CN105678749A - Fresh fish direction discrimination method based on visual sense - Google Patents
Fresh fish direction discrimination method based on visual sense Download PDFInfo
- Publication number
- CN105678749A CN105678749A CN201511024610.6A CN201511024610A CN105678749A CN 105678749 A CN105678749 A CN 105678749A CN 201511024610 A CN201511024610 A CN 201511024610A CN 105678749 A CN105678749 A CN 105678749A
- Authority
- CN
- China
- Prior art keywords
- fish
- circle
- center
- flake
- candidate
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30128—Food products
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a fresh fish direction discrimination method based on visual sense, and belongs to the visual sense positioning detection field. The fresh fish direction discrimination method based on visual sense comprises the steps: acquiring a gray level image of the fish, and establishing an x, y coordinate system; converting the gray level image into a binary image; acquiring the contour boundary of the fish body; utilizing a Hough gradient method to detect all the circles which conform to the preset fisheye radius scope in the binary image; finding out candidate circles for the fisheye; when a plurality of candidate circles appear, according to the features of the fisheye, preferably selecting the center of one candidate circle as the center of the fisheye; acquiring the coordinate (X0, Y0) of the center of the fisheye, wherein the coordinate of the point which is closes to the center of the fisheye on the contour boundary of the fish body is (X1, Y1); and the coordinate of the X axis corresponding to the vertical center line of the contour boundary of the fish body is: X2=(Xmax+Xmin)/2; and determining the direction of the fish head and fish belly. The fresh fish direction discrimination method based on visual sense can quickly determine the direction of the fish head and fish belly, and can improve the degree of automation for deep processing of fish.
Description
Technical field
The invention belongs to vision localization detection field, particularly to a kind of fresh fish discriminating direction method of view-based access control model.
Background technology
China's aquaculture production accounts for the 70% of Gross World Product, but 1/3rd of processing capacity also not enough total output. In process technology and equipment, there is bigger gap with the advanced country such as German, Japanese.
In Fish are processed, conventional manufacturing procedure includes cleaning, decaptitate, go dirty, open back of the body etc., owing to operation is numerous and diverse, equipment technology backwardness, current most of the aquatic products processing enterprises are still based on manual operations. Along with the continuous lifting of labor cost, mechanization and automatization are the inevitable directions of the aquatic products processing enterprises development. In recent years, along with the development of science and technology, from the cleaning of Fish, scale, dirty to cutting off, going, all research and develop and defined automation equipment, can be used for large-scale production. But the sequence of Fish is put, especially fish the back of the body and fish belly towards, remain the difficult point of industrialized production, it is necessary to manually put. Not only production efficiency is low, also cannot realize full-automatic production continuously.
Summary of the invention
Present invention aims to the problems referred to above, it is provided that a kind of fresh fish discriminating direction method of mechanically-based vision, it may be achieved fish belly towards automatic identification, for Fish Mechanized Processing.
The object of the present invention is achieved like this:
A kind of fresh fish discriminating direction method of view-based access control model, the method comprises the steps:
Step 1, shoots the photo to obtain fish with camera, then photo is converted to gray level image, and sets up x, y coordinate system;
Step 2, adopts artificial selection's method or maximum variance between clusters to determine the binary-state threshold of described gray level image, with by described greyscale image transitions for bianry image;
Step 3, carries out contour encoding to described bianry image, to obtain the profile border of fish body;
Step 4, presets the radius of flake, adopts Hough gradient method to detect all circles meeting default flake radius of the gray level image being arranged in fish body profile border;
Step 5, preset the flake center distance range to nearest fish body profile border, then calculate the center of circle distance to nearest fish body profile border of all circles met in default flake radius, and it is round as the candidate of flake to retain the circle meeting within the scope of predeterminable range;
Step 6, when only one of which candidate's bowlder, directly performs step 7;
And when multiple candidate's bowlders occur, it is necessary to according to the center as flake, the center of circle of flake feature preferably candidate circle; Concrete preferred process is as follows:
The center of circle of statistics candidate's circle is in the gray average μ specified in radius and variances sigma2:
Wherein, ZiCenter of circle gray value of i-th pixel in appointment radius for candidate's circle;
By μ+k σ2As index, k is predetermined coefficient, and the center of circle of candidate's circle that selective goal is minimum is as the center of flake;
According to processing fingerling class, whether step 7, judge that fish mouth is opened and flake center can be made to drop on the profile border of fish mouth to the point that fish body profile border is nearest; If meeting, then perform step 8; If will not, then perform step 9;
Step 8, Graham scanning method is adopted to obtain the convex closure winding thread on fish body profile border, then the oculocentric coordinate (X0 of fish is obtained, Y0), obtain the coordinate (X1 of point nearest to this flake center on convex closure winding thread, Y1), the x-axis coordinate calculating the median vertical line obtaining convex closure winding thread corresponding after obtaining x-axis coordinate range Xmin and the Xmax that convex closure winding thread is corresponding is X2=(Xmax+Xmin)/2;
Step 9, obtain the oculocentric coordinate (X0 of fish, Y0), obtain the coordinate (X1 of point nearest to this flake center on fish body profile border, Y1), the x-axis coordinate calculating the median vertical line obtaining fish body profile border corresponding after obtaining x-axis coordinate range Xmin and the Xmax that fish body silhouette edge bound pair is answered is X2=(Xmax+Xmin)/2;
Step 10, it is judged that the direction of fish head and fish belly; Work as X0>X2, represent fish head forward; As X0<X2, represent fish head backward; Work as Y0>Y1, represent fish belly upwards; Y0<Y1, represents that fish belly is downward.
Wherein, step 4 comprises the following specific steps that:
Step 4.1, adopts Canny algorithm to search all edges of the gray level image being arranged in fish body profile border, then adopts Sobel operator to calculate the partial gradient of each marginal point, calculate process as follows:
If original image is:
Wherein, Z5For the gray value of marginal point to be calculated, all the other are the gray value of the point in field around marginal point to be calculated;
Using Sobel operator to calculate gray value is Z5The approximate partial derivative of marginal point:
Gx=(Z7+2Z8+Z9)-(Z1+2Z2+Z3)
;
Gy=(Z3+2Z6+Z9)-(Z1+2Z4+Z7)
Wherein, GxIt is Z for gray value5Marginal point at x Directional partial derivative, GyIt is Z for gray value5Marginal point at y Directional partial derivative, then gray value is Z5The gradient magnitude of marginal pointGradient direction angle
Step 4.2, adds up each difference on the straight line of each marginal point gradient direction in two dimension accumulator, and these marginal points of labelling;
Step 4.3, to score in accumulator more than given threshold value and more than the some descending of its neighbour, it is thus achieved that the center of circle list of candidate's circle;
Step 4.4, the center of circle that each candidate is justified is ranked up to the distance of marginal point, from the maximum radius of default flake to least radius, selects the radius that edge pixel is supported most, if the center of circle of this candidate circle is subject to enough marginal point supports, then retains this circle.
The invention have the benefit that the direction using this method can quickly judge fish head and fish belly, improve the automaticity of Fish deep processing, both improve production efficiency, and reduced again labor intensity.
Accompanying drawing explanation
Fig. 1 is the FB(flow block) of the present invention.
Fig. 2 is the gray level image being placed in x, y coordinate system.
Fig. 3 is binary image after Threshold segmentation.
Fig. 4 is the image multiple candidates circle occur.
Fig. 5 is preferred rear flake location drawing picture.
Fig. 6 is the image after obtaining convex closure winding thread.
Fig. 7 is the x after labelling coordinate figure, y coordinate system image.
Detailed description of the invention
Below in conjunction with specific embodiments and the drawings, the present invention is expanded on further.
As it is shown in figure 1, a kind of fresh fish discriminating direction method of view-based access control model, the method comprises the steps:
Step 1, shoots the photo to obtain fish with camera, then photo is converted to gray level image, and sets up x, y coordinate system, as shown in Figure 2.
Step 2, adopts artificial selection's method or maximum variance between clusters to determine the binary-state threshold of described gray level image, with by described greyscale image transitions for bianry image, as shown in Figure 3.
Step 3, carries out contour encoding based on SATOSHISUZUK algorithm to described bianry image, to obtain the profile border of fish body. Owing to fish body occupies maximum area in picture, fish body profile border should be area and meets the maximum outline of pre-conditioned (minimum area≤contour area≤maximum area).
Step 4, presets the radius of flake, adopts Hough gradient method to detect all circles meeting default flake radius of the gray level image being arranged in fish body profile border. Its concrete steps include:
Step 4.1, adopts Canny algorithm to search all edges of the gray level image being arranged in fish body profile border, then adopts Sobel operator to calculate the partial gradient of each marginal point, calculate process as follows:
If original image is:
Wherein, Z5For the gray value of marginal point to be calculated, all the other are the gray value of the point in field around marginal point to be calculated;
Using Sobel operator to calculate gray value is Z5The approximate partial derivative of marginal point:
Gx=(Z7+2Z8+Z9)-(Z1+2Z2+Z3)
;
Gy=(Z3+2Z6+Z9)-(Z1+2Z4+Z7)
Wherein, GxIt is Z for gray value5Marginal point at x Directional partial derivative, GyIt is Z for gray value5Marginal point at y Directional partial derivative, then gray value is Z5The gradient magnitude of marginal pointGradient direction angle
Step 4.2, adds up each difference on the straight line of each marginal point gradient direction in two dimension accumulator, and these marginal points of labelling;
Step 4.3, to score in accumulator more than given threshold value and more than the some descending of its neighbour, it is thus achieved that the center of circle list of candidate's circle;
Step 4.4, the center of circle that each candidate is justified is ranked up to the distance of marginal point, from the maximum radius of default flake to least radius, selects the radius that edge pixel is supported most, if the center of circle of this candidate circle is subject to enough marginal point supports, then retains this circle.
Step 5, preset the flake center distance range to nearest fish body profile border, then calculate the center of circle distance to nearest fish body profile border of all circles met in default flake radius, and it is round as the candidate of flake to retain the circle meeting within the scope of predeterminable range.
Step 6, when only one of which candidate's bowlder, directly performs step 7.
And would be likely to occur error owing to gradient direction calculates, it is therefore more likely that there is multiple candidate circle, as shown in Figure 4. At this time, it may be necessary to according to flake feature (in most cases the crystalline lens of fresh fish is the part and color even that fish body gray scale is minimum), it is preferable that the center of circle of candidate's circle is as the center of flake. Concrete preferred process is as follows:
The center of circle of statistics candidate's circle is in the gray average μ specified in radius and variance
Wherein, ZiCenter of circle gray value of i-th pixel in appointment radius for candidate's circle;
By μ+k σ2As index, k is predetermined coefficient, and the center of circle of candidate's circle that selective goal is minimum is as the center of flake, and result is as shown in Figure 5.
Step 7, the fish of the Partial Species such as picture Carnis Pseudosciaenae is under mouth splayed condition, fish mouth is oculocentric closest from fish, now the walking direction of fish belly can be produced interference, thus, whether now need to judge that fish mouth is opened can make flake center to the point that fish body profile border is nearest drop on the profile border of fish mouth; If meeting, then perform step 8; If will not, then perform step 9.
Step 8, Graham scanning method is adopted to obtain the convex closure winding thread on fish body profile border, as shown in Figure 6, then the oculocentric coordinate (X0 of fish is obtained, Y0), the coordinate (X1, Y1) of point nearest to this flake center on convex closure winding thread is obtained, the x-axis coordinate calculating the median vertical line obtaining convex closure winding thread corresponding after obtaining x-axis coordinate range Xmin and the Xmax that convex closure winding thread is corresponding is X2=(Xmax+Xmin)/2, as shown in Figure 6.
Step 9, obtain the oculocentric coordinate (X0 of fish, Y0), obtain the coordinate (X1 of point nearest to this flake center on fish body profile border, Y1), the x-axis coordinate calculating the median vertical line obtaining fish body profile border corresponding after obtaining x-axis coordinate range Xmin and the Xmax that fish body silhouette edge bound pair is answered is X2=(Xmax+Xmin)/2, as shown in Figure 7.
Step 10, according to the coordinate figure that step 8 or step 9 obtain, it is judged that the direction of fish head and fish belly. Work as X0>X2, represent fish head forward; As X0<X2, represent fish head backward; Work as Y0>Y1, represent fish belly upwards; Y0<Y1, represents that fish belly is downward.
The present invention can quickly judge the direction of flake and fish belly, for Fish Mechanized Processing, can improve the automaticity of Fish processing, effectively reduces the human cost of processing, and reliability is high.
Claims (2)
1. the fresh fish discriminating direction method of a view-based access control model, it is characterised in that the method comprises the steps:
Step 1, shoots the photo to obtain fish with camera, then photo is converted to gray level image, and sets up x, y coordinate system;
Step 2, adopts artificial selection's method or maximum variance between clusters to determine the binary-state threshold of described gray level image, with by described greyscale image transitions for bianry image;
Step 3, carries out contour encoding to described bianry image, to obtain the profile border of fish body;
Step 4, presets the radius of flake, adopts Hough gradient method to detect all circles meeting default flake radius of the gray level image being arranged in fish body profile border;
Step 5, preset the flake center distance range to nearest fish body profile border, then calculate the center of circle distance to nearest fish body profile border of all circles met in default flake radius, and it is round as the candidate of flake to retain the circle meeting within the scope of predeterminable range;
Step 6, when only one of which candidate's bowlder, directly performs step 7;
And when multiple candidate's bowlders occur, it is necessary to according to the center as flake, the center of circle of flake feature preferably candidate circle; Concrete preferred process is as follows:
The center of circle of statistics candidate's circle is in the gray average μ specified in radius and variances sigma2:
Wherein, ZiCenter of circle gray value of i-th pixel in appointment radius for candidate's circle;
By μ+k σ2As index, k is predetermined coefficient, and the center of circle of candidate's circle that selective goal is minimum is as the center of flake;
According to processing fingerling class, whether step 7, judge that fish mouth is opened and flake center can be made to drop on the profile border of fish mouth to the point that fish body profile border is nearest; If meeting, then perform step 8;If will not, then perform step 9;
Step 8, Graham scanning method is adopted to obtain the convex closure winding thread on fish body profile border, then the oculocentric coordinate (X0 of fish is obtained, Y0), obtain the coordinate (X1 of point nearest to this flake center on convex closure winding thread, Y1), the x-axis coordinate calculating the median vertical line obtaining convex closure winding thread corresponding after obtaining x-axis coordinate range Xmin and the Xmax that convex closure winding thread is corresponding is X2=(Xmax+Xmin)/2;
Step 9, obtain the oculocentric coordinate (X0 of fish, Y0), obtain the coordinate (X1 of point nearest to this flake center on fish body profile border, Y1), the x-axis coordinate calculating the median vertical line obtaining fish body profile border corresponding after obtaining x-axis coordinate range Xmin and the Xmax that fish body silhouette edge bound pair is answered is X2=(Xmax+Xmin)/2;
Step 10, it is judged that the direction of fish head and fish belly; As X0 > X2, represent fish head forward; As X0 < X2, represent fish head backward; As Y0 > Y1, represent fish belly upwards; Y0 < Y1, represents that fish belly is downward.
2. the fresh fish discriminating direction method of a view-based access control model, it is characterised in that described step 4 comprises the following specific steps that:
Step 4.1, adopts Canny algorithm to search all edges of the gray level image being positioned at fish body profile border, then adopts Sobel operator to calculate the partial gradient of each marginal point, calculate process as follows:
If original image is:
Wherein, Z5For the gray value of marginal point to be calculated, all the other are the gray value of the point in field around marginal point to be calculated;
Using Sobel operator to calculate gray value is Z5The approximate partial derivative of marginal point:
Gx=(Z7+2Z8+Z9)-(Z1+2Z2+Z3)
;
Gy=(Z3+2Z6+Z9)-(Z1+2Z4+Z7)
Wherein, GxIt is Z for gray value5Marginal point at x Directional partial derivative, GyIt is Z for gray value5Marginal point at y Directional partial derivative, then gray value is Z5The gradient magnitude of marginal pointGradient direction angle
Step 4.2, adds up each difference on the straight line of each marginal point gradient direction in two dimension accumulator, and these marginal points of labelling;
Step 4.3, to score in accumulator more than given threshold value and more than the some descending of its neighbour, it is thus achieved that the center of circle list of candidate's circle;
Step 4.4, the center of circle that each candidate is justified is ranked up to the distance of marginal point, from the maximum radius of default flake to least radius, selects the radius that edge pixel is supported most, if the center of circle of this candidate circle is subject to enough marginal point supports, then retains this circle.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201511024610.6A CN105678749B (en) | 2015-12-30 | 2015-12-30 | A kind of fresh fish discriminating direction method of view-based access control model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201511024610.6A CN105678749B (en) | 2015-12-30 | 2015-12-30 | A kind of fresh fish discriminating direction method of view-based access control model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105678749A true CN105678749A (en) | 2016-06-15 |
CN105678749B CN105678749B (en) | 2018-03-30 |
Family
ID=56189848
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201511024610.6A Active CN105678749B (en) | 2015-12-30 | 2015-12-30 | A kind of fresh fish discriminating direction method of view-based access control model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105678749B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106417104A (en) * | 2016-08-31 | 2017-02-22 | 北京农业信息技术研究中心 | Cultured fish swimming direction detecting system and method |
CN109127462A (en) * | 2018-08-22 | 2019-01-04 | 深圳控石智能系统有限公司 | A kind of sausage intelligent sorting method of view-based access control model guidance |
CN109345554A (en) * | 2018-09-12 | 2019-02-15 | 南京农业大学 | A kind of adhesion mushroom vision in-situ measuring method based on RGB-D camera |
CN109900719A (en) * | 2019-03-04 | 2019-06-18 | 华中科技大学 | A kind of visible detection method of blade surface knife mark |
CN112017201A (en) * | 2020-08-07 | 2020-12-01 | 湖北省农业科学院农产品加工与核农技术研究所 | Method for judging head and tail postures of fish body in processing and conveying |
CN112017200A (en) * | 2020-08-07 | 2020-12-01 | 湖北省农业科学院农产品加工与核农技术研究所 | Fish posture judgment method based on visual recognition |
CN115336624A (en) * | 2022-08-16 | 2022-11-15 | 中国水产科学研究院渔业机械仪器研究所 | Intelligent fish body cutting device based on image recognition and control method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6697535B1 (en) * | 1999-04-30 | 2004-02-24 | Cognex Technology And Investment Corporation | Method for refining a parameter of a contour in an image |
CN102278977A (en) * | 2011-08-05 | 2011-12-14 | 中国科学院自动化研究所 | Infrared visual positioning system of robotic fish |
CN104298996A (en) * | 2014-08-08 | 2015-01-21 | 中国科学院自动化研究所 | Underwater active vision tracking method applied to bionic robot fish |
CN104881045A (en) * | 2015-06-17 | 2015-09-02 | 中国科学院自动化研究所 | Bionic robot fish three-dimensional tracking method based on embedded visual guidance |
-
2015
- 2015-12-30 CN CN201511024610.6A patent/CN105678749B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6697535B1 (en) * | 1999-04-30 | 2004-02-24 | Cognex Technology And Investment Corporation | Method for refining a parameter of a contour in an image |
CN102278977A (en) * | 2011-08-05 | 2011-12-14 | 中国科学院自动化研究所 | Infrared visual positioning system of robotic fish |
CN104298996A (en) * | 2014-08-08 | 2015-01-21 | 中国科学院自动化研究所 | Underwater active vision tracking method applied to bionic robot fish |
CN104881045A (en) * | 2015-06-17 | 2015-09-02 | 中国科学院自动化研究所 | Bionic robot fish three-dimensional tracking method based on embedded visual guidance |
Non-Patent Citations (4)
Title |
---|
HANMEI HONG 等: "Visual quality detection of a aquatic products using machine vision", 《AQUACULTURAL ENGINEERING》 * |
VASSILIS M.PAPADAKIS 等: "Sub-second analysis of fish behavior using a novel computer-vision system", 《AQUACULTURAL ENGINEERING》 * |
YANGHONG ZHONG 等: "Vision-based 3D Tracking System for Fish Interactive Analysis", 《2015 BY THE INFORMATION PROCESSING SOCIETY OF JAPAN》 * |
庞静洁: "基于视频的鱼类运动目标检测与跟踪方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106417104A (en) * | 2016-08-31 | 2017-02-22 | 北京农业信息技术研究中心 | Cultured fish swimming direction detecting system and method |
CN109127462A (en) * | 2018-08-22 | 2019-01-04 | 深圳控石智能系统有限公司 | A kind of sausage intelligent sorting method of view-based access control model guidance |
CN109127462B (en) * | 2018-08-22 | 2020-09-04 | 深圳控石智能系统有限公司 | Intelligent sausage sorting method based on visual guidance |
CN109345554A (en) * | 2018-09-12 | 2019-02-15 | 南京农业大学 | A kind of adhesion mushroom vision in-situ measuring method based on RGB-D camera |
CN109900719A (en) * | 2019-03-04 | 2019-06-18 | 华中科技大学 | A kind of visible detection method of blade surface knife mark |
CN112017201A (en) * | 2020-08-07 | 2020-12-01 | 湖北省农业科学院农产品加工与核农技术研究所 | Method for judging head and tail postures of fish body in processing and conveying |
CN112017200A (en) * | 2020-08-07 | 2020-12-01 | 湖北省农业科学院农产品加工与核农技术研究所 | Fish posture judgment method based on visual recognition |
CN112017200B (en) * | 2020-08-07 | 2022-07-05 | 湖北省农业科学院农产品加工与核农技术研究所 | Fish posture judgment method based on visual recognition |
CN112017201B (en) * | 2020-08-07 | 2024-03-19 | 湖北省农业科学院农产品加工与核农技术研究所 | Fish body head and tail gesture judging method in processing and conveying |
CN115336624A (en) * | 2022-08-16 | 2022-11-15 | 中国水产科学研究院渔业机械仪器研究所 | Intelligent fish body cutting device based on image recognition and control method |
Also Published As
Publication number | Publication date |
---|---|
CN105678749B (en) | 2018-03-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105678749A (en) | Fresh fish direction discrimination method based on visual sense | |
CN107798330B (en) | Weld image feature information extraction method | |
Fabic et al. | Fish population estimation and species classification from underwater video sequences using blob counting and shape analysis | |
CN107941808B (en) | 3D printing forming quality detection system and method based on machine vision | |
CN114821114B (en) | Groove cutting robot image processing method based on vision system | |
CN108960011B (en) | Partially-shielded citrus fruit image identification method | |
CN104156951B (en) | A kind of white blood cell detection method for BAL fluid smear | |
CN103914827B (en) | The visible detection method of weather strip for automobile profile defects | |
CN102110294B (en) | Method and system for processing image of diseased fish body | |
CN108564573B (en) | Automatic fish head and tail image identification method for spanish mackerel cutting machine | |
CN107403435B (en) | Color image edge extraction method based on RGB color space | |
CN107256398B (en) | Feature fusion based individual milk cow identification method | |
CN110930357A (en) | In-service steel wire rope surface defect detection method and system based on deep learning | |
CN111080696B (en) | Underwater sea cucumber identification and positioning method based on computer vision | |
CN108133471B (en) | Robot navigation path extraction method and device based on artificial bee colony algorithm | |
CN109781737B (en) | Detection method and detection system for surface defects of hose | |
CN110738631A (en) | prawn shape information detection method and detection system based on images | |
CN104515479A (en) | Medium plate surface plate shape measuring system and method | |
CN111583193A (en) | Pistachio nut framework extraction device based on geometric contour template matching and algorithm thereof | |
CN109544513A (en) | A kind of steel pipe end surface defect extraction knowledge method for distinguishing | |
CN101464998A (en) | Non-gauss veins noise smooth filtering method for textile industry | |
CN111369497B (en) | Walking type tree fruit continuous counting method and device | |
CN110298856B (en) | Non-contact type fish body length estimation method | |
CN111815542A (en) | Tree ring image medullary center positioning and ring measuring method | |
CN108171713A (en) | The robot of view-based access control model quickly identifies and three-dimensional visual positioning method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |