CN110852989A - Quality flaw detection of tile photographed picture - Google Patents
Quality flaw detection of tile photographed picture Download PDFInfo
- Publication number
- CN110852989A CN110852989A CN201910945013.9A CN201910945013A CN110852989A CN 110852989 A CN110852989 A CN 110852989A CN 201910945013 A CN201910945013 A CN 201910945013A CN 110852989 A CN110852989 A CN 110852989A
- Authority
- CN
- China
- Prior art keywords
- image
- tile
- sample image
- detected
- unqualified
- 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
Images
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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a quality flaw detection method of a tile photographed picture, which comprises the following steps: s1, collecting an image of a tile to be detected, and matching the image of the detected tile with an image library; s2, calling a corresponding cutting template according to the sample image to cut the image of the detected ceramic tile to obtain a detected image module, wherein the cutting template is a cutting mode for cutting the sample image according to color distribution; s3, extracting a characteristic value of the image detection module, comparing the extracted characteristic value with a characteristic value of the sample image module in a preset comparison mode, and judging whether the detected ceramic tile is qualified or not according to a comparison difference value; and S4, collecting unqualified tile image data, and updating the data of the comparison mode adopted in the S4 according to the unqualified tile image data. The method and the device divide the image of the ceramic tile according to the color distribution, so that the color approximate area is divided in the same module, the accuracy of the characteristic value is improved, and the qualification rate of the product is improved.
Description
Technical Field
The invention relates to the technical field of tile detection, in particular to a quality flaw detection method for a tile photographed picture.
Background
The surface color of the ceramic tile mainly aims at the fact that the deviation of the surface color of the ceramic tile and the surface color of a standard ceramic tile is caused by the reasons of raw material change, the change of a firing system, different polishing depths and the like in the processing process of the ceramic tile, the flaws of the ceramic tile on the color are caused, flaw detection is carried out on the surface color of the ceramic tile, and unqualified products can be screened out. The existing tile color difference detection method is limited by diversification of tile color patterns, and generally adopts the method of directly holding standard colors to carry out manual judgment or uniformly dividing images after the images are collected by a camera to carry out color difference detection. The first method can achieve the detection purpose, but human eyes are fatigued, cannot always align with the color of the tile in the production process, and the manual detection efficiency and accuracy are low. The second approach solves the problem of human eye fatigue, but due to the variety of tiles, detection of color distribution by uniformly dividing tiles can have large differences. At present, a lot of tile color difference detection methods based on vision exist, but when color feature information of a tile is counted, the interference features of the edge field pixel level cannot be distinguished, and in addition, when feature values are compared, a set threshold value is fixed and unchanged, so that the qualified rate of the tile is low.
Disclosure of Invention
The invention aims at least one defect in the prior art, and provides a quality defect detection method for a tile photographed picture.
The technical scheme adopted by the invention is as follows:
the quality flaw detection method for the tile photographed picture comprises the following steps:
s1, collecting an image of a tile to be detected, matching the image of the detected tile with an image library, and displaying a sample image matched with the image of the detected tile;
s2, calling a corresponding cutting template according to the sample image to cut the image of the detected ceramic tile to obtain a detected image module, wherein the cutting template is a cutting mode for cutting the sample image according to color distribution;
s3, extracting a characteristic value of the image detection module, comparing the extracted characteristic value with a characteristic value of the sample image module in a preset comparison mode, and judging whether the detected ceramic tile is qualified or not according to a comparison difference value;
and S4, collecting image data of unqualified tiles, and updating the data of the comparison mode adopted in the S4 according to the unqualified tile image data.
The method cuts the tile image to be detected by adopting a cutting mode of cutting the image according to color distribution, so that the area with similar color is conveniently cut into the same module, and the characteristic value of the subsequently extracted and detected image module is more accurate.
Further, the comparison in step S3 is to compare the comparison difference with a preset threshold, where the comparison difference is smaller than the threshold and is qualified, and the comparison difference is larger than the threshold and is unqualified.
Further, the step S4 includes marking, counting and sorting the feature values of the unqualified detected image modules. The characteristic values of the unqualified detection image modules are marked, counted and sequenced, so that the detection image modules of the unqualified ceramic tiles can be more intuitively known to have higher unqualified rate.
Further, the comparison mode also comprises the step of judging each characteristic value according to the sorting order, and when one characteristic value is judged to be unqualified, the judgment is directly judged to be unqualified. The characteristic values of the corresponding detection image modules are detected according to the sequence of the unqualified detection image modules, and when a certain characteristic value is judged to be unqualified, the characteristic value is directly judged to be unqualified, so that the detection rate and efficiency are increased to a certain extent.
Further, the updating is to adjust the threshold value of the feature value comparison and adjust the sorting size of the feature values according to the collected image data. And the threshold value for comparing the characteristic values is adjusted according to the collected image data, so that the qualification rate and the quality of the product can be improved.
Further, the step of forming the image library in step S1 is:
s11, collecting an image of a standard ceramic tile as a sample image;
s12, cutting the sample image according to the color distribution of the ceramic tile to form a sample image module;
s13, taking the cutting mode of each sample image as a cutting template of the sample image;
s14, extracting a characteristic value of each sample image module;
and S15, storing the sample image, the sample image module, the cutting template and the characteristic value.
Further, the shape of the sample image module in step S12 differs according to the region of the color distribution.
Further, the step S12 includes preprocessing the sample image module. The sample image module is preprocessed, so that the noise of the image can be reduced, and the robustness of the image is improved.
Further, the storage manner in step S15 is centralized storage, that is, the sample image modules and the cutting manner of the same sample image form an association relationship with the sample image, and each feature value forms an association relationship with the corresponding sample image module. The sample image module and the characteristic value are stored in a centralized manner, and the same ceramic tile is associated, so that centralized management is facilitated, and the utilization rate of a storage space is improved.
Further, the step S4 includes generating a detection report according to the collected unqualified tile image data, and uploading and storing the detection report. And a detection report is generated according to the unqualified tile image data, so that a producer can more intuitively know the problem of the unqualified tile.
Further, both the sample image and the detected tile image are acquired in a particular environment. Therefore, images shot by all tiles are ensured to be in the same luminous environment, the shot images are ensured to be uniform, and errors among the images are reduced.
Compared with the prior art, the invention has the beneficial effects that:
the method and the device divide the standard tile image according to the color distribution to form the dividing template, divide the tile image to be detected according to the dividing template, ensure that the detection image module and the sample image module have the same dividing area, divide the image according to the color distribution, facilitate the cutting of the area with approximate color into the same module, and improve the accuracy of the characteristic value of the image module.
Collecting unqualified tile images, dynamically adjusting the threshold value for comparing the characteristic values according to the collected tile image data, and improving the qualification rate and quality of products.
And the image detection modules are detected according to the sequence of the unqualified characteristic values, so that the image detection efficiency is improved.
Drawings
Fig. 1 is a flow chart of the detection of the tile image to be detected according to the present invention.
FIG. 2 is a flow chart of sample image formation of the present invention.
Detailed Description
The drawings are only for purposes of illustration and are not to be construed as limiting the invention. For a better understanding of the following embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
Example 1
As shown in fig. 1, the present embodiment provides a method for detecting quality defects of a photographed tile picture, including:
s1, collecting an image of a tile to be detected, matching the image of the detected tile with an image library, and displaying a sample image matched with the image of the detected tile;
s2, calling a corresponding cutting template according to the sample image to cut the image of the detected ceramic tile to obtain a detected image module, wherein the cutting template is a cutting mode for cutting the sample image according to color distribution;
s3, extracting a characteristic value of the image detection module, comparing the extracted characteristic value with a characteristic value of the sample image module in a preset comparison mode, and judging whether the detected ceramic tile is qualified or not according to a comparison difference value;
and S4, collecting image data of unqualified tiles, and updating the data of the comparison mode adopted in the S4 according to the unqualified tile image data.
When the ceramic tile needs to be detected, a camera is used for shooting the ceramic tile, the shot image is uploaded to a system, the system matches the image library according to the image of the detected ceramic tile, the sample image with the highest matching rate with the detected ceramic tile image is displayed, a cutting template corresponding to the sample image is called to cut the image of the detected ceramic tile to obtain a detected image module, and the cutting template is a cutting mode for cutting the sample image according to color distribution. And then extracting the characteristic value of the detection image module, comparing the characteristic value of the detection image module with the characteristic value of the sample image, and judging whether the ceramic tile is qualified or not according to the comparison result. And storing the image data of the unqualified ceramic tiles into a database, and adjusting the preset data according to the unqualified ceramic tile image data. Through the implementation mode, in the process of detecting the quality flaws of the ceramic tiles, the similar colors can be cut into the same area by adopting the cutting mode of cutting the images according to the color distribution, so that the accuracy of characteristic value extraction is improved, and the qualification rate of products is improved.
In this embodiment, the comparison in step S3 is performed by comparing the comparison difference with a preset threshold, where the comparison difference is smaller than the threshold and is qualified, and the comparison difference is larger than the threshold and is unqualified.
In this embodiment, the step S4 further includes marking, counting, and sorting the feature values of the unqualified detected image modules. The characteristic values of the unqualified detection image modules are marked, counted and sequenced, so that the detection image modules of the unqualified ceramic tiles can be more intuitively known to have higher unqualified rate.
In this embodiment, the comparing method further includes determining each feature value according to the sorting order, and directly determining that a certain feature value is not qualified when the certain feature value is determined to be not qualified. The characteristic values of the corresponding detection image modules are detected according to the sequence of the unqualified detection image modules, and when a certain characteristic value is judged to be unqualified, the characteristic value is directly judged to be unqualified, so that the detection rate and efficiency are increased to a certain extent.
In this embodiment, the updating is to adjust the threshold of the eigenvalue comparison and adjust the sort size of the eigenvalue according to the collected image data. And the threshold value for comparing the characteristic values is adjusted according to the collected image data, so that the qualification rate and the quality of the product can be improved.
In this embodiment, the step of forming the image library in step S1 is:
s11, collecting an image of a standard ceramic tile as a sample image;
s12, cutting the sample image according to the color distribution of the ceramic tile to form a sample image module;
s13, taking the cutting mode of each sample image as a cutting template of the sample image;
s14, extracting a characteristic value of each sample image module;
and S15, storing the sample image, the sample image module, the cutting template and the characteristic value.
Specifically, as shown in fig. 2, which is a flowchart of sample image formation, a standard tile is photographed and an image is uploaded to a system as a sample image, the system cuts the sample image according to color distribution of the standard tile to obtain one or more sample image modules in different areas and extract a feature value of each sample image module, a cutting mode for cutting the sample image according to the color distribution is used as a cutting template, and the sample image, the sample image modules, the cutting template, and the feature value are stored in a database.
In this embodiment, the shape of the sample image module in step S12 is different depending on the region of the color distribution.
In this embodiment, the step S12 further includes preprocessing the sample image module. Specifically, in the process of preprocessing the sample image, the sample image is enhanced firstly, then image filtering and binarization processing are performed, and the sample image module is preprocessed, so that the noise of the image can be reduced, and the robustness of the image can be improved.
In this embodiment, the storage manner in step S15 is centralized storage, that is, the sample image modules and the cutting manner of the same sample image form an association relationship with the sample image, and each feature value forms an association relationship with the corresponding sample image module. Specifically, sample image modules of the same tile are stored in a project file, names of all sample image modules of the same tile are associated with sample images according to the naming of all the sample image modules, characteristic values are stored in files of corresponding sample image modules and named, and the names of the sample image modules are associated with the strokes of the sample image modules. The sample image module and the characteristic value are stored in a centralized manner, and the same ceramic tile is associated, so that centralized management is facilitated, and the utilization rate of a storage space is improved.
In this embodiment, the step S4 further includes generating a detection report according to the collected unqualified tile image data, and uploading and storing the detection report. And a detection report is generated according to the unqualified tile image data, so that a producer can more intuitively know the problem of the unqualified tile.
In this embodiment, both the sample image and the detected tile image are acquired in a particular environment. Therefore, images shot by all tiles are ensured to be in the same luminous environment, the shot images are ensured to be uniform, and errors among the images are reduced.
In the specific implementation process of the embodiment, when a detection person uses the system to detect a tile for the first time, an image of a standard tile needs to be recorded as a sample image, the sample image is obtained by shooting the standard tile in a characteristic environment, and after the sample image is recorded into the system, the system divides the sample image according to the color distribution of the sample image to obtain sample image modules with different areas, so that color approximate areas can be conveniently cut into the same sample image module, and the subsequently extracted characteristic value is more accurate. And then, preprocessing the sample image module, including image enhancement, image filtering and image binarization processing, extracting the characteristic values of the preprocessed sample image module, and presetting a corresponding threshold value according to each characteristic value. And (3) storing the sample image modules and the characteristic values thereof in a centralized manner, namely storing the sample image modules and the characteristic values of the same tile in the same project file, namely forming an association relationship between the sample image modules and the cutting mode of the same sample image and the sample image, and forming an association relationship between each characteristic value and the corresponding sample image module. The sample image module and the characteristic value are stored in a centralized manner, and the same ceramic tile is associated, so that centralized management is facilitated, and the utilization rate of a storage space is improved. And storing the cutting mode of cutting the sample image according to the color distribution in a project file corresponding to the standard ceramic tile. When a detection person detects the ceramic tile, the ceramic tile to be detected is shot in the same specific environment as the sample image, the shot detection ceramic tile image is uploaded to the system, the system performs primary matching on the detection ceramic tile image and the sample image library, and the sample image with the highest matching rate with the detection ceramic tile image is displayed. And then calling a cutting template corresponding to the sample image to cut the detected tile image to obtain a detected image module, preprocessing the detected image module, extracting a characteristic value of the preprocessed detected image module, comparing the characteristic value of the detected image module with the characteristic value of the corresponding sample image module, determining that the detected tile image module is qualified if the comparison difference is smaller than a preset threshold value, and determining that the detected tile image module is unqualified if the comparison difference is larger than the preset threshold value. And finally, collecting data of the ceramic tiles detected to be unqualified to form a dynamic collection process: dynamically adjusting the threshold value of the characteristic value comparison according to the collected data; the method comprises the steps of marking and counting unqualified characteristic values, sequencing unqualified detection image modules according to a counting value, and sequencing judgment of the characteristic values by a system when detecting the detection tiles when accumulating to a certain value, wherein the sequencing can be single value sequencing, parallel sequencing for equal counting values, or sequencing according to a range value, namely parallel sequencing is carried out on the characteristic values of the counting value in a certain range, and parallel judgment is carried out during judgment. And judging the characteristic values in sequence, directly judging the characteristic values as unqualified when judging that one characteristic value is unqualified, finishing the whole judging process, and judging the detected ceramic tile as qualified when all the characteristic values are known to be qualified. And finally, generating a detection report of the same kind of ceramic tiles according to the collected unqualified ceramic tile image data, and generating the detection report according to the unqualified ceramic tile image data, so that a producer can more intuitively know the problem of the unqualified ceramic tiles.
Claims (10)
1. A quality flaw detection method for a tile photographed picture is characterized by comprising the following steps:
s1, collecting an image of a tile to be detected, matching the image of the detected tile with an image library, and displaying a sample image matched with the image of the detected tile;
s2, calling a corresponding cutting template according to the sample image to cut the image of the detected ceramic tile to obtain a detected image module, wherein the cutting template is a cutting mode for cutting the sample image according to color distribution;
s3, extracting a characteristic value of the image detection module, comparing the extracted characteristic value with a characteristic value of the sample image module in a preset comparison mode, and judging whether the detected ceramic tile is qualified or not according to a comparison difference value;
and S4, collecting image data of unqualified tiles, and updating the data of the comparison mode adopted in the S4 according to the unqualified tile image data.
2. The method as claimed in claim 1, wherein the comparison in step S3 is performed by comparing the comparison difference with a predetermined threshold, and the comparison difference is smaller than the threshold, the comparison result is qualified, and the comparison difference is greater than the threshold, the comparison result is unqualified.
3. The method for detecting the quality defect of the photographed picture of the ceramic tile as claimed in claim 1, wherein said step S4 further comprises marking, counting and sorting the feature values of the defective image modules.
4. The method for detecting the quality defect of the photographed picture of the ceramic tile as claimed in claim 2, wherein the comparison method further comprises the steps of judging the characteristic values in a row order, and directly judging that one of the characteristic values is unqualified when the characteristic value is judged to be unqualified.
5. The method as claimed in any one of claims 1 to 4, wherein the updating is to adjust the threshold of the eigenvalue comparison and adjust the rank of the eigenvalue according to the collected image data.
6. The method for detecting the quality defect of the photographed picture of the ceramic tile as claimed in claim 1, wherein the step of forming the image library in the step S1 is:
s11, collecting an image of a standard ceramic tile as a sample image;
s12, cutting the sample image according to the color distribution of the ceramic tile to form a sample image module;
s13, taking the cutting mode of each sample image as a cutting template of the sample image;
s14, extracting a characteristic value of each sample image module;
and S15, storing the sample image, the sample image module, the cutting template and the characteristic value.
7. The method as claimed in claim 6, wherein the shape of the pattern image block in step S12 is different according to the color distribution.
8. The method as claimed in claim 6, wherein the step S12 further comprises preprocessing the sample image block.
9. The method as claimed in claim 6, wherein the storage manner in step S15 is centralized storage, that is, sample image blocks and cutting manners of the same sample image are associated with the sample image, and each feature value is associated with the corresponding sample image block.
10. The method for detecting the quality defect of the photographed tile picture according to claim 1, wherein said step S4 further comprises generating a detection report according to the collected unqualified tile image data, and uploading and storing the detection report.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910945013.9A CN110852989B (en) | 2019-09-30 | 2019-09-30 | Quality flaw detection of tile photographed picture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910945013.9A CN110852989B (en) | 2019-09-30 | 2019-09-30 | Quality flaw detection of tile photographed picture |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110852989A true CN110852989A (en) | 2020-02-28 |
| CN110852989B CN110852989B (en) | 2022-05-13 |
Family
ID=69596263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910945013.9A Active CN110852989B (en) | 2019-09-30 | 2019-09-30 | Quality flaw detection of tile photographed picture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110852989B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111942034A (en) * | 2020-08-26 | 2020-11-17 | 佛山欧神诺陶瓷有限公司 | Inkjet printing apparatus and inkjet printing control method |
| CN113533332A (en) * | 2021-08-10 | 2021-10-22 | 江西离子型稀土工程技术研究有限公司 | Praseodymium-neodymium alloy qualitative quality inspection method based on section image characteristics |
| CN114841915A (en) * | 2022-03-14 | 2022-08-02 | 阿里巴巴（中国）有限公司 | Tile flaw detection method and system based on artificial intelligence and storage medium |
| CN117274249A (en) * | 2023-11-20 | 2023-12-22 | 江西省中鼐科技服务有限公司 | Ceramic tile appearance detection method and system based on artificial intelligent image technology |
| CN118294464A (en) * | 2024-05-31 | 2024-07-05 | 圣奥科技股份有限公司 | Plank detection system and method for furniture production |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103793903A (en) * | 2012-10-29 | 2014-05-14 | 方正国际软件(北京)有限公司 | Image segmentation method and system utilizing barcode identification |
| CN106872487A (en) * | 2017-04-21 | 2017-06-20 | 佛山市南海区广工大数控装备协同创新研究院 | The surface flaw detecting method and device of a kind of view-based access control model |
| CN106952257A (en) * | 2017-03-21 | 2017-07-14 | 南京大学 | A kind of curved surface label open defect detection method based on template matches and Similarity Measure |
| CN107633507A (en) * | 2017-09-02 | 2018-01-26 | 南京理工大学 | LCD defect inspection methods based on contour detecting and characteristic matching |
| US20180089533A1 (en) * | 2016-09-27 | 2018-03-29 | Abbyy Development Llc | Automated methods and systems for locating document subimages in images to facilitate extraction of information from the located document subimages |
| CN107966444A (en) * | 2017-10-12 | 2018-04-27 | 常州信息职业技术学院 | Textile flaw detection method based on template |
| CN108827974A (en) * | 2018-06-28 | 2018-11-16 | 广东科达洁能股份有限公司 | A kind of ceramic tile defect inspection method and system |
| CN109523505A (en) * | 2018-09-18 | 2019-03-26 | 深圳市智信精密仪器有限公司 | A kind of ceramic brick surface patterned defect detection method based on machine vision |
| CN109829496A (en) * | 2019-01-30 | 2019-05-31 | 广州市载道信息科技有限公司 | A kind of physical measurement classification method and equipment |
-
2019
- 2019-09-30 CN CN201910945013.9A patent/CN110852989B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103793903A (en) * | 2012-10-29 | 2014-05-14 | 方正国际软件(北京)有限公司 | Image segmentation method and system utilizing barcode identification |
| US20180089533A1 (en) * | 2016-09-27 | 2018-03-29 | Abbyy Development Llc | Automated methods and systems for locating document subimages in images to facilitate extraction of information from the located document subimages |
| CN106952257A (en) * | 2017-03-21 | 2017-07-14 | 南京大学 | A kind of curved surface label open defect detection method based on template matches and Similarity Measure |
| CN106872487A (en) * | 2017-04-21 | 2017-06-20 | 佛山市南海区广工大数控装备协同创新研究院 | The surface flaw detecting method and device of a kind of view-based access control model |
| CN107633507A (en) * | 2017-09-02 | 2018-01-26 | 南京理工大学 | LCD defect inspection methods based on contour detecting and characteristic matching |
| CN107966444A (en) * | 2017-10-12 | 2018-04-27 | 常州信息职业技术学院 | Textile flaw detection method based on template |
| CN108827974A (en) * | 2018-06-28 | 2018-11-16 | 广东科达洁能股份有限公司 | A kind of ceramic tile defect inspection method and system |
| CN109523505A (en) * | 2018-09-18 | 2019-03-26 | 深圳市智信精密仪器有限公司 | A kind of ceramic brick surface patterned defect detection method based on machine vision |
| CN109829496A (en) * | 2019-01-30 | 2019-05-31 | 广州市载道信息科技有限公司 | A kind of physical measurement classification method and equipment |
Non-Patent Citations (5)
| Title |
|---|
| CB YU等: "Finger-vein image recognition combining modified hausdorff distance with minutiae feature matching", 《JOURNAL OF BIOMEDICAL SCIENCE & ENGINEERING》 * |
| 孟祥萍等: "指纹纹线的提取和匹配", 《计算机工程与应用》 * |
| 崔玲玲等: "融合双特征图信息的图像显著性检测方法", 《中国图像图形学报》 * |
| 王明超: "瓷砖表面瑕疵检测方法的研究", 《中国优秀硕士学位论文全文数据库 (信息科技辑)》 * |
| 马旭: "数字医疗图像特征分割常用方法比较研究", 《辽宁大学学报（自然科学版）》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111942034A (en) * | 2020-08-26 | 2020-11-17 | 佛山欧神诺陶瓷有限公司 | Inkjet printing apparatus and inkjet printing control method |
| CN111942034B (en) * | 2020-08-26 | 2021-12-03 | 佛山欧神诺陶瓷有限公司 | Inkjet printing apparatus and inkjet printing control method |
| CN113533332A (en) * | 2021-08-10 | 2021-10-22 | 江西离子型稀土工程技术研究有限公司 | Praseodymium-neodymium alloy qualitative quality inspection method based on section image characteristics |
| CN114841915A (en) * | 2022-03-14 | 2022-08-02 | 阿里巴巴（中国）有限公司 | Tile flaw detection method and system based on artificial intelligence and storage medium |
| CN117274249A (en) * | 2023-11-20 | 2023-12-22 | 江西省中鼐科技服务有限公司 | Ceramic tile appearance detection method and system based on artificial intelligent image technology |
| CN117274249B (en) * | 2023-11-20 | 2024-03-01 | 江西省中鼐科技服务有限公司 | Ceramic tile appearance detection method and system based on artificial intelligent image technology |
| CN118294464A (en) * | 2024-05-31 | 2024-07-05 | 圣奥科技股份有限公司 | Plank detection system and method for furniture production |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110852989B (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110852989B (en) | Quality flaw detection of tile photographed picture | |
| CN109872300B (en) | Visual saliency detection method for appearance defects of friction plate | |
| CN112036755B (en) | Supervision method and system for quality detection of building engineering | |
| CN116559183B (en) | Method and system for improving defect judging efficiency | |
| CN104537651B (en) | Proportion detecting method and system for cracks in road surface image | |
| CN116645367B (en) | Steel plate cutting quality detection method for high-end manufacturing | |
| CN103914708A (en) | Food variety detection method and system based on machine vision | |
| CN116703911B (en) | LED lamp production quality detecting system | |
| CN113935666B (en) | Building decoration wall tile abnormity evaluation method based on image processing | |
| CN107490583A (en) | A kind of intermediate plate defect inspection method based on machine vision | |
| CN109215009A (en) | Continuous casting billet surface image defect inspection method based on depth convolutional neural networks | |
| CN106780464A (en) | A kind of fabric defect detection method based on improvement Threshold segmentation | |
| CN110554052A (en) | artificial board surface defect detection method and system | |
| CN104749801B (en) | High Precision Automatic optical detecting method and system | |
| CN116402742A (en) | Visual detection method and system for surface defects of automobile sheet metal part | |
| CN210775265U (en) | Artificial board surface defect detection system | |
| CN107220946B (en) | Real-time removing method for bad block degree images on rock conveyer belt | |
| CN114049340A (en) | Diamond saw wire abrasion on-line detection method and device based on machine vision | |
| CN116703834B (en) | Method and device for judging and grading excessive sintering ignition intensity based on machine vision | |
| CN117237747A (en) | Hardware defect classification and identification method based on artificial intelligence | |
| CN110458231B (en) | Ceramic product detection method, device and equipment | |
| CN109544514B (en) | Sawn timber identity identification method, device and equipment integrating apparent features | |
| CN110516725A (en) | The detection method of plank fringe spacing and color based on machine vision | |
| CN116309564A (en) | Method and system for detecting appearance defects of battery cells based on artificial intelligent image recognition | |
| CN110866917A (en) | Tablet type and arrangement mode identification method based on machine vision |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |