conversion to CIE Lab color space, and the mean value extracted in RGB image under all pixel G passages, extract the mean value under all pixel L passages, a passage in CIE Lab image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210573966.5A CN103076288B (en) | 2012-12-26 | 2012-12-26 | A kind of flesh of fish self-grading device based on computer vision and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210573966.5A CN103076288B (en) | 2012-12-26 | 2012-12-26 | A kind of flesh of fish self-grading device based on computer vision and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103076288A CN103076288A (en) | 2013-05-01 |
| CN103076288B true CN103076288B (en) | 2015-08-26 |
Family
ID=48152892
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210573966.5A Active CN103076288B (en) | 2012-12-26 | 2012-12-26 | A kind of flesh of fish self-grading device based on computer vision and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103076288B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109975292A (en) * | 2019-03-18 | 2019-07-05 | 江苏大学 | A kind of atlantic salmon and rainbow trout method for quick identification based on machine vision |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103487383A (en) * | 2013-09-27 | 2014-01-01 | 浙江工商大学 | Method for measuring meat color luster of tuna by using colorimeter |
| CN103940751B (en) * | 2014-04-23 | 2016-11-09 | 成都大学 | A kind of assessment method of cold fresh meat quality based on image processing techniques |
| CN107392117B (en) * | 2015-03-10 | 2019-10-08 | 利辛县宝贵食品有限公司 | Fish body freshness identifying system |
| CN104865198A (en) * | 2015-04-28 | 2015-08-26 | 青岛农业大学 | Machine vision-based fresh fat liver quality automatic grading system and method |
| CN105606473B (en) * | 2016-01-20 | 2018-06-15 | 华中农业大学 | A kind of red grape rigidity nondestructive testing method based on machine vision |
| CN108416782B (en) * | 2018-04-19 | 2023-09-26 | 云南佳叶现代农业发展有限公司 | Tobacco leaf grading method and system based on visual identification and illumination correction |
| CN108828167A (en) * | 2018-07-05 | 2018-11-16 | 中国水产科学研究院黄海水产研究所 | A kind of rapid method for assessment for eating northern bluefin tuna meat grade raw |
| CN110006899A (en) * | 2019-04-12 | 2019-07-12 | 华中农业大学 | The lossless vision detection and classification method of lime-preserved egg inside quality |
| CN110827267B (en) * | 2019-11-08 | 2023-04-07 | 武汉海晨工控技术有限公司 | Vinasse blending quality visual analysis method and system in brewing process of raw wine in white spirit factory |
| CN112101957A (en) * | 2020-01-01 | 2020-12-18 | 袁艳荣 | Information recording platform based on big data storage |
| CN111229649B (en) * | 2020-01-20 | 2023-09-01 | 湖北民族大学 | Article grading method and automatic grading system |
| CN111795655B (en) * | 2020-08-04 | 2021-10-08 | 苏交科集团股份有限公司 | Contour mark installation angle measuring device and measuring method thereof |
| CN112634274B (en) * | 2021-03-10 | 2021-06-04 | 广东省农业科学院动物科学研究所 | Machine vision-based fattening pork quality evaluation method and system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101158651A (en) * | 2007-11-16 | 2008-04-09 | 南京农业大学 | Device and method for detecting apples inside and outside quality based on laser image analyse |
| CN101551341B (en) * | 2009-04-09 | 2011-02-16 | 浙江大学 | Meat online non-destructive testing method and apparatus based on integration of image and spectrum information |
| CN101561402B (en) * | 2009-05-07 | 2011-06-22 | 浙江大学 | Machine vision-based real-time detection and grading method and grading device for pork appearance quality |
| CN101706445B (en) * | 2009-11-10 | 2011-05-18 | 吉林大学 | Image processing mthod for beef marbling grain grade scoring |
-
2012
- 2012-12-26 CN CN201210573966.5A patent/CN103076288B/en active Active
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109975292A (en) * | 2019-03-18 | 2019-07-05 | 江苏大学 | A kind of atlantic salmon and rainbow trout method for quick identification based on machine vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103076288A (en) | 2013-05-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103076288B (en) | A kind of flesh of fish self-grading device based on computer vision and method | |
| CN105181912B (en) | A kind of Noninvasive Measuring Method of Freshness in rice storage | |
| CN102147402A (en) | Machine vision technology based method for rapidly detecting egg freshness | |
| CN105651713B (en) | A kind of green vegetables leaf chlorophyll quantitative detecting method based on computer image analysis | |
| CN102156129B (en) | Beef quality intelligent grading system and method based on machine vision | |
| CN108181316B (en) | Bamboo strip defect detection method based on machine vision | |
| CN103149214B (en) | Method for detecting flaw on surface of fruit | |
| CN110070526A (en) | Defect inspection method based on the prediction of deep neural network temperature figure | |
| US8260005B2 (en) | Portable tool for determining meat quality | |
| CN110705655A (en) | Tobacco leaf classification method based on coupling of spectrum and machine vision | |
| CN108318494A (en) | The red online vision detection and classification devices and methods therefor for putting forward fruit powder | |
| CN109089992A (en) | A kind of classification method and system of the Estimation of The Fish Freshness based on machine vision | |
| CN110009609B (en) | Method for rapidly detecting yellow rice | |
| CN109975292A (en) | A kind of atlantic salmon and rainbow trout method for quick identification based on machine vision | |
| CN114441452B (en) | Optical fiber pigtail detection method | |
| CN100376888C (en) | Method and device for computer vision detection and classification of beef carcase quality | |
| Motonaga et al. | A method of making digital fruit color charts for cultivation management and quality control | |
| CN106960188B (en) | Weather image classification method and device | |
| CN206331487U (en) | A kind of agricultural product volume rapid measurement device based on machine vision | |
| CN111316934B (en) | Bee birth and fertility analyzer and method | |
| CN107239761A (en) | Fruit tree branch pulling effect evaluation method based on skeleton Corner Detection | |
| CN112686105B (en) | Fog concentration grade identification method based on video image multi-feature fusion | |
| CN112529922A (en) | Method for accurately separating RGB image backgrounds of multicolor blades in open environment | |
| CN110929787A (en) | Apple objective grading system based on images | |
| CN114820616B (en) | Equipment state detection method and device for flashing mode indicator light |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |