Intelligent Industrial Inspection Method and System Based on Image Processing

Abstract

The present invention is applicable to the field of industrial detection technology, and provides an intelligent industrial detection method and system based on image processing. The present invention generates a target grayscale image by performing detection shooting and grayscale processing on the target process product; selecting a first standard point, a second standard point and multiple detection points; obtaining the point coordinate data of the first standard point, the second standard point and multiple detection points, and calculating multiple detection ratios and multiple detection angles; performing detection analysis, generating industrial detection results, and performing post-process processing. It is possible to perform detection shooting, select a first standard point, a second standard point and multiple detection points, calculate multiple detection ratios and multiple detection angles, perform detection analysis and processing on multiple detection ratios and multiple detection angles, and there is no need to perform comprehensive detection analysis on multiple features of the product, which solves the diversity and complexity of detection, and greatly reduces the amount of calculation, reduces the calculation cost, and can effectively avoid the risk of freezing and crashing.

Description

Technical Field

[0001] The present invention belongs to the field of industrial detection technology, and in particular relates to an intelligent industrial detection method and system based on image processing. Background Art

[0002] Industrial inspection based on image processing, also known as visual automatic inspection technology, is a technology that detects industrial products by taking images of them and using computers to process and analyze the images.

[0003] Industrial inspection based on image processing is often used in quality inspection of product assembly.

[0004] In the prior art, automatic visual inspection of product assembly usually requires comprehensive inspection and analysis of multiple features such as the product's edges, corners, textures, and shapes. Due to the variety of industrial products, inspection is diverse and complex, and the amount of calculation is large, requiring a large amount of computing resources, high computing costs, and prone to the risk of jamming and crashing. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide an intelligent industrial inspection method and system based on image processing, aiming to solve the technical problems existing in the prior art mentioned in the background technology.

[0006] The embodiment of the present invention is implemented as follows:

[0007] An intelligent industrial detection method based on image processing, the method specifically comprises the following steps:

[0008] Detect and photograph the target process product to obtain a target photographed image and target process data, and grayscale the target photographed image to generate a target grayscale image;

[0009] According to the target process data, selecting a first standard point, a second standard point and a plurality of detection points from the target grayscale image;

[0010] Construct a detection coordinate system, obtain the point coordinate data of the first standard point, the second standard point and multiple detection points, and calculate multiple detection ratios and multiple detection angles;

[0011] Detection and analysis are performed on the plurality of detection ratios and the plurality of detection angles to generate industrial detection results, and post-processing is performed.

[0012] As a further limitation of the technical solution of the embodiment of the present invention, the detecting and photographing of the target process product, obtaining the target photographed image and the target process data, and graying the target photographed image to generate the target grayscale image specifically includes the following steps:

[0013] Determine the current industrial process and match the target process data;

[0014] According to the target process data, periodically generate detection shooting instructions;

[0015] According to the inspection and shooting instructions, the target process product is inspected and photographed to obtain the target photographed image;

[0016] The target captured image is grayscaled to generate a target grayscale image.

[0017] As a further limitation of the technical solution of the embodiment of the present invention, the selecting the first standard point, the second standard point and the plurality of detection points from the target grayscale image according to the target process data specifically comprises the following steps:

[0018] According to the target process data, the target grayscale image is divided into a previous process area and a plurality of current process areas;

[0019] Selecting a first standard point and a second standard point from the previous process area;

[0020] A plurality of inspection points are selected from the plurality of current process areas.

[0021] As a further limitation of the technical solution of the embodiment of the present invention, the selecting the first standard point and the second standard point from the previous process area specifically includes the following steps:

[0022] Obtain standard feature data;

[0023] According to the standard feature data, feature matching is performed in the previous process area to determine a first regular pattern and a second regular pattern;

[0024] Marking the centroid of the first regular figure as a first standard point;

[0025] The centroid of the second regular figure is marked as a second standard point.

[0026] As a further limitation of the technical solution of the embodiment of the present invention, the construction of the detection coordinate system, obtaining the point coordinate data of the first standard point, the second standard point and the multiple detection points, and calculating the multiple detection ratios and the multiple detection angles specifically include the following steps:

[0027] Selecting a positioning origin from the target grayscale image;

[0028] Based on the positioning origin, construct a detection coordinate system;

[0029] Acquire point coordinate data of a first standard point, a second standard point and a plurality of detection points in the detection coordinate system;

[0030] According to the point coordinate data, multiple detection ratios and multiple detection angles are calculated.

[0031] As a further limitation of the technical solution of the embodiment of the present invention, the calculation formulas of the multiple detection ratios are:

[0032] ;

[0033] in, For the The detection ratio corresponding to the detection points is For the The coordinates of the detection points, are the coordinates of the first standard point, is the coordinate of the second standard point;

[0034] The calculation formulas for the multiple detection angles are:

[0035] ;

[0036] in, For the The detection angle corresponding to each detection point.

[0037] As a further limitation of the technical solution of the embodiment of the present invention, the detection and analysis of the multiple detection ratios and the multiple detection angles, generating industrial detection results, and performing post-process processing specifically include the following steps:

[0038] According to the target process data, matching standard inspection data;

[0039] Based on the standard detection data, a plurality of the detection ratios and a plurality of the detection angles are compared and analyzed to generate an industrial detection result;

[0040] According to the industrial test results, judging whether the test meets the standards;

[0041] When the test is up to standard, the normal conduction of the post-process is carried out;

[0042] When the detection does not meet the standards, an abnormal conduction is performed to terminate the process.

[0043] An intelligent industrial inspection system based on image processing, the system includes a detection shooting processing module, an image point selection module, a ratio angle calculation module and a detection analysis processing module, wherein:

[0044] The detection and shooting processing module is used to detect and shoot the target process product, obtain the target shot image and target process data, and grayscale the target shot image to generate a target grayscale image;

[0045] An image point selection module, used to select a first standard point, a second standard point and a plurality of detection points from the target grayscale image according to the target process data;

[0046] A ratio angle calculation module is used to construct a detection coordinate system, obtain point coordinate data of a first standard point, a second standard point and multiple detection points, and calculate multiple detection ratios and multiple detection angles;

[0047] The detection analysis and processing module is used to detect and analyze the multiple detection ratios and the multiple detection angles, generate industrial detection results, and perform post-process processing.

[0048] As a further limitation of the technical solution of the embodiment of the present invention, the detection and shooting processing module specifically includes:

[0049] A data matching unit is used to determine the current industrial process and match the target process data;

[0050] An instruction generation unit, used for periodically generating detection and shooting instructions according to the target process data;

[0051] A detection and shooting unit, used to detect and shoot the target process product according to the detection and shooting instruction, and obtain the target shot image;

[0052] The grayscale processing unit is used to perform grayscale processing on the target captured image to generate a target grayscale image.

[0053] As a further limitation of the technical solution of the embodiment of the present invention, the proportional angle calculation module specifically includes:

[0054] An origin selection unit, used for selecting a positioning origin from the target grayscale image;

[0055] A coordinate system construction unit, used to construct a detection coordinate system based on the positioning origin;

[0056] A coordinate acquisition unit, used for acquiring point coordinate data of a first standard point, a second standard point and a plurality of detection points in the detection coordinate system;

[0057] The ratio angle calculation unit is used to calculate multiple detection ratios and multiple detection angles according to the point coordinate data.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] The embodiment of the present invention generates a target grayscale image by performing detection, shooting and grayscale processing on the target process product; selecting a first standard point, a second standard point and multiple detection points; obtaining the point coordinate data of the first standard point, the second standard point and multiple detection points, and calculating multiple detection ratios and multiple detection angles; performing detection analysis, generating industrial detection results, and performing post-process processing. It is possible to perform detection shooting, select a first standard point, a second standard point and multiple detection points, calculate multiple detection ratios and multiple detection angles, perform detection analysis and processing on multiple detection ratios and multiple detection angles, and there is no need to perform comprehensive detection and analysis of multiple features on the product, which solves the problem of diversity and complexity of detection, and greatly reduces the amount of calculation, reduces the calculation cost, and can effectively avoid the risk of freezing and crashing. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A flow chart of an intelligent industrial detection method based on image processing provided by an embodiment of the present invention is shown;

[0061] Figure 2 A flowchart of detection, shooting and grayscale processing in the method provided by an embodiment of the present invention is shown;

[0062] Figure 3 A flowchart showing the selection of a first standard point, a second standard point and a plurality of detection points in the method provided by an embodiment of the present invention is shown;

[0063] Figure 4 A flowchart of selecting the first standard point and the second standard point in the method provided by an embodiment of the present invention is shown;

[0064] Figure 5 A flow chart showing calculation of multiple detection ratios and multiple detection angles in the method provided by an embodiment of the present invention is shown;

[0065] Figure 6 A flow chart showing detection, analysis and processing in the method provided by an embodiment of the present invention is shown;

[0066] Figure 7 An application architecture diagram of an intelligent industrial detection system based on image processing provided by an embodiment of the present invention is shown;

[0067] Figure 8 It shows a structural block diagram of a detection and shooting processing module in a system provided by an embodiment of the present invention;

[0068] Fig. 9 The structure block diagram of the proportional angle calculation module in the system provided by the embodiment of the present invention is shown. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0070] It is understandable that in the prior art, automatic visual inspection of product assembly usually requires comprehensive inspection and analysis of multiple features of the product, such as edges, corners, textures, shapes, etc. Due to the variety of industrial products, inspection is diverse and complex, and the amount of calculation is large, which requires a lot of computing resources, has high computing costs, and is prone to the risk of freezing and crashing.

[0071] To solve the above problems, the intelligent industrial inspection method and system based on image processing disclosed in the embodiment of the present invention, by inspecting and shooting the target process product, obtaining the target shooting image and the target process data, and graying the target shooting image to generate the target gray image; according to the target process data, from the target gray image, select the first standard point, the second standard point and multiple inspection points; construct the inspection coordinate system, obtain the point coordinate data of the first standard point, the second standard point and multiple inspection points, calculate multiple inspection ratios and multiple inspection angles; inspect and analyze the multiple inspection ratios and multiple inspection angles, generate industrial inspection results, and perform post-process processing. It is possible to perform inspection shooting, select the first standard point, the second standard point and multiple inspection points, calculate multiple inspection ratios and multiple inspection angles, inspect and analyze and process the multiple inspection ratios and multiple inspection angles, without the need to perform comprehensive inspection and analysis of multiple features on the product, solve the diversity and complexity of inspection, and greatly reduce the amount of calculation, reduce the calculation cost, and effectively avoid the risk of jamming and crash.

[0072] Specifically, Figure 1 A flow chart of an intelligent industrial detection method based on image processing provided by an embodiment of the present invention is shown.

[0073] In a preferred embodiment of the present invention, an intelligent industrial detection method based on image processing comprises the following steps:

[0074] Step S101: inspect and photograph a target process product to obtain a target photographed image and target process data, and grayscale the target photographed image to generate a target grayscale image.

[0075] In an embodiment of the present invention, during the assembly production process of the target industrial product, the current industrial process is determined, the target process data corresponding to the current industrial process is matched, and the working cycle of the process equipment is determined by identifying the target process data. Then, according to the working cycle, detection and shooting instructions are periodically generated, and then according to the detection and shooting instructions, the target process product after the current industrial process is completed is inspected and shot, and a target shot image is obtained, and then the target shot image is grayed to generate a target grayscale image.

[0076] It is understandable that the target process data records the processing parts of the current industrial process and the corresponding processing technology, process equipment and other data.

[0077] It can be understood that the target process product is a temporary name for the target industrial product under the current industrial process.

[0078] Specifically, Figure 2 The flowchart of the detection, shooting and grayscale processing in the method provided by the embodiment of the present invention is shown.

[0079] Among them, in another preferred embodiment provided by the present invention, the detection and shooting of the target process product, obtaining the target shot image and the target process data, and graying the target shot image to generate the target grayscale image specifically includes the following steps:

[0080] Step S1011, determine the current industrial process and match the target process data.

[0081] Step S1012: periodically generate detection and shooting instructions according to the target process data.

[0082] Step S1013: perform inspection and photography of the target process product according to the inspection and photography instruction to obtain a target photographed image.

[0083] Step S1014: grayscale the target captured image to generate a target grayscale image.

[0084] Furthermore, the intelligent industrial detection method based on image processing also includes the following steps:

[0085] Step S102: selecting a first standard point, a second standard point and a plurality of detection points from the target grayscale image according to the target process data.

[0086] In an embodiment of the present invention, by identifying the target process data, multiple processed parts of the current industrial process are determined, the target grayscale image is divided into a previous process area and multiple current process areas, and standard feature data is obtained. Then, according to the standard feature data, feature matching is performed in the previous process area to determine the first regular figure and the second regular figure, and the centroid of the first regular figure is marked as the first standard point, and the centroid of the second regular figure is marked as the second standard point. Similarly, multiple detection regular figures are selected from multiple current process areas, and the centroids of the multiple detection regular figures are marked as detection points to achieve the selection of multiple detection points.

[0087] It can be understood that the previous process area is the image area where the product appearance remains unchanged after the previous industrial process is completed; multiple current process areas correspond to multiple processed parts and are multiple image areas where the product appearance changes after the current industrial process is completed, for example: in the current industrial process, the image area with the shell and the image area with the bracket.

[0088] It can be understood that the first regular pattern, the second regular pattern and the multiple detection regular patterns can be regular patterns such as circles, squares, triangles, trapezoids, etc., among which the circle can be the image of a screw hole, a sealant opening, etc., and the square can be the image of the outer wall of the shell, the coding area, etc.

[0089] Specifically, Figure 3 A flow chart of selecting a first standard point, a second standard point and a plurality of detection points in a method provided by an embodiment of the present invention is shown.

[0090] In another preferred embodiment of the present invention, the step of selecting a first standard point, a second standard point and a plurality of detection points from the target grayscale image according to the target process data specifically comprises the following steps:

[0091] Step S1021 : dividing the target grayscale image into a previous process area and multiple current process areas according to the target process data.

[0092] Step S1022: Select a first standard point and a second standard point from the previous process area.

[0093] Specifically, Figure 4 A flow chart of selecting the first standard point and the second standard point in the method provided by an embodiment of the present invention is shown.

[0094] In another preferred embodiment of the present invention, the step of selecting the first standard point and the second standard point from the previous process area specifically comprises the following steps:

[0095] Step S10221, obtain standard feature data.

[0096] Step S10222: perform feature matching in the previous process area according to the standard feature data to determine a first regular pattern and a second regular pattern.

[0097] Step S10223: Mark the centroid of the first regular figure as a first standard point.

[0098] Step S10224: mark the centroid of the second regular figure as a second standard point.

[0099] Furthermore, the step of selecting a first standard point, a second standard point, and a plurality of detection points from the target grayscale image according to the target process data further comprises the following steps:

[0100] Step S1023: Select multiple inspection points from the multiple current process areas.

[0101] Furthermore, the intelligent industrial detection method based on image processing also includes the following steps:

[0102] Step S103: construct a detection coordinate system, obtain point coordinate data of the first standard point, the second standard point and multiple detection points, and calculate multiple detection ratios and multiple detection angles.

[0103] In the embodiment of the present invention, a positioning origin is selected from the target grayscale image, and then a detection coordinate system is constructed based on the positioning origin, and then point coordinate data of the first standard point, the second standard point and multiple detection points in the detection coordinate system are obtained. According to the point coordinate data, multiple detection ratios and multiple detection angles are calculated. Specifically, the calculation formulas of the multiple detection ratios are:

[0104] ;

[0105] in, For the The detection ratio corresponding to the detection points is For the The coordinates of the detection points, are the coordinates of the first standard point, is the coordinate of the second standard point;

[0106] The calculation formulas for the multiple detection angles are:

[0107] ;

[0108] in, For the The detection angle corresponding to each detection point.

[0109] It can be understood that the lower left corner of the target grayscale image can be selected as the positioning origin.

[0110] Specifically, Figure 5 A flow chart of calculating multiple detection ratios and multiple detection angles in the method provided by an embodiment of the present invention is shown.

[0111] Among them, in another preferred embodiment provided by the present invention, the construction of the detection coordinate system, obtaining the point coordinate data of the first standard point, the second standard point and the multiple detection points, and calculating the multiple detection ratios and the multiple detection angles specifically include the following steps:

[0112] Step S1031: Select a positioning origin from the target grayscale image.

[0113] Step S1032: construct a detection coordinate system based on the positioning origin.

[0114] Step S1033: obtaining point coordinate data of the first standard point, the second standard point and multiple detection points in the detection coordinate system.

[0115] Step S1034: Calculate multiple detection ratios and multiple detection angles according to the point coordinate data.

[0116] Furthermore, the intelligent industrial detection method based on image processing also includes the following steps:

[0117] Step S104: perform detection and analysis on the multiple detection ratios and the multiple detection angles, generate industrial detection results, and perform post-processing.

[0118] In an embodiment of the present invention, standard inspection data is matched according to target process data, and then based on the standard inspection data, a comparison and analysis is performed on multiple inspection ratios and multiple inspection angles to generate industrial inspection results. Based on the industrial inspection results, it is determined whether the inspection meets the standards. If it is determined that the inspection meets the standards, normal conduction to the later position of the process is performed, so that the target process product enters the workstation of the next processing process; if it is determined that the inspection does not meet the standards, abnormal conduction of process termination is performed, so that the target process product leaves the normal processing process flow and prepares for subsequent parts recycling.

[0119] It is understandable that the standard detection data includes detection standard ranges corresponding to multiple detection ratios and multiple detection angles.

[0120] It can be understood that by taking the positions of the first standard point and the second standard point as standard references and calculating the corresponding detection ratio and detection angle of the detection point, the distance and orientation of the detection point can be verified, and then the installation of the corresponding processed parts can be verified (for example: if the processed parts are installed incorrectly, the distance and orientation of the detection point will be incorrect; if the processed parts are not installed tightly, the distance and orientation of the detection point will be incorrect), without the need for comprehensive detection and analysis of multiple features such as the edges, corners, textures, and shapes of the product. This is convenient, fast, highly accurate, and requires less calculation.

[0121] Specifically, Figure 6 A flow chart showing detection, analysis and processing in the method provided in an embodiment of the present invention is shown.

[0122] Among them, in another preferred embodiment provided by the present invention, the detection and analysis of the multiple detection ratios and the multiple detection angles, generating industrial detection results, and performing post-process processing specifically include the following steps:

[0123] Step S1041, matching standard inspection data according to the target process data.

[0124] Step S1042: Based on the standard detection data, compare and analyze the multiple detection ratios and the multiple detection angles to generate industrial detection results.

[0125] Step S1043: Determine whether the detection meets the standard based on the industrial detection result.

[0126] Step S1044: When the detection standard is met, normal conduction of the subsequent process is carried out.

[0127] Step S1045: When the detection does not meet the standard, abnormal conduction is performed to terminate the process.

[0128] Furthermore, Figure 7 The application architecture diagram of the intelligent industrial detection system based on image processing provided by an embodiment of the present invention is shown.

[0129] Specifically, in another preferred embodiment provided by the present invention, the intelligent industrial detection system based on image processing includes:

[0130] The detection and shooting processing module 101 is used to detect and shoot the target process product, obtain the target shot image and target process data, and perform grayscale processing on the target shot image to generate a target grayscale image.

[0131] In an embodiment of the present invention, during the assembly production process of the target industrial product, the detection and shooting processing module 101 determines the current industrial process, matches the target process data corresponding to the current industrial process, identifies the target process data, determines the working cycle of the process equipment, and then periodically generates detection and shooting instructions according to the working cycle, and then, according to the detection and shooting instructions, detects and shoots the target process product after the current industrial process is assembled, obtains the target shot image, and then grayscales the target shot image to generate a target grayscale image.

[0132] Furthermore, Figure 8 It shows a structural block diagram of the detection and shooting processing module 101 in the system provided by an embodiment of the present invention.

[0133] Specifically, in another preferred embodiment of the present invention, the detection and shooting processing module 101 specifically includes:

[0134] The data matching unit 1011 is used to determine the current industrial process and match the target process data.

[0135] The instruction generating unit 1012 is used to periodically generate detection and shooting instructions according to the target process data.

[0136] The detection and shooting unit 1013 is used to detect and shoot the target process product according to the detection and shooting instruction to obtain the target shot image.

[0137] The grayscale processing unit 1014 is used to perform grayscale processing on the target captured image to generate a target grayscale image.

[0138] Furthermore, the intelligent industrial detection system based on image processing also includes:

[0139] The image point selection module 102 is used to select a first standard point, a second standard point and a plurality of detection points from the target grayscale image according to the target process data.

[0140] In an embodiment of the present invention, the image point selection module 102 identifies the target process data, determines multiple processed parts of the current industrial process, divides the target grayscale image into a previous process area and multiple current process areas, and obtains standard feature data, and then performs feature matching in the previous process area according to the standard feature data to determine the first regular figure and the second regular figure, marks the centroid of the first regular figure as the first standard point, and then marks the centroid of the second regular figure as the second standard point. Similarly, multiple detection regular figures are selected from multiple current process areas, and the centroids of the multiple detection regular figures are marked as detection points to achieve the selection of multiple detection points.

[0141] The ratio angle calculation module 103 is used to construct a detection coordinate system, obtain point coordinate data of the first standard point, the second standard point and multiple detection points, and calculate multiple detection ratios and multiple detection angles.

[0142] In the embodiment of the present invention, the ratio angle calculation module 103 selects a positioning origin from the target grayscale image, and then constructs a detection coordinate system based on the positioning origin, and then obtains the point coordinate data of the first standard point, the second standard point and the multiple detection points in the detection coordinate system, and calculates multiple detection ratios and multiple detection angles according to the point coordinate data. Specifically, the calculation formula of the multiple detection ratios is:

[0143] ;

[0144] in, For the The detection ratio corresponding to the detection points is For the The coordinates of the detection points, are the coordinates of the first standard point, is the coordinate of the second standard point;

[0145] The calculation formulas for the multiple detection angles are:

[0146] ;

[0147] in, For the The detection angle corresponding to each detection point.

[0148] Furthermore, Fig. 9 The structure block diagram of the proportional angle calculation module 103 in the system provided by the embodiment of the present invention is shown.

[0149] Specifically, in another preferred embodiment of the present invention, the ratio angle calculation module 103 specifically includes:

[0150] The origin selection unit 1031 is used to select a positioning origin from the target grayscale image.

[0151] The coordinate system construction unit 1032 is used to construct a detection coordinate system based on the positioning origin.

[0152] The coordinate acquisition unit 1033 is used to acquire point coordinate data of the first standard point, the second standard point and multiple detection points in the detection coordinate system.

[0153] The ratio angle calculation unit 1034 is used to calculate multiple detection ratios and multiple detection angles according to the point coordinate data.

[0154] Furthermore, the intelligent industrial detection system based on image processing also includes:

[0155] The detection analysis processing module 104 is used to detect and analyze the multiple detection ratios and the multiple detection angles, generate industrial detection results, and perform post-process processing.

[0156] In an embodiment of the present invention, the detection analysis processing module 104 matches the standard detection data according to the target process data, and then compares and analyzes multiple detection ratios and multiple detection angles based on the standard detection data to generate industrial detection results. According to the industrial detection results, it is determined whether the detection is up to standard. If it is determined that the detection is up to standard, normal conduction to the later position of the process is performed, so that the target process product enters the workstation of the next processing process; if it is determined that the detection is not up to standard, abnormal conduction of process termination is performed, so that the target process product leaves the normal processing process flow and prepares for subsequent parts recovery.

[0157] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0158] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0159] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. An intelligent industrial detection method based on image processing, characterized in that: The method specifically comprises the following steps: Detect and photograph the target process product to obtain a target photographed image and target process data, and grayscale the target photographed image to generate a target grayscale image; According to the target process data, selecting a first standard point, a second standard point and a plurality of detection points from the target grayscale image; The target grayscale image is divided into a previous process area and multiple current process areas, and standard feature data is obtained. Then, according to the standard feature data, feature matching is performed in the previous process area to determine the first regular figure and the second regular figure, and the centroid of the first regular figure is marked as the first standard point, and the centroid of the second regular figure is marked as the second standard point. Similarly, multiple detection regular figures are selected from multiple current process areas, and the centroids of the multiple detection regular figures are marked as detection points; the previous process area is an image area where the product appearance remains unchanged after the previous industrial process is completed; the multiple current process areas are multiple image areas corresponding to multiple processed parts, where the product appearance changes after the current industrial process is completed; Construct a detection coordinate system, obtain the point coordinate data of the first standard point, the second standard point and multiple detection points, and calculate multiple detection ratios and multiple detection angles; From the target grayscale image, select the positioning origin, build a detection coordinate system, obtain the point coordinate data of the first standard point, the second standard point and multiple detection points in the detection coordinate system, and calculate multiple detection ratios and multiple detection angles. Specifically, the calculation formula for multiple detection ratios is: ； in, For the The detection ratio corresponding to the detection points is For the The coordinates of the detection points, are the coordinates of the first standard point, is the coordinate of the second standard point; The calculation formula for multiple detection angles is: ； in, For the The detection angle corresponding to each detection point; Detection and analysis are performed on the plurality of detection ratios and the plurality of detection angles to generate industrial detection results, and post-processing is performed.

2. The intelligent industrial detection method based on image processing according to claim 1 is characterized in that: The detecting and photographing of the target process product, obtaining the target photographed image and the target process data, and graying the target photographed image to generate the target grayscale image specifically comprises the following steps: Determine the current industrial process and match the target process data; According to the target process data, periodically generate detection shooting instructions; According to the inspection and shooting instructions, the target process product is inspected and photographed to obtain the target photographed image; The target captured image is grayscaled to generate a target grayscale image.

3. The intelligent industrial detection method based on image processing according to claim 1 is characterized in that: The detecting and analyzing of the plurality of detection ratios and the plurality of detection angles, generating industrial detection results, and performing post-process processing specifically include the following steps: According to the target process data, matching standard inspection data; Based on the standard detection data, a plurality of the detection ratios and a plurality of the detection angles are compared and analyzed to generate an industrial detection result; According to the industrial test results, judging whether the test meets the standards; When the test is up to standard, the normal conduction of the post-process is carried out; When the detection does not meet the standards, an abnormal conduction is performed to terminate the process.

4. Intelligent industrial inspection system based on image processing, characterized by: The system includes a detection and shooting processing module, an image point selection module, a ratio angle calculation module and a detection analysis processing module, wherein: The detection and shooting processing module is used to detect and shoot the target process product, obtain the target shot image and target process data, and grayscale the target shot image to generate a target grayscale image; An image point selection module, used to select a first standard point, a second standard point and a plurality of detection points from the target grayscale image according to the target process data; The image point selection module divides the target grayscale image into a previous process area and multiple current process areas, and obtains standard feature data, and then performs feature matching in the previous process area according to the standard feature data to determine the first regular figure and the second regular figure, and marks the centroid of the first regular figure as the first standard point, and then marks the centroid of the second regular figure as the second standard point. Similarly, multiple detection regular figures are selected from multiple current process areas, and the centroids of the multiple detection regular figures are marked as detection points; the previous process area is an image area where the product appearance remains unchanged after the previous industrial process is completed; the multiple current process areas are multiple image areas corresponding to multiple processed parts, where the product appearance changes after the current industrial process is completed; A ratio angle calculation module is used to construct a detection coordinate system, obtain point coordinate data of a first standard point, a second standard point and multiple detection points, and calculate multiple detection ratios and multiple detection angles; The ratio angle calculation module selects the positioning origin from the target grayscale image, constructs a detection coordinate system, obtains the point coordinate data of the first standard point, the second standard point and multiple detection points in the detection coordinate system, and calculates multiple detection ratios and multiple detection angles. Specifically, the calculation formula for multiple detection ratios is: ； in, For the The detection ratio corresponding to the detection points is For the The coordinates of the detection points, are the coordinates of the first standard point, is the coordinate of the second standard point; The calculation formula for multiple detection angles is: ； in, For the The detection angle corresponding to each detection point; The detection analysis and processing module is used to detect and analyze the multiple detection ratios and the multiple detection angles, generate industrial detection results, and perform post-process processing.

5. The intelligent industrial detection system based on image processing according to claim 4 is characterized in that: The detection and shooting processing module specifically includes: A data matching unit is used to determine the current industrial process and match the target process data; An instruction generation unit, used for periodically generating detection and shooting instructions according to the target process data; A detection and shooting unit, used to detect and shoot the target process product according to the detection and shooting instruction, and obtain the target shot image; The grayscale processing unit is used to perform grayscale processing on the target captured image to generate a target grayscale image.

6. The intelligent industrial detection system based on image processing according to claim 4 is characterized in that: The proportional angle calculation module specifically includes: An origin selection unit, used for selecting a positioning origin from the target grayscale image; A coordinate system construction unit, used to construct a detection coordinate system based on the positioning origin; A coordinate acquisition unit, used for acquiring point coordinate data of a first standard point, a second standard point and a plurality of detection points in the detection coordinate system; The ratio angle calculation unit is used to calculate multiple detection ratios and multiple detection angles according to the point coordinate data.

Images