CN113327282A - Office swivel chair punching position and connection point identification method - Google Patents
Office swivel chair punching position and connection point identification method Download PDFInfo
- Publication number
- CN113327282A CN113327282A CN202110389543.7A CN202110389543A CN113327282A CN 113327282 A CN113327282 A CN 113327282A CN 202110389543 A CN202110389543 A CN 202110389543A CN 113327282 A CN113327282 A CN 113327282A
- Authority
- CN
- China
- Prior art keywords
- image
- particle swarm
- optimal solution
- value
- punching
- 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
- 238000004080 punching Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 28
- 239000002245 particle Substances 0.000 claims abstract description 43
- 239000011159 matrix material Substances 0.000 claims abstract description 22
- 238000005457 optimization Methods 0.000 claims abstract description 18
- 230000009466 transformation Effects 0.000 claims abstract description 4
- 238000012545 processing Methods 0.000 claims description 10
- 238000003066 decision tree Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000013145 classification model Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 230000001133 acceleration Effects 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000003707 image sharpening Methods 0.000 claims description 3
- 239000002023 wood Substances 0.000 abstract description 4
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23211—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with adaptive number of clusters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/48—Extraction of image or video features by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation
Abstract
The invention provides a method for identifying punching positions and connection points of office swivel chairs, which comprises the following steps: collecting the images of the swivel chair wood boards, and determining the accurate positions of the arcs by utilizing a random Hough transformation algorithm; extracting an image by taking a circle center coordinate as a central point and 2R as a radius, constructing a gray level co-occurrence matrix according to the extracted image, calculating characteristic parameters of the matrix, inputting characteristic parameter data into a particle swarm optimization algorithm, obtaining a global optimal solution and a local optimal solution of a generated particle swarm by utilizing the particle swarm optimization algorithm, and updating the speed and the position of particles; dividing the global optimal solution and the local optimal solution of the particle swarm obtained by the particle swarm optimization algorithm into a plurality of clustering centers to obtain corresponding membership functions; acquiring a membership matrix according to a result obtained by the membership function, and solving an optimal clustering number; the collected images are divided into two types of 'punching positions' and 'connection points' according to labels and are determined as final classification categories.
Description
Technical Field
The invention relates to the field of images, in particular to a method for identifying punching positions and connection points of office swivel chairs.
Background
The invention provides an office swivel chair punching position and connection point identification method based on image processing and algorithm identification, which is based on the fact that identification of punching positions and connection points on the existing office swivel chair production line mainly depends on manual experience judgment or a traditional geometric measurement positioning method and has the problems of low identification accuracy, poor standardization degree and the like.
Disclosure of Invention
The method provided by the invention mainly aims to overcome the defects in the prior art and provides a method for identifying the punching position and the connection point of the office swivel chair.
The invention adopts the following technical scheme:
a method for recognizing punching positions and connection points of office swivel chairs comprises the following steps:
acquiring a chair board image, and performing image preprocessing operation and data enhancement operation on the chair board image;
detecting the circular arc by using a random Hough transformation algorithm according to the number of falling points falling on the image, and determining the coordinate of the detected circular arc and the radius of the circular arc so as to determine the accurate position of the circular arc;
taking the circle center coordinate as a central point and 2R as a radius to extract an image, constructing a gray level co-occurrence matrix according to the extracted image, calculating characteristic parameters of the matrix, and carrying out normalization processing on characteristic parameter data;
inputting the characteristic parameter data after normalization processing into a particle swarm optimization algorithm, obtaining a global optimal solution and a local optimal solution of the generated particle swarm by using the particle swarm optimization algorithm, and updating the speed and the position of the particles;
dividing the global optimal solution and the local optimal solution of the particle swarm obtained by the particle swarm optimization algorithm into a plurality of clustering centers, inputting the clustering centers into a fuzzy core clustering algorithm, performing iterative computation, searching the optimal solution by using a Lagrange multiplier method, and obtaining a corresponding membership function;
according to a maximum membership principle, solving the maximum value of a membership function as a termination condition of a fuzzy core clustering algorithm, if the termination condition is met, acquiring a membership matrix according to the result obtained by the membership function, classifying data samples, and taking the cluster number corresponding to the maximum value of a clustering effectiveness index function as an optimal cluster number;
and constructing a gradient lifting decision tree classification model according to the optimal clustering number, dividing the acquired image into two types of a punching position and a connecting point according to labels, inputting the two types of punching positions and connecting points into the gradient lifting decision tree classification model to obtain a class confidence probability value, calculating an average value of classification result confidence degrees of a classifier, and determining the class with high confidence degree as a final classification class.
Specifically, the image preprocessing operation and the data enhancement operation on the images of the boards of the swivel chair specifically include:
image pre-processing operations include, but are not limited to: carrying out image graying processing, Gaussian filtering and image sharpening;
data enhancement operations include, but are not limited to: image turning and rotating, image brightness adjusting and image regularization processing.
Specifically, the characteristic parameters include: energy value Eng, entropy value Ent, moment of inertia Mot, autocorrelation Cor.
Specifically, the energy value Eng is specifically calculated as:
let a two-dimensional image size be M × N, where M and N denote pixel coordinate values at row and column positions on the image, and where p (M, N) denotes an element value of the gray level co-occurrence matrix at the (M, N) position.
Specifically, the entropy value Ent is specifically calculated as:
specifically, the moment of inertia Mot is specifically calculated as:
specifically, the autocorrelation Cor is specifically calculated as:
wherein, mumμnAs mean, σ, of pixels in horizontal and vertical directions in the imagemσnIs the standard deviation of the pixels in the horizontal and vertical directions in the image.
Specifically, the obtaining of the global optimal solution and the local optimal solution of the generated particle swarm by using the particle swarm optimization algorithm and the updating of the speed and the position of the particle specifically include:
wherein, PkAnd VkRespectively representing the position and velocity, Z, of the kth particlekIndicating the current local optimum position, ZgIs the whole particle swarmThe optimal position is determined, lambda is the inertia weight, n is the iteration number, alpha1And alpha2As an acceleration constant, c1And c2Is a random number between 0 and 1.
Specifically, the particle swarm global optimal solution and the particle swarm local optimal solution obtained by the particle swarm optimization algorithm are divided into a plurality of clustering centers, the clustering centers are input into the fuzzy core clustering algorithm, iterative computation is carried out, the optimal solution is searched by using a Lagrange multiplier method, and a corresponding membership function is obtained, wherein the membership function specifically comprises the following steps:
wherein K represents a Gaussian kernel function, xe(e-1, 2, …, n) represents the data characteristic parameter of the e-th input, s represents the number of categories of the input characteristic parameters, ci,cj(i, j ═ 1,2, …, Q) respectively represent the ith and jth cluster centers, g represents the weighted value index, and the range of values is [1.5,2.5]。
Specifically, the cluster number corresponding to the maximum value of the cluster validity index function is used as the optimal cluster number, wherein the cluster validity index function has a specific formula:
wherein, CN is the cluster number corresponding to the maximum value of the Index of the cluster validity Index function, C in the formula is the number of data samples, UijAnd (4) representing the membership function value of the ith clustering center and the jth sample.
As can be seen from the above description of the present invention, compared with the prior art, the present invention has the following advantages:
(1) the invention utilizes random Hough transformation algorithm to detect the punching position and the connection point position in the office swivel chair, namely obtaining the detection circular arc of the punching position and the connection point; for the detected arc, performing category judgment on the detected arc by utilizing a particle swarm optimization algorithm, a fuzzy kernel clustering algorithm and a gradient lifting decision tree classification model; the positioning and the type judgment of the punching position and the connecting point position of the office swivel chair wood board are realized by detecting and classifying the image of the office swivel chair, and a high-precision identification and positioning method is provided for the punching and riveting links in the industrial automatic production of the swivel chair.
Drawings
Fig. 1 is a flowchart of an office swivel chair punching position and connection point identification algorithm according to an embodiment of the present invention.
Detailed Description
The invention is further described below by means of specific embodiments.
As shown in fig. 1, a flow chart of an office swivel chair punching position and connection point identification algorithm provided in the embodiment of the present application is as follows:
(1) starting a program, acquiring a board image of the swivel chair by a machine vision system, and performing image preprocessing operation (including image graying, Gaussian filtering and image sharpening) and data enhancement operation (including image turning and rotation, image brightness adjustment and image regularization);
(2) detecting the circular arc by using a Random Hough Transform (RHT) algorithm according to the number of falling points falling on the image, determining the coordinates (xc, yc) of the detected circular arc and the radius R of the circular arc so as to determine the accurate position of the circular arc and provide a visual identification detection basis for the wood board punching and locking operation of a connecting point;
(3) taking the circle center coordinate as a central point and 2R as a radius to extract an image, constructing a gray level co-occurrence matrix according to the extracted image, calculating characteristic parameters of the matrix, and taking the characteristic parameters as input image characteristics of a subsequent classifier, wherein the characteristic parameters needing to be calculated comprise an energy value Eng, an entropy value Ent, an inertia moment Mot and an autocorrelation Cor, and p (m, n) represents the element value of the gray level co-occurrence matrix at the (m, n) position;
the energy value Eng is calculated by the following method, which is the sum of squares of matrix element values and can represent the uniformity degree and the texture thickness of the image gray distribution:
setting a two-dimensional image size to M × N, where M and N represent pixel coordinate values at row and column positions on the image;
the entropy value Ent is calculated as the following, which is the product of the matrix element values and their logarithm values, and represents the degree of non-uniformity or complexity of the texture in the image:
the moment of inertia Mot is calculated as follows, and the value of the moment of inertia Mot reflects the definition degree of an image and the depth degree of texture grooves:
the autocorrelation Cor reflects the uniformity of the image texture, using the mean value of pixels μ in the horizontal and vertical directions in the imagem、μnAnd standard deviation sigmam、σnThe characterization is carried out, and the calculation method is as follows:
constructing gray level co-occurrence matrixes in eight directions in the range of 0-360 degrees of the image by taking 45 degrees as step lengths, calculating characteristic parameters of each gray level co-occurrence matrix, including an energy value Eng, an entropy value Ent and an inertia moment Mot, and forming a characteristic vector by autocorrelation Cor, and taking the characteristic vector as an input characteristic of a classifier algorithm formed on the basis of particle swarm optimization, mean clustering and gradient lifting decision trees;
(4) after the characteristic parameters of the matrix are obtained through calculation, normalization processing is carried out on the characteristic parameter data, and then the characteristic data are input into a Particle Swarm Optimization (PSO) algorithm: let X ═ X be the sample to be classified in the image sample set1,x2,…,xnIn which xi(i ═ 1,2, …, n) is the normalized data feature, the global optimal solution and the local optimal solution of the generated particle swarm are obtained using the Particle Swarm Optimization (PSO) algorithm, and the velocity and position of the particle are updated using the following equations:
wherein, PkAnd VkRespectively representing the position and velocity, Z, of the kth particlekIndicating the current local optimum position, ZgFor the global optimal position of the whole particle swarm, lambda is the inertia weight, n is the iteration number, alpha1And alpha2As an acceleration constant, c1And c2A random number between 0 and 1;
(5) obtaining global optimal and local optimal solutions of the particle swarm through a Particle Swarm Optimization (PSO) algorithm, dividing the optimal solutions into W clustering centers, inputting the W clustering centers into a fuzzy kernel clustering (KFCM) algorithm, performing iterative computation, searching the optimal solution by using a Lagrange multiplier method, and obtaining corresponding membership function muie(i-1, 2; e-1, 2, …, n) representing the membership function of the ith class corresponding to the ith sample, wherein the classes are the 'punching position' and the 'connection point position'; the membership function is as follows, where s represents the number of classes of the input characteristic parameter, ci,cj(i, j ═ 1,2, …, Q) respectively represent the ith and jth cluster centers, g represents the weighted value index, and the range of values is [1.5,2.5]；
(6) According to the principle of maximum membership degree, solving the maximum value of the membership degree function as the termination condition of the fuzzy core clustering algorithm, if the maximum value reaches the termination condition of the fuzzy core clustering algorithmAnd under the termination condition, acquiring a membership matrix (C rows and N columns) according to the result obtained by the membership function, dividing the data samples into C types, taking the cluster number CN corresponding to the maximum value of the cluster validity Index function Index as the optimal cluster number, wherein C is the number of the data samples, and U is the number of the data samplesijAnd (4) representing the membership function value of the ith clustering center and the jth sample.
(7) Construction according to the number of clusters CN
And (3) dividing the acquired image into two types of a punching position and a connecting point according to labels, inputting data into the M gradient lifting decision tree classification models to obtain M category confidence probability values, calculating the average value of the classification result confidence of the M classifiers, determining the category with high confidence as a final classification category, and finally ending the algorithm process.
The method for identifying the punching position and the connecting point of the office swivel chair can be used for positioning and identifying the punching of the office swivel chair wood plate image and the connecting point, and provides a high-precision visual identification basis for the punching rivet operation in the subsequent automatic swivel chair production
The above description is only an embodiment of the present invention, but the design concept of the present invention is not limited thereto, and any insubstantial modifications made by using the design concept should fall within the scope of infringing the present invention.
Claims (10)
1. A method for recognizing punching positions and connection points of office swivel chairs is characterized by comprising the following steps:
acquiring a chair board image, and performing image preprocessing operation and data enhancement operation on the chair board image;
detecting the circular arc by using a random Hough transformation algorithm according to the number of falling points falling on the image, and determining the coordinate of the detected circular arc and the radius of the circular arc so as to determine the accurate position of the circular arc;
taking the circle center coordinate as a central point and 2R as a radius to extract an image, constructing a gray level co-occurrence matrix according to the extracted image, calculating characteristic parameters of the matrix, and carrying out normalization processing on characteristic parameter data;
inputting the characteristic parameter data after normalization processing into a particle swarm optimization algorithm, obtaining a global optimal solution and a local optimal solution of the generated particle swarm by using the particle swarm optimization algorithm, and updating the speed and the position of the particles;
dividing the global optimal solution and the local optimal solution of the particle swarm obtained by the particle swarm optimization algorithm into a plurality of clustering centers, inputting the clustering centers into a fuzzy core clustering algorithm, performing iterative computation, searching the optimal solution by using a Lagrange multiplier method, and obtaining a corresponding membership function;
according to a maximum membership principle, solving the maximum value of a membership function as a termination condition of a fuzzy core clustering algorithm, if the termination condition is met, acquiring a membership matrix according to the result obtained by the membership function, classifying data samples, and taking the cluster number corresponding to the maximum value of a clustering effectiveness index function as an optimal cluster number;
and constructing a gradient lifting decision tree classification model according to the optimal clustering number, dividing the acquired image into two types of a punching position and a connecting point according to labels, inputting the two types of punching positions and connecting points into the gradient lifting decision tree classification model to obtain a class confidence probability value, calculating an average value of classification result confidence degrees of a classifier, and determining the class with high confidence degree as a final classification class.
2. The method for identifying punching positions and connection points of an office swivel chair as claimed in claim 1, wherein the image preprocessing operation and the data enhancement operation on the images of the swivel chair plank specifically comprise:
image pre-processing operations include, but are not limited to: carrying out image graying processing, Gaussian filtering and image sharpening;
data enhancement operations include, but are not limited to: image turning and rotating, image brightness adjusting and image regularization processing.
3. The method as claimed in claim 1, wherein the characteristic parameters include: energy value Eng, entropy value Ent, moment of inertia Mot, autocorrelation Cor.
4. The method for identifying the punching position and the connecting point of the office swivel chair as claimed in claim 1, wherein the energy value Eng is specifically calculated as:
let a two-dimensional image size be M × N, where M and N denote pixel coordinate values at row and column positions on the image, and p (M, N) denotes an element value of the gray level co-occurrence matrix at the (M, N) position.
5. The method for identifying the punching position and the connecting point of the office swivel chair according to claim 1, wherein the entropy value Ent is specifically calculated as:
let a two-dimensional image size be M × N, where M and N denote pixel coordinate values at row and column positions on the image, and p (M, N) denotes an element value of the gray level co-occurrence matrix at the (M, N) position.
6. The method for identifying the punching position and the connecting point of the office swivel chair according to claim 1, wherein the moment of inertia Mot is specifically calculated as:
let a two-dimensional image size be M × N, where M and N denote pixel coordinate values at row and column positions on the image, and p (M, N) denotes an element value of the gray level co-occurrence matrix at the (M, N) position.
7. The method as claimed in claim 1, wherein the autocorrelation Cor is calculated by:
wherein, a two-dimensional image is set to be M × N, wherein M and N represent pixel coordinate values at row and column positions on the image, p (M, N) represents an element value of the gray level co-occurrence matrix at the (M, N) position, and μmμnAs mean, σ, of pixels in horizontal and vertical directions in the imagemσnIs the standard deviation of the pixels in the horizontal and vertical directions in the image.
8. The method for identifying the punching position and the connection point of the office swivel chair according to claim 1, wherein the global optimal solution and the local optimal solution of the generated particle swarm are obtained by utilizing a particle swarm optimization algorithm, and the speed and the position of the particle are updated, specifically:
wherein, PkAnd VkRespectively representing the position and velocity, Z, of the kth particlekIndicating the current local optimum position, ZgFor the global optimal position of the whole particle swarm, lambda is the inertia weight, n is the iteration number, alpha1And alpha2As an acceleration constant, c1And c2Is a random number between 0 and 1.
9. The method for identifying punching positions and connection points of office swivels as claimed in claim 1, wherein the particle swarm global optimal and local optimal solutions obtained by the particle swarm optimization algorithm are divided into a plurality of clustering centers, the clustering centers are input to a fuzzy kernel clustering algorithm, iterative computation is performed, the optimal solution is found by using a lagrange multiplier method, and corresponding membership functions are obtained, wherein the membership functions specifically are as follows:
wherein K represents a Gaussian kernel function, xe(e 1,2, …, s) represents the data characteristic variable of the e-th input, s represents the class of the input characteristic variable, ci,cj(i, j ═ 1,2, …, Q) respectively represent the ith and jth cluster centers, g represents the weighted value index, and the range of values is [1.5,2.5]。
10. The method for identifying punching positions and connecting points of office swivels as claimed in claim 1, wherein the cluster number corresponding to the maximum value of the cluster validity indicator function is taken as the optimal cluster number, wherein the cluster validity indicator function has a specific formula:
wherein, CN is the cluster number corresponding to the maximum value of the Index of the cluster validity Index function, C in the formula is the number of data samples, UijAnd (4) representing the membership function value of the ith clustering center and the jth sample.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110389543.7A CN113327282A (en) | 2021-04-12 | 2021-04-12 | Office swivel chair punching position and connection point identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110389543.7A CN113327282A (en) | 2021-04-12 | 2021-04-12 | Office swivel chair punching position and connection point identification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113327282A true CN113327282A (en) | 2021-08-31 |
Family
ID=77414703
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110389543.7A Pending CN113327282A (en) | 2021-04-12 | 2021-04-12 | Office swivel chair punching position and connection point identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113327282A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117788472A (en) * | 2024-02-27 | 2024-03-29 | 南京航空航天大学 | Method for judging corrosion degree of rivet on surface of aircraft skin based on DBSCAN algorithm |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593855A (en) * | 2013-12-04 | 2014-02-19 | 西安电子科技大学 | Clustered image splitting method based on particle swarm optimization and spatial distance measurement |
| CN105844278A (en) * | 2016-04-15 | 2016-08-10 | 浙江理工大学 | Multi-feature fused fabric scanning pattern recognition method |
| CN106296709A (en) * | 2016-08-18 | 2017-01-04 | 北京航空航天大学 | A kind of based on population and the cell image segmentation method of fuzzy means clustering |
| CN109274948A (en) * | 2017-07-17 | 2019-01-25 | 深圳市道通智能航空技术有限公司 | Image color correction method, apparatus, storage medium and computer equipment |
| CN110119747A (en) * | 2019-05-14 | 2019-08-13 | 安徽理工大学 | A kind of coal rock detection method based on radioscopic image |
| CN111476303A (en) * | 2020-04-09 | 2020-07-31 | 国网河北省电力有限公司电力科学研究院 | Line loss analysis method of fuzzy C-means clustering based on quantum optimization particle swarm |
-
2021
- 2021-04-12 CN CN202110389543.7A patent/CN113327282A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593855A (en) * | 2013-12-04 | 2014-02-19 | 西安电子科技大学 | Clustered image splitting method based on particle swarm optimization and spatial distance measurement |
| CN105844278A (en) * | 2016-04-15 | 2016-08-10 | 浙江理工大学 | Multi-feature fused fabric scanning pattern recognition method |
| CN106296709A (en) * | 2016-08-18 | 2017-01-04 | 北京航空航天大学 | A kind of based on population and the cell image segmentation method of fuzzy means clustering |
| CN109274948A (en) * | 2017-07-17 | 2019-01-25 | 深圳市道通智能航空技术有限公司 | Image color correction method, apparatus, storage medium and computer equipment |
| CN110119747A (en) * | 2019-05-14 | 2019-08-13 | 安徽理工大学 | A kind of coal rock detection method based on radioscopic image |
| CN111476303A (en) * | 2020-04-09 | 2020-07-31 | 国网河北省电力有限公司电力科学研究院 | Line loss analysis method of fuzzy C-means clustering based on quantum optimization particle swarm |
Non-Patent Citations (1)
| Title |
|---|
| 杨懿: "车身焊点质量检测机器人视觉定位算法研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117788472A (en) * | 2024-02-27 | 2024-03-29 | 南京航空航天大学 | Method for judging corrosion degree of rivet on surface of aircraft skin based on DBSCAN algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111680542B (en) | Steel coil point cloud identification and classification method based on multi-scale feature extraction and Pointnet neural network | |
| CN108171184B (en) | Method for re-identifying pedestrians based on Simese network | |
| CN110222767B (en) | Three-dimensional point cloud classification method based on nested neural network and grid map | |
| CN110929713B (en) | Steel seal character recognition method based on BP neural network | |
| CN113470090A (en) | Multi-solid-state laser radar external reference calibration method based on SIFT-SHOT characteristics | |
| CN112036384B (en) | Sperm head shape recognition method, device and equipment | |
| CN111915418A (en) | Internet financial fraud online detection method and device | |
| CN113012098B (en) | Iron tower angle steel punching defect detection method based on BP neural network | |
| CN114926410A (en) | Method for detecting appearance defects of brake disc | |
| CN110866547B (en) | Automatic classification system and method for traditional Chinese medicine decoction pieces based on multiple features and random forests | |
| CN115294109A (en) | Real wood board production defect identification system based on artificial intelligence, and electronic equipment | |
| Solak et al. | A new method for classifying nuts using image processing and k‐means++ clustering | |
| CN108520539B (en) | Image target detection method based on sparse learning variable model | |
| CN113327282A (en) | Office swivel chair punching position and connection point identification method | |
| CN109583307A (en) | A kind of Cashmere and Woolens fiber recognition method based on local feature Yu word packet model | |
| CN101561875B (en) | Method for positioning two-dimensional face images | |
| CN102930291A (en) | Automatic K adjacent local search heredity clustering method for graphic image | |
| CN116704270A (en) | Intelligent equipment positioning marking method based on image processing | |
| CN109829511B (en) | Texture classification-based method for detecting cloud layer area in downward-looking infrared image | |
| CN109815887B (en) | Multi-agent cooperation-based face image classification method under complex illumination | |
| CN114863189B (en) | Intelligent image identification method based on big data | |
| CN113627531B (en) | Method for determining pear ring rot resistance based on support vector machine classification algorithm | |
| CN115690401A (en) | Improved rotating frame target detection method fusing CIOU adaptive sample allocation | |
| CN114757916A (en) | Industrial CT image defect classification method based on feature extraction and BP network | |
| CN115272305A (en) | Button hole defect detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210831 |