CN106305567A - Image identification method used for judging river crabs - Google Patents
Image identification method used for judging river crabs Download PDFInfo
- Publication number
- CN106305567A CN106305567A CN201610691336.6A CN201610691336A CN106305567A CN 106305567 A CN106305567 A CN 106305567A CN 201610691336 A CN201610691336 A CN 201610691336A CN 106305567 A CN106305567 A CN 106305567A
- Authority
- CN
- China
- Prior art keywords
- crab
- point
- image
- carapace
- eigenvalue
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses an image identification method used for judging river crabs. The image identification method comprises the steps of acquiring a river crab image, pre-processing the image, constructing a river crab mode pattern and calculating a characteristic value, constructing a group mode pattern and calculating a group characteristic value, judging river crab individuals and the like; the river crab mode pattern is constructed and the characteristic value is calculated through analyzing river crab image data, and the mode patterns and the characteristic values of the individuals and groups which are to be judged are compared, so that the groups of river crabs to be judged are judged. The image identification method provided by the invention is applied to family breeding of the river crabs and the difficulties that the river crabs are difficult to mark in a river crab growth process and mixed breeding families are not easy to distinguish can be avoided; a good family meeting a breeding target is selected through judging the belonged families of the good river crabs obtained by mixed breeding; mode pattern and characteristic value archives of different river crab production places can be constructed in advance and are used for detecting the production places of the river crabs; the image identification method can also be applied to the judgment of other crab types.
Description
Technical field
The present invention relates to field of image recognition, particularly relate to a kind of image-recognizing method differentiated for crab.
Background technology
Crab formal name used at school Eriocheir sinensis (Eriocheir sinsensis H.), be under the jurisdiction of Arthropoda, Crustachia,
Decapoda, Grapsidae, bending of leg Eriocheir sinensis subfamily, Eriocheir.Crab meat flavour is delicious, nutritious, is loved by the people, from 1991
Since Nian China's culture of Chinese mitten crab yield quickly, steady growth, become the Aquatic product economic living that China is particularly important, culture of Chinese mitten crab
It it is one of most active pillar industry with development prospect in China's aquaculture.Many areas, the whole nation using crab industry as tune
Whole agricultural structure, increase farmers' income, developing rural economy, the pillar industry built a Harmonious Society.Along with culture of Chinese mitten crab
The development of industry, breeding enterprise and raiser are the most vigorous to the demand of the excellent juvenile crab of selection-breeding, but some technology rings
Joint but governs the development of crab breeding industry, especially restricts the development of family selective breeding pattern.
Labelling technique is exactly the key point restricting family selective breeding mode development, the existing labelling side being used in shell-fish
Method has, cut appendage, tatoo, deep tissues vital staining, line mark, fluorochrome mark, fluorescent board labelling, eye mark etc., but by
Shell in Eriocheir sinensis class growth course habit frequently, and the application effect of these methods is all difficult to ensure that and accurately distinguishes family, at present
Widespread practice is that each family needs to separate cultivation, the most again it cannot be guaranteed that the concordance of its breeding environment, and then impact choosing
Educate effect.Some experts and scholars develop molecular marking technique, but are limited to the factor such as technical costs and operating time, it is also difficult to
Large-scale application.
Summary of the invention
In order to overcome above-mentioned the deficiencies in the prior art, the invention provides a kind of image recognition side differentiated for crab
Method.Crab is not directly marked by the present invention, by contrast crab defect individual and colony's view data feature, it is achieved select
The purpose of superior families.
The technical solution adopted in the present invention is the family relation being differentiated crab by the method for image comparison, thus real
The purpose of existing labelling crab, specifically includes.
(1) gather crab image, refer to obtain the carapace photo of crab.Acquisition mode can include but not limited to high definition
Camera is taken pictures, photographic head is taken pictures, scanner scanning.
(2) Image semantic classification, refers to that the image to every crab carries out calculation system.First carapace image is carried out
Dimension-reduction treatment, retains carapace edge contour, the central point of protrusion of surface and the central point in special color region, and remainder is saturating
Brightization；Secondly carapace contour line reveals process, then selected characteristic point in units of pixel: to incise in the middle of carapace volume tooth
Depression deepest point is that initial point sets up plane coordinate system, draws the coordinate of each edge pixel point, being positioned at carapace edge wheel
Turning point on profile is set to A point, starts to be denoted as clockwise A from initial pointi(i=1,2,3 ... n), pixel between each two A point
The 25%th of quantity the, 50%, 75% point of point is set to B point, starts to be denoted as clockwise B from initial pointi(i=1,2,3 ... n), cephalothorax
In first, the location of the core of each projection is C point, starts to be denoted as clockwise C from initial pointi(i=1,2,3 ... n), on carapace
The location of the core in special color region is D point, starts to be denoted as clockwise D from initial pointi(i=1,2,3 ... n).
(3) build crab ideograph and eigenvalue, start to link A point successively from initial point and B point forms closed curve, become
Crab ideograph；Measure Ai、Bi、Ci、DiEach point is denoted as LA respectively to the distance of initial pointi、LBi、LCi、LDi, i=1、2、3……
N, measures the first side facewidth and is denoted as L0, calculate eigenvalue Tj, Tj=Lj/L0,j=Ai、Bi、Ci、Di。
(4) building known crab Population pattern figure and eigenvalue, repetition step (1), (2) are every in obtaining known crab colony
The view data of individuality, statistical analysis obtains LAi、LBi, meansigma methods `LAi、`LBi, and then obtain A point and B point average bit
Put `A point and `B point, start to link `A point successively from initial point and `B point forms closed curve, become the pattern of known crab colony
Figure；Statistical analysis obtains eigenvalue TjMeansigma methods `Tj, become the eigenvalue of known crab colony.
(5) crab individuality differentiates, first crab individual mode figure to be discriminated and known crab Population pattern figure is compared
Right, the two initial point is overlapped, scales crab individual mode figure to be discriminated and move rotation, the part coincided together is thought
The two coincide, if the two goodness of fit reaches more than 75%, then carries out eigenvalue comparison, if 0.95 < Tj/`Tj< think for 1.05
Second it coincide, if the eigenvalue goodness of fit reaches more than 90%, then it is assumed that crab individuality to be discriminated is the member in colony.
Detailed description of the invention
Below in conjunction with embodiment, the concrete application of the present invention is elaborated.
Embodiment 1: selection-breeding big specification crab.
First the crab family of selection-breeding to be carried out is carried out family individually to cultivate and mixed breed simultaneously, and preset big specification
Standard.
Secondly applying the present invention to select after becoming Eriocheir sinensis or button Eriocheir sinensis harvesting, concrete steps include.
(1) gather crab image, refer to obtain the carapace photo of crab；Acquisition mode can include but not limited to high definition
Camera is taken pictures, photographic head is taken pictures, scanner scanning.
(2) Image semantic classification, refers to that the image to every crab carries out calculation system；First carapace image is carried out
Dimension-reduction treatment, retains carapace edge contour, the central point of protrusion of surface and the central point in special color region, and remainder is saturating
Brightization；Secondly carapace contour line reveals process, then selected characteristic point in units of pixel: to incise in the middle of carapace volume tooth
Depression deepest point is that initial point sets up plane coordinate system, draws the coordinate of each edge pixel point, being positioned at carapace edge wheel
Turning point on profile is set to A point, starts to be denoted as clockwise A from initial pointi(i=1,2,3 ... n), pixel between each two A point
The 25%th of quantity the, 50%, 75% point of point is set to B point, starts to be denoted as clockwise B from initial pointi(i=1,2,3 ... n), cephalothorax
In first, the location of the core of each projection is C point, starts to be denoted as clockwise C from initial pointi(i=1,2,3 ... n), on carapace
The location of the core in special color region is D point, starts to be denoted as clockwise D from initial pointi(i=1,2,3 ... n).
(3) build crab ideograph and eigenvalue, start to link A point successively from initial point and B point forms closed curve, become
Crab ideograph；Measure Ai、Bi、Ci、DiEach point is denoted as LA respectively to the distance of initial pointi、LBi、LCi、LDi, i=1、2、3……
N, measures the first side facewidth and is denoted as L0, calculate eigenvalue Tj, Tj=Lj/L0,j=Ai、Bi、Ci、Di。
(4) build family ideograph and family eigenvalue, obtain every river of each family according to step (1), (2), (3)
The view data of Eriocheir sinensis, obtains LA according to family statistical analysisi、LBi, meansigma methods `LAi、`LBi, and then obtain A point and B point and put down
All position `A point and `B points, start to link `A point successively from initial point and `B point form closed curve, become the ideograph of family；System
Meter analysis obtains eigenvalue TjMeansigma methods `Tj, become the eigenvalue of family.
(5) crab individuality differentiates, first selects the crab conduct meeting default big specification standards from mixed breed crab
Crab to be discriminated, obtains view data according to step (1), (2), (3), secondly by crab individual mode figure to be discriminated successively and family
It is that ideograph is compared, the two initial point is overlapped, scale crab individual mode figure to be discriminated and move rotation, overlap
Part together thinks that the two coincide, if the two goodness of fit reaches more than 75%, then carries out eigenvalue comparison, if 0.95 <
Tj/`Tj< think the most identical for 1.05, if the eigenvalue goodness of fit reaches more than 90%, then it is assumed that crab individuality to be discriminated is family
Set member, finally selects to differentiate that the family that success rate is high enters follow-up selection-breeding step.
Embodiment 2: differentiate the crab place of production.
(1) gathering crab image, refer to obtain the carapace photo of crab, acquisition mode can include but not limited to high definition
Camera is taken pictures, photographic head is taken pictures, scanner scanning.
(2) Image semantic classification, refers to that the image to every crab carries out calculation system.First carapace image is carried out
Dimension-reduction treatment, retains carapace edge contour, the central point of protrusion of surface and the central point in special color region, and remainder is saturating
Brightization；Secondly carapace contour line reveals process, then selected characteristic point in units of pixel: to incise in the middle of carapace volume tooth
Depression deepest point is that initial point sets up plane coordinate system, draws the coordinate of each edge pixel point, being positioned at carapace edge wheel
Turning point on profile is set to A point, starts to be denoted as clockwise A from initial pointi(i=1,2,3 ... n), pixel between each two A point
The 25%th of quantity the, 50%, 75% point of point is set to B point, starts to be denoted as clockwise B from initial pointi(i=1,2,3 ... n), cephalothorax
In first, the location of the core of each projection is C point, starts to be denoted as clockwise C from initial pointi(i=1,2,3 ... n), on carapace
The location of the core in special color region is D point, starts to be denoted as clockwise D from initial pointi(i=1,2,3 ... n).
(3) build crab ideograph and eigenvalue, start to link A point successively from initial point and B point forms closed curve, become
Crab ideograph；Measure Ai、Bi、Ci、DiEach point is denoted as LA respectively to the distance of initial pointi、LBi、LCi、LDi, i=1、2、3……
N, measures the first side facewidth and is denoted as L0, calculate eigenvalue Tj, Tj=Lj/L0,j=Ai、Bi、Ci、Di。
(4) build different sources crab ideograph and eigenvalue, go each 100 of the crab of different sources, according to step
(1), (2), (3) obtain the view data of every crab, obtain LA according to place of production statistical analysisi、LBi, meansigma methods `LAi、`
LBi, and then obtain A point and B point mean place `A point and `B point, start to link A point successively from initial point and `B point forms Guan Bi song
Line, becomes the ideograph of different sources crab；Statistical analysis obtains eigenvalue TjMeansigma methods `Tj, become different sources crab
Eigenvalue.
(5) the crab place of production differentiates, is first according to step (1), (2), the view data of (3) acquisition crab to be discriminated, its
Secondary crab ideograph to be discriminated and different sources crab ideograph are compared, the two initial point is overlapped, scales river to be discriminated
Eriocheir sinensis individual mode figure also moves rotation, and the part coincided together thinks that the two coincide, if the two goodness of fit reaches 75%
Above, then carry out eigenvalue comparison, if 0.95 < Tj/`Tj< think the most identical for 1.05, if the eigenvalue goodness of fit reaches
More than 90%, then it is assumed that crab individuality to be discriminated is that this place of production produces.
Claims (8)
1. the image-recognizing method differentiated for crab, including gathering crab image, Image semantic classification, structure river to be discriminated
The ideograph of Eriocheir sinensis and eigenvalue, structure known crab Population pattern figure and population characteristic value, crab individuality differentiate five steps, its
Feature is, it is first crab individual mode figure to be discriminated and known crab Population pattern figure to be compared that crab individuality differentiates, as
The two goodness of fit of fruit reaches more than 75%, then carries out eigenvalue comparison, if the eigenvalue goodness of fit reaches more than 90%, then it is assumed that treat
Differentiate that crab individuality is the member in known colony.
The image-recognizing method differentiated for crab the most according to claim 1, is characterized in that, gathers crab image and refers to
Obtain the carapace photo of crab.
The image-recognizing method differentiated for crab the most according to claim 1, is characterized in that, Image semantic classification is the most right
Carapace image carries out dimension-reduction treatment, retains in carapace edge contour, the central point of protrusion of surface and special color region
Heart point, remainder transparence；Secondly carapace contour line reveals process, then selected characteristic point in units of pixel: with head
Incising depression deepest point in the middle of cuirass volume tooth is that initial point sets up plane coordinate system, draws the coordinate of each edge pixel point,
The turning point being positioned on carapace edge wheel profile is set to A point, starts to be denoted as clockwise A from initial pointi(i=1,2,3 ... n), often
Between two A points, the 25%th of pixel quantity the, 50%, 75% point is set to B point, starts to be denoted as clockwise B from initial pointi(i=1,2,
3 ... n), it is C point the location of the core of each projection on carapace, starts to be denoted as clockwise C from initial pointi(i=1,2,
3 ... n), it is D point the location of the core in special color region on carapace, starts to be denoted as clockwise D from initial pointi(i=1,2,
3 ... n).
The image-recognizing method differentiated for crab the most according to claim 1, is characterized in that, start chain successively from initial point
Connect A point and B point forms closed curve, become crab ideograph.
The image-recognizing method differentiated for crab the most according to claim 1, is characterized in that, measure Ai、Bi、Ci、Di
Each point is denoted as LA respectively to the distance of initial pointi、LBi、LCi、LDi, i=1,2,3 ... n, measure the first side facewidth and be denoted as L0, meter
Calculate eigenvalue Tj, Tj=Lj/L0,j=Ai、Bi、Ci、Di。
The image-recognizing method differentiated for crab the most according to claim 1, is characterized in that, it is thus achieved that every in crab colony
The view data of individuality, statistical analysis obtains LAi、LBi, meansigma methods `LAi、`LBi, and then obtain A point and B point average bit
Put `A point and `B point, start to link `A point successively from initial point and `B point forms closed curve, become the ideograph of crab colony；System
Meter analysis obtains eigenvalue TjMeansigma methods `Tj, become the eigenvalue of crab colony.
The image-recognizing method differentiated for crab the most according to claim 1, is characterized in that, crab individuality mould to be discriminated
It is that the two initial point is overlapped that formula figure and known crab Population pattern figure are compared, and scales crab individual mode figure to be discriminated and goes forward side by side
Row is mobile to be rotated, and the part coincided together thinks that the two coincide.
The image-recognizing method differentiated for crab the most according to claim 1, is characterized in that, crab eigenvalue to be discriminated
TjWith known crab population characteristic value `TjCompare and be divided by both 18, if 0.95 < Tj/`Tj< think that the two is kissed for 1.05
Close.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610691336.6A CN106305567A (en) | 2016-08-21 | 2016-08-21 | Image identification method used for judging river crabs |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610691336.6A CN106305567A (en) | 2016-08-21 | 2016-08-21 | Image identification method used for judging river crabs |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106305567A true CN106305567A (en) | 2017-01-11 |
Family
ID=57743236
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610691336.6A Pending CN106305567A (en) | 2016-08-21 | 2016-08-21 | Image identification method used for judging river crabs |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106305567A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108898161A (en) * | 2018-06-07 | 2018-11-27 | 中国水产科学研究院淡水渔业研究中心 | A kind of Coilia fishes population method of discrimination based on Otolith Morphology |
| CN109214834A (en) * | 2018-09-10 | 2019-01-15 | 百度在线网络技术（北京）有限公司 | Product traceability method and apparatus based on block chain |
| CN110235815A (en) * | 2019-06-21 | 2019-09-17 | 山东省海洋生物研究院 | The accurate screening technique of turbot grace bodily form family based on geometric state |
| CN113610540A (en) * | 2021-07-09 | 2021-11-05 | 北京农业信息技术研究中心 | River crab anti-counterfeiting tracing method and system |
| CN114241248A (en) * | 2022-02-24 | 2022-03-25 | 北京市农林科学院信息技术研究中心 | River crab origin tracing method and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1609807A (en) * | 2003-10-20 | 2005-04-27 | 上海科技馆 | Method and apparatus for simulating fishing with computer and video equipment |
| US20090105952A1 (en) * | 2007-10-19 | 2009-04-23 | Grace Ted V | Watercraft automation and aquatic effort data utilization |
| CN204168907U (en) * | 2014-10-29 | 2015-02-25 | 中国水产科学研究院淡水渔业研究中心 | A kind of fish-catching apparatus of automatic selection fishing kind |
| CN104969875A (en) * | 2015-07-23 | 2015-10-14 | 中山大学深圳研究院 | Pet behavior detection system based on image change |
| CN105028356A (en) * | 2015-07-25 | 2015-11-11 | 李学新 | Underwater fish catching method based on image processing |
| CN105164521A (en) * | 2013-02-06 | 2015-12-16 | 加拿大北大西洋海鲜渔业公司 | Imaging for determination of crustacean physical attributes |
| CN105145423A (en) * | 2015-07-31 | 2015-12-16 | 江苏省淡水水产研究所 | Method for quickly determining various ecdysis periods of river crabs |
-
2016
- 2016-08-21 CN CN201610691336.6A patent/CN106305567A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1609807A (en) * | 2003-10-20 | 2005-04-27 | 上海科技馆 | Method and apparatus for simulating fishing with computer and video equipment |
| US20090105952A1 (en) * | 2007-10-19 | 2009-04-23 | Grace Ted V | Watercraft automation and aquatic effort data utilization |
| CN105164521A (en) * | 2013-02-06 | 2015-12-16 | 加拿大北大西洋海鲜渔业公司 | Imaging for determination of crustacean physical attributes |
| CN204168907U (en) * | 2014-10-29 | 2015-02-25 | 中国水产科学研究院淡水渔业研究中心 | A kind of fish-catching apparatus of automatic selection fishing kind |
| CN104969875A (en) * | 2015-07-23 | 2015-10-14 | 中山大学深圳研究院 | Pet behavior detection system based on image change |
| CN105028356A (en) * | 2015-07-25 | 2015-11-11 | 李学新 | Underwater fish catching method based on image processing |
| CN105145423A (en) * | 2015-07-31 | 2015-12-16 | 江苏省淡水水产研究所 | Method for quickly determining various ecdysis periods of river crabs |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108898161A (en) * | 2018-06-07 | 2018-11-27 | 中国水产科学研究院淡水渔业研究中心 | A kind of Coilia fishes population method of discrimination based on Otolith Morphology |
| CN109214834A (en) * | 2018-09-10 | 2019-01-15 | 百度在线网络技术（北京）有限公司 | Product traceability method and apparatus based on block chain |
| CN110235815A (en) * | 2019-06-21 | 2019-09-17 | 山东省海洋生物研究院 | The accurate screening technique of turbot grace bodily form family based on geometric state |
| CN113610540A (en) * | 2021-07-09 | 2021-11-05 | 北京农业信息技术研究中心 | River crab anti-counterfeiting tracing method and system |
| CN113610540B (en) * | 2021-07-09 | 2024-02-02 | 北京农业信息技术研究中心 | River crab anti-counterfeiting tracing method and system |
| CN114241248A (en) * | 2022-02-24 | 2022-03-25 | 北京市农林科学院信息技术研究中心 | River crab origin tracing method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106305567A (en) | Image identification method used for judging river crabs | |
| Zion et al. | Real-time underwater sorting of edible fish species | |
| CN102013021B (en) | Tea tender shoot segmentation and identification method based on color and region growth | |
| Sacchi et al. | Photographic identification in reptiles: a matter of scales | |
| CN111294565A (en) | Intelligent pig raising monitoring method and management terminal | |
| Costa et al. | Preliminary evidences of colour differences in European sea bass reared under organic protocols | |
| CN113379769A (en) | Intelligent defense platform for crop diseases and insect pests | |
| Colihueque et al. | Skin color characterization in rainbow trout by use of computer-based image analysis | |
| Yu et al. | Chicken embryo fertility detection based on PPG and convolutional neural network | |
| CN109191461A (en) | A kind of Countryside Egg recognition methods and identification device based on machine vision technique | |
| CN115100513B (en) | Method and system for estimating food intake of breeding object based on computer vision recognition | |
| CN110414369B (en) | Cow face training method and device | |
| CN108133184B (en) | Fish identification and vaccine injection method based on fractal theory and BP algorithm | |
| CN116912260B (en) | Broiler chicken breeding health state detection method based on artificial intelligence | |
| Ibaraki et al. | Image analysis for plants: basic procedures and techniques | |
| Sazzad et al. | RoseNet: Rose leave dataset for the development of an automation system to recognize the diseases of rose | |
| CN104101570B (en) | A kind of infra-red sepectrometry detects the method in the hairtail place of production | |
| Zhang et al. | Analysis of selective breeding of nacre color in two strains of Hyriopsis cumingii Lea based on the cielab colorspace | |
| Tu et al. | A study of broccoli grading system based on machine vision and neural networks | |
| VinceCruz-Abeledo et al. | Comparison of morphometric identification of species in juvenile mangrove crabs (Genus Scylla) by automated and local approaches | |
| Kaur et al. | Tea Leaf Disease Detection using Deep Learning Convolutional Neural Network Model | |
| Guerrero et al. | Postcranial anomalies of Eocene freshwater pleurodiran and cryptodiran turtles from the Spanish Duero Basin | |
| CN112598079A (en) | Method for identifying cephalopod population and species | |
| CN109658409A (en) | A kind of Leaf of banana leaf spot detection method | |
| Wang et al. | A grapevine virus disease detection method based on convolution neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170111 |