Industrial quality monitoring system based on machine vision
Through machine vision -based industrial quality monitoring systems, real -time collection and matching workpiece images, constructing and comparing standard workpieces models, solving inaccurate problems in the quality monitoring of workpieces, and achieving efficient and accurate workpiece quality monitoring.
Patent Information
- Application Number
- CN202411779754.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-05
- Publication Date
- 2025-01-21
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The prior art has problems with inaccurate workpiece quality monitoring, especially when the workpiece shape is abnormal, it is difficult to react quickly and accurately, resulting in insufficient accuracy of workpiece quality monitoring and cannot meet the quality requirements of industrial processing.
Using the industrial quality monitoring system based on machine vision, obtain workpiece type and standard workpiece through the first collection module, collect standard workpiece feature information for the second collection module, build a standard workpiece model, divide the module to designate management area, extract feature information, extract the module, extract the module, extract the module, extract the module Real -time collecting workpiece images and matching standard workpieces models, judging the module based on comparison information to determine the qualifications of the workpiece and prompts unqualified workpieces.
It improves the efficiency of the quality of workpieces, enhances the accuracy and rapid response capacity of the quality monitoring of the workpiece, can accurately monitor the unusually small position in the workpiece, and meet the high requirements of the quality of the workpiece in industrial processing.
Smart Images
Figure CN119338327A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of industrial quality, and in particular to an industrial quality monitoring system based on machine vision. Background Art
[0002] At present, in industrial processing, the quality of workpieces is still an issue that people need to pay attention to. At present, the monitoring of workpiece quality problems will also be carried out by visual images. However, the current means of monitoring workpiece quality through visual images has the problem of inaccuracy. When there are abnormal problems in the shape of part of the workpiece, it is impossible to respond quickly and accurately, and it is difficult to monitor the abnormality of the workpiece. There will still be small abnormalities in the workpiece, which makes it difficult to achieve the accuracy of workpiece quality monitoring and cannot meet the quality requirements of industrial processing workpieces. Summary of the invention
[0003] The purpose of the present invention is to provide an industrial quality monitoring system based on machine vision to solve the shortcomings of the background technology.
[0004] In order to achieve the above object, the present invention provides the following technical solution: an industrial quality monitoring system based on machine vision, comprising: A first acquisition module is used to acquire workpiece types and obtain corresponding standard workpieces based on the workpiece types; A second acquisition module is connected to the first acquisition module and is used to acquire standard workpiece feature information based on the workpiece type, wherein the standard workpiece feature information includes workpiece size and corresponding workpiece shape information, and to construct a standard workpiece model based on the standard workpiece feature information; A division module, connected to the second acquisition module, is used to define a management area for the standard workpiece, extract characteristic information of the management area, bind the characteristic information with the workpiece type of the standard workpiece as comparison information in the standard workpiece model and store it in a database; An extraction module, connected to the division module, is used to collect workpiece images in the industrial processing process in real time based on machine vision, and determine a corresponding standard workpiece model based on the workpiece type corresponding to the workpiece image; The judgment module is connected to the extraction module and is used to judge the eligibility of workpieces in the industrial processing process based on the comparison information and to prompt unqualified workpieces.
[0005] In a preferred embodiment, the first acquisition module includes: A type acquisition unit is used to collect the types of workpieces processed in industrial processing, and number the workpiece types to obtain workpiece type codes; The regularization unit is used to manage standard workpieces of different sizes based on the workpiece type code, and uses the workpiece type code and the corresponding standard workpieces of different sizes as workpiece class information.
[0006] In a preferred embodiment, the second acquisition module includes: A feature acquisition unit, used for collecting workpiece dimensions and corresponding workpiece shape information corresponding to the workpiece type as standard workpiece feature information; The construction unit is used to respectively construct standard workpiece models based on the standard workpiece feature information corresponding to the standard workpieces of different sizes in the workpiece information, and establish a corresponding relationship between the standard workpiece model and the standard workpiece in the workpiece information.
[0007] In a preferred embodiment, the division module includes: A division unit, used for obtaining a monitoring surface of a standard workpiece based on preset information, and dividing a working area in the monitoring surface into management areas; An extraction unit, used to extract shape features and size information of standard workpieces in the management area, and to perform feature marking on the shape features of the workpieces based on preset conditions to obtain feature information, wherein the feature information is lines depicting the shape of the workpieces and identification points on the lines; The binding unit is used to bind the feature information with the workpiece type of the standard workpiece as comparison information and mark it in the standard workpiece model and store it in a database.
[0008] In a preferred embodiment, the division unit includes: A selection unit is used to select a surface in the standard workpiece as a base surface, and starting from the base surface, use the angle between adjacent surfaces in the standard workpiece that do not meet the preset angle as a division angle, and obtain multiple monitoring surfaces in the standard workpiece based on the division angle; The information determination unit is used to determine the working area in the monitoring surface based on the workpiece usage information and divide it into management areas.
[0009] In a preferred embodiment, the extraction unit comprises: An information extraction unit, used to extract shape features and size information of standard workpieces in the management area and mark them in the corresponding standard workpiece model; A drawing unit, used for drawing lines on the standard workpiece based on the shape characteristics and size information of the standard workpiece to obtain lines for drawing the shape of the workpiece, and setting identification points on the lines for drawing the shape of the workpiece based on preset conditions; A management unit, used for taking the lines depicting the shape of the workpiece and the identification points on the lines as feature information; The purpose of collecting standard workpieces here is to obtain a database. The feature information is bound to the workpiece type of the standard workpiece and marked as comparison information in the standard workpiece model and stored in the database. This can prepare data for quality monitoring of standard workpieces, and subsequently quickly obtain the quality of the workpiece through visual images during the industrial processing process, greatly improving the efficiency of workpiece quality resolution.
[0010] In a preferred embodiment, the extraction module comprises: A real-time acquisition unit, used to acquire images of workpieces in industrial processing in real time based on machine vision; The matching unit is used to match the workpiece type of the workpiece image in the database to obtain the corresponding standard workpiece model.
[0011] In a preferred embodiment, the judgment module includes: A monitoring unit, used for acquiring a corresponding monitoring surface in a standard workpiece model based on a workpiece image, and covering the workpiece image based on the corresponding monitoring surface in the standard workpiece model to obtain a coverage surface; A judgment unit, used for acquiring the workpiece quality of the workpiece image in the coverage surface based on the identification points in the comparison information to obtain coverage information, wherein the coverage information is the number of overlaps between the identification points and the workpiece part in the workpiece image; The prompting unit is used to treat the workpieces corresponding to the overlap number that does not meet the preset number as unqualified workpieces and to prompt the unqualified workpieces.
[0012] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention can prepare data for quality monitoring of standard workpieces, and subsequently quickly acquire the quality of workpieces through visual images during industrial processing, greatly improving the efficiency of workpiece quality resolution; by performing quality monitoring of workpieces through multiple identification points, even positions with smaller abnormalities in the workpiece can be monitored, thereby improving the accuracy of workpiece quality monitoring, and being able to quickly and accurately respond to workpiece quality monitoring, greatly improving the efficiency of workpiece quality monitoring. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0014] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION
[0015] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0016] Example 1, please refer to Figure 1 As shown, the industrial quality monitoring system based on machine vision described in this embodiment includes: A first acquisition module is used to acquire workpiece types and obtain corresponding standard workpieces based on the workpiece types; A second acquisition module is connected to the first acquisition module and is used to acquire standard workpiece feature information based on the workpiece type, wherein the standard workpiece feature information includes workpiece size and corresponding workpiece shape information, and to construct a standard workpiece model based on the standard workpiece feature information; A division module, connected to the second acquisition module, is used to define a management area for the standard workpiece, extract characteristic information of the management area, bind the characteristic information with the workpiece type of the standard workpiece as comparison information in the standard workpiece model and store it in a database; An extraction module, connected to the division module, is used to collect workpiece images in the industrial processing process in real time based on machine vision, and determine a corresponding standard workpiece model based on the workpiece type corresponding to the workpiece image; A judgment module, connected to the extraction module, is used to judge the eligibility of workpieces in the industrial processing process based on the comparison information and to prompt unqualified workpieces; Further explanation: At present, in industrial processing, the quality problem of workpieces is still an issue that people need to pay attention to. At present, the monitoring of workpiece quality problems will also be monitored by visual images. However, the current means of monitoring workpiece quality through visual images has the problem of inaccuracy. When there is an abnormality in the shape of part of the workpiece, it is impossible to respond quickly and accurately, and it is difficult to monitor the abnormality of the workpiece. There will still be small abnormalities in the workpiece, which is difficult to achieve the accuracy of workpiece quality monitoring and cannot meet the quality requirements of industrial processing workpieces. In this application, data for quality monitoring of standard workpieces can be prepared, and the quality of the workpiece can be quickly acquired through visual images in the subsequent industrial processing process, which greatly improves the efficiency of workpiece quality resolution; by monitoring the quality of the workpiece through multiple identification points, the position with smaller abnormalities in the workpiece can also be monitored, which improves the accuracy of workpiece quality monitoring, and can respond quickly and accurately to the quality monitoring of the workpiece, greatly improving the efficiency of workpiece quality monitoring; In one embodiment, the first acquisition module includes: A type acquisition unit is used to collect the types of workpieces processed in industrial processing, and number the workpiece types to obtain workpiece type codes; A regularization unit is used to manage standard workpieces of different sizes based on workpiece type codes, and use the workpiece type codes and corresponding standard workpieces of different sizes as workpiece class information; To further illustrate, firstly, the type of workpiece for industrial processing is obtained, and the corresponding workpiece type is encoded to obtain the workpiece type code. After encoding, standard workpieces of the same type but different sizes are regularized under the workpiece type code, and the workpiece type code and the corresponding standard workpieces of different sizes are used as workpiece type information. For example, the workpiece for industrial processing is a connector of equipment, and the connectors have the same shape but different sizes. Such connectors of different sizes and the same shape are the same workpiece type, and they need to be classified and managed under the same workpiece type. The types of workpieces can be managed, which is convenient for subsequent monitoring of workpieces, and can greatly improve the convenience of monitoring the quality of workpieces. In one embodiment, the second acquisition module includes: A feature acquisition unit, used for collecting workpiece dimensions and corresponding workpiece shape information corresponding to the workpiece type as standard workpiece feature information; A construction unit is used to respectively construct standard workpiece models based on the standard workpiece feature information corresponding to the standard workpieces of different sizes in the workpiece information, and establish a corresponding relationship between the standard workpiece models and the standard workpieces in the workpiece information; Further explanation: based on the workpiece type, the workpiece size and the corresponding workpiece shape information are collected and used as the standard workpiece feature information. The workpiece size and the corresponding workpiece shape information collected here are standard workpieces without quality problems. Then, standard workpiece models of standard workpieces of different sizes in the workpiece class information are constructed according to the workpiece type. After obtaining the standard workpiece models of multiple standard workpieces of different sizes, the corresponding relationship between the standard workpiece model and the standard workpiece in the workpiece class information is established, which can facilitate the subsequent monitoring efficiency of the workpiece in the industrial processing process. The corresponding standard workpiece model can be directly obtained through the corresponding relationship for the quality monitoring of the workpiece; In one embodiment, the partitioning module comprises: A division unit, used for obtaining a monitoring surface of a standard workpiece based on preset information, and dividing a working area in the monitoring surface into management areas; An extraction unit, used to extract shape features and size information of standard workpieces in the management area, and to perform feature marking on the shape features of the workpieces based on preset conditions to obtain feature information, wherein the feature information is lines depicting the shape of the workpieces and identification points on the lines; A binding unit, used for binding the feature information with the workpiece type of the standard workpiece as comparison information in the standard workpiece model and storing it in a database; In one embodiment, the dividing unit comprises: A selection unit is used to select a surface in the standard workpiece as a base surface, and starting from the base surface, use the angle between adjacent surfaces in the standard workpiece that do not meet the preset angle as a division angle, and obtain multiple monitoring surfaces in the standard workpiece based on the division angle; An information determination unit, used to determine a working area in a monitoring surface based on workpiece usage information and divide it into management areas; In one embodiment, the extraction unit comprises: An information extraction unit, used to extract shape features and size information of standard workpieces in the management area and mark them in the corresponding standard workpiece model; A drawing unit, used for drawing lines on the standard workpiece based on the shape characteristics and size information of the standard workpiece to obtain lines for drawing the shape of the workpiece, and setting identification points on the lines for drawing the shape of the workpiece based on preset conditions; A management unit, used for taking the lines depicting the shape of the workpiece and the identification points on the lines as feature information; It is further explained that since the standard workpiece has multiple faces, a face is arbitrarily selected as the base face, and face division is started based on the base face. The angle between adjacent faces in the standard workpiece that does not meet the preset angle is used as the division angle, and multiple monitoring faces in the standard workpiece are obtained based on the division angle. The multiple monitoring faces are divided here in order to make the quality monitoring of the corresponding monitoring faces more accurate during data collection, and to ensure that the workpiece quality is monitored on the corresponding monitoring faces during the subsequent real-time collection of workpiece images, thereby improving the effectiveness of workpiece quality monitoring. Afterwards, the working area in the monitoring face is divided to obtain a management area. The working area here is the main position of the workpiece that affects the use behavior of the workpiece. Afterwards, the shape characteristics and size information of the standard workpiece in the management area are obtained, and the shape of the workpiece is depicted according to the shape characteristics and size information of the standard workpiece to obtain the lines for depicting the shape of the workpiece. Based on the preset conditions, the shape of the workpiece is depicted in the workpiece shape. Identification points are set on the lines. The preset conditions here are the turning points of the lines depicting the shape of the workpiece and the equidistant division of the lines depicting the shape of the workpiece when there are no turning points of the lines depicting the shape of the workpiece. Here, the turning points of the lines depicting the shape of the workpiece and the nodes of the equidistant division of the lines depicting the shape of the workpiece are used as the set identification points. Then, the lines depicting the shape of the workpiece and the identification points on the lines are used as feature information. Then, the feature information is bound to the corresponding standard workpiece as comparison information and stored in a database. Here, the collection of standard workpieces is to obtain a database. The feature information is bound to the workpiece type of the standard workpiece as comparison information and marked in the standard workpiece model and stored in the database. The data for quality monitoring of the standard workpiece can be prepared. The quality of the workpiece can be quickly obtained through visual images in the subsequent industrial processing process, which greatly improves the efficiency of workpiece quality resolution. In one embodiment, the extraction module comprises: A real-time acquisition unit, used to acquire images of workpieces in industrial processing in real time based on machine vision; A matching unit, used for matching the workpiece type of the workpiece image in a database to obtain a corresponding standard workpiece model; Further explanation: In the process of industrial processing of workpieces, it is necessary to collect workpiece images in the industrial processing process in real time, and then match the workpiece type in the workpiece image in the database to obtain the corresponding standard workpiece model. Here, the workpiece type is first obtained, and the corresponding standard workpiece model is obtained based on the workpiece type. Then, the corresponding standard workpiece model is obtained according to the size of the workpiece image. The standard workpiece model here is a model for quality judgment of the workpiece image collected in real time, which can quickly identify the workpiece quality and has good workpiece quality monitoring efficiency; In one embodiment, the determination module includes: A monitoring unit, used for acquiring a corresponding monitoring surface in a standard workpiece model based on a workpiece image, and covering the workpiece image based on the corresponding monitoring surface in the standard workpiece model to obtain a coverage surface; A judgment unit, used for acquiring the workpiece quality of the workpiece image in the coverage surface based on the identification points in the comparison information to obtain coverage information, wherein the coverage information is the number of overlaps between the identification points and the workpiece part in the workpiece image; A prompting unit, used for treating the workpieces corresponding to the overlap number that does not meet the preset number as unqualified workpieces and giving prompts for the unqualified workpieces; It is further explained that after the workpiece image is acquired, the corresponding monitoring surface can be determined based on the shooting surface of the workpiece image corresponding to the standard workpiece model, and then the monitoring surface corresponding to the standard workpiece model is overlaid with the workpiece image to obtain the coverage surface. The coverage operation here is: the monitoring surface corresponding to the standard workpiece model is put on the collected workpiece image for overlapping coverage, and the workpiece quality of the workpiece image in the coverage surface is obtained based on the recognition points in the comparison information to obtain the coverage information, wherein the coverage information is the number of overlaps between the recognition points and the workpiece part in the workpiece image. Here, the adhesion of the workpiece image is obtained through the recognition points after the overlapping coverage. For example, the recognition points are in the standard workpiece model. If the corresponding position in the model is at the corner of a line, and the recognition point is not at the corner of the corresponding line in the workpiece image after the corresponding monitoring surface of the standard workpiece model is covered with the workpiece image, then there is no overlap, otherwise it is overlap. Here, the coverage information is used as a judgment, and the workpieces corresponding to the number of overlaps that do not meet the preset number are regarded as unqualified workpieces, and the unqualified workpieces are prompted. By performing quality monitoring of the workpiece through multiple recognition points, even the smaller abnormal positions in the workpiece can be monitored, thereby improving the accuracy of workpiece quality monitoring, and being able to respond quickly and accurately to the quality monitoring of the workpiece, greatly improving the efficiency of workpiece quality monitoring.
[0017] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. An industrial quality monitoring system based on machine vision, characterized in that: include: A first acquisition module is used to acquire workpiece types and obtain corresponding standard workpieces based on the workpiece types; A second acquisition module is connected to the first acquisition module and is used to acquire standard workpiece feature information based on the workpiece type, wherein the standard workpiece feature information includes workpiece size and corresponding workpiece shape information, and to construct a standard workpiece model based on the standard workpiece feature information; A division module, connected to the second acquisition module, is used to define a management area for the standard workpiece, extract characteristic information of the management area, bind the characteristic information with the workpiece type of the standard workpiece as comparison information in the standard workpiece model and store it in a database; An extraction module, connected to the division module, is used to collect workpiece images in the industrial processing process in real time based on machine vision, and determine a corresponding standard workpiece model based on the workpiece type corresponding to the workpiece image; The judgment module is connected to the extraction module and is used to judge the eligibility of workpieces in the industrial processing process based on the comparison information and to prompt unqualified workpieces.
2. The industrial quality monitoring system based on machine vision according to claim 1, characterized in that: The first acquisition module includes: A type acquisition unit is used to collect the types of workpieces processed in industrial processing, and number the workpiece types to obtain workpiece type codes; The regularization unit is used to manage standard workpieces of different sizes based on the workpiece type code, and uses the workpiece type code and the corresponding standard workpieces of different sizes as workpiece class information.
3. The industrial quality monitoring system based on machine vision according to claim 1, characterized in that: The second acquisition module includes: A feature acquisition unit, used for collecting workpiece dimensions and corresponding workpiece shape information corresponding to the workpiece type as standard workpiece feature information; The construction unit is used to respectively construct standard workpiece models based on the standard workpiece feature information corresponding to the standard workpieces of different sizes in the workpiece information, and establish a corresponding relationship between the standard workpiece model and the standard workpiece in the workpiece information.
4. The industrial quality monitoring system based on machine vision according to claim 1, characterized in that: The division module comprises: A division unit, used for obtaining a monitoring surface of a standard workpiece based on preset information, and dividing a working area in the monitoring surface into management areas; An extraction unit, used to extract shape features and size information of standard workpieces in the management area, and to perform feature marking on the shape features of the workpieces based on preset conditions to obtain feature information, wherein the feature information is lines depicting the shape of the workpieces and identification points on the lines; The binding unit is used to bind the feature information with the workpiece type of the standard workpiece as comparison information and mark it in the standard workpiece model and store it in a database.
5. The industrial quality monitoring system based on machine vision according to claim 4, characterized in that: The division unit comprises: A selection unit is used to select a surface in the standard workpiece as a base surface, and starting from the base surface, use the angle between adjacent surfaces in the standard workpiece that do not meet the preset angle as a division angle, and obtain multiple monitoring surfaces in the standard workpiece based on the division angle; The information determination unit is used to determine the working area in the monitoring surface based on the workpiece usage information and divide it into management areas.
6. The industrial quality monitoring system based on machine vision according to claim 4, characterized in that: The extraction unit comprises: An information extraction unit, used to extract shape features and size information of standard workpieces in the management area and mark them in the corresponding standard workpiece model; A drawing unit, used for drawing lines on the standard workpiece based on the shape characteristics and size information of the standard workpiece to obtain lines for drawing the shape of the workpiece, and setting identification points on the lines for drawing the shape of the workpiece based on preset conditions; The management unit is used to use the lines depicting the shape of the workpiece and the identification points on the lines as feature information.
7. The industrial quality monitoring system based on machine vision according to claim 1, characterized in that: The extraction module comprises: A real-time acquisition unit, used to acquire images of workpieces in industrial processing in real time based on machine vision; The matching unit is used to match the workpiece type of the workpiece image in the database to obtain the corresponding standard workpiece model.
8. The industrial quality monitoring system based on machine vision according to claim 1, characterized in that: The judging module comprises: A monitoring unit, used for acquiring a corresponding monitoring surface in a standard workpiece model based on a workpiece image, and covering the workpiece image based on the corresponding monitoring surface in the standard workpiece model to obtain a coverage surface; A judgment unit, used for acquiring the workpiece quality of the workpiece image in the coverage surface based on the identification points in the comparison information to obtain coverage information, wherein the coverage information is the number of overlaps between the identification points and the workpiece part in the workpiece image; The prompting unit is used to treat the workpieces corresponding to the overlap number that does not meet the preset number as unqualified workpieces and to prompt the unqualified workpieces.