An intelligent control system for a production line

Abstract

The application discloses a production line intelligent control system, and relates to the technical field of intelligent control systems, comprising: a monitoring and collecting terminal, which is used for collecting product images after processing nodes on a production line process, defining a processing area on the product images based on processing information of a target processing node, and obtaining a target processing node processing area image; an image multi-dimensional processing module, which performs contour detection processing on the processing area image to obtain a contour deviation detection image, analyzes and obtains a contour precision deviation coefficient based on the contour deviation detection image, and performs regional distribution processing on the processing area image to obtain a plurality of sub-region deviation detection images. The application accurately analyzes and evaluates the product processing condition from two dimensions in a distributed manner, so that the defective products and waste products can be accurately identified, the defective products and waste products can be automatically controlled to be removed to prevent them from entering the next processing node, the processing efficiency of the production line is significantly improved, and resource waste is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control systems, and particularly relates to a production line intelligent control system. BACKGROUND

[0002] The production line intelligent control system and the control method are important components of modern manufacturing industry, and they have important significance for improving production efficiency, ensuring product quality and reducing production cost. The production line intelligent control system is a highly integrated system, which usually includes multiple levels and modules, such as enterprise management level, production management level, process control level, device control level and detection driving level. These levels and modules share information and work collaboratively through advanced communication technology and network technology, and use intelligent control technology and engineering methods to control and optimize the production process in real time, ensuring the stability and efficiency of the production process.

[0003] For example, a Chinese patent with the name of "Mechanical processing line product processing quality monitoring system" and the authorization announcement number of CN212160443U and the authorization announcement date of 2020.12.15, which includes a monitoring gantry, a central processor, an AI recognition module and a wireless transmission module. The inside bottom of the monitoring gantry is provided with a conveying belt, the inside top and the inside left and right side walls of the monitoring gantry are provided with monitoring probes, the top of the monitoring gantry is provided with a central processor, the central processor is bidirectionally connected with a data storage module and an image comparison module, the central processor is electrically connected with the monitoring probes, the central processor is bidirectionally connected with the AI recognition module, the central processor is bidirectionally connected with the wireless transmission module, and the wireless transmission module is bidirectionally connected with a terminal service station.

[0004] The existing production line detects the product at each processing node by setting a monitoring module to collect product photos, and then uses an AI recognition module or a machine learning model to evaluate and analyze the product processing state. When analyzing and processing the collected images, there are many features on the images, which affects the efficiency and accuracy of the analysis, and cannot meet the accuracy and timeliness requirements of the production line processing. SUMMARY

[0005] The present application aims to provide a production line intelligent control system to solve the above problems in the prior art.

[0006] To achieve the above purpose, the present application adopts the following technical scheme: a production line intelligent control system, comprising:

[0007] The monitoring acquisition terminal is used for acquiring a product image processed by a processing node on a production line, and based on processing information of a target processing node, a processing area is demarcated on the product image to obtain a target processing node processing area image;

[0008] The image multi-dimensional processing module performs contour detection processing on the processing area image to obtain a contour deviation detection image, analyzes the contour deviation detection image to obtain a contour precision deviation coefficient, performs regional distribution processing on the processing area image to obtain a plurality of sub-region deviation detection images, calculates a sub-region deviation defect coefficient of each sub-region based on the sub-region deviation detection images, and integrates and calculates a region precision deviation coefficient;

[0009] The processing condition analysis unit is used for integrating and calculating a product processing defect coefficient, and evaluating a current processing condition of the product based on the product processing defect coefficient;

[0010] The rejection processing unit is used for calling the current processing condition of the product, and performing a rejection operation on the product based on the current processing condition of the product.

[0011] As a further description of the above technical solutions:

[0012] The contour detection processing on the processing area image to obtain a contour deviation detection image specifically includes:

[0013] An edge image is extracted by performing edge detection on the processing area image, and a current processing area image contour line is obtained based on the edge image;

[0014] The current processing area image contour line is mapped on a blank image background layer;

[0015] A detection contour line is constructed, the detection contour line is mapped on the blank image background layer and covers the current processing area image contour line, and a contour deviation detection image is obtained by removing the detection contour line region.

[0016] As a further description of the above technical solutions:

[0017] The contour precision deviation coefficient obtained based on the contour deviation detection image analysis specifically includes:

[0018] The processing area standard contour line is marked on the contour deviation detection image;

[0019] Pixels outside the standard contour line in the contour deviation detection image are extracted, and the gray values of the pixels are calculated and marked as an external contour deviation value Pw;

[0020] Pixels inside the standard contour line in the contour deviation detection image are extracted, and the gray values of the pixels are calculated and marked as an internal contour deviation value Pn;

[0021] The external contour deviation value and the internal contour deviation value are integrated to obtain a contour precision deviation coefficient PL.

[0022] As a further description of the above technical solution:

[0023] The region distribution processing is performed on the processing region image to obtain a plurality of sub-region deviation detection images, and the sub-region deviation defect coefficients of the sub-region deviation detection images are calculated based on the sub-region deviation detection images and are integrated to obtain a region precision deviation coefficient. Specifically, the region precision deviation coefficient is calculated as follows:

[0024] The processing region image is called, and the contour line of the current processing region image is mapped on the processing region image. The standard region detection image is obtained by cropping and removing the contour line region of the current processing region image.

[0025] Based on the processing information of the target processing node, the regions with the same processing standard are divided on the standard region detection image to obtain a plurality of sub-region deviation detection images, and the deviation defect coefficients of the sub-region deviation detection images are distributedly calculated. Then, the deviation defect coefficients of the sub-region deviation detection images are summed to obtain a region precision deviation coefficient.

[0026] As a further description of the above technical solution:

[0027] The distributed calculation of the deviation defect coefficients of the sub-region deviation detection images is specifically as follows:

[0028] An m-size pixel detection region is constructed on each sub-region deviation detection image, and the pixel gray value of each pixel detection region is calculated to obtain a gray value data set U ∈ (X1, X2, X3...X m ) of the sub-region deviation detection image. m X

[0029] The self-deviation coefficient and the standard deviation coefficient are calculated respectively through the gray value data set of the sub-region deviation detection image.

[0030] The self-deviation coefficient and the standard deviation coefficient are summed to calculate the deviation defect coefficient of the sub-region deviation detection image.

[0031] As a further description of the above technical solution:

[0032] The self-deviation coefficient and the standard deviation coefficient are calculated respectively through the gray value data set of the sub-region deviation detection image. Specifically, the self-deviation coefficient and the standard deviation coefficient are calculated as follows:

[0033] The gray value data set U ∈ (X1, X2, X3...X m ) of the sub-region deviation detection image is called.

[0034] The self-deviation coefficient is calculated, P z represents the self-bias coefficient of the sub-region bias detection image, X m represents the pixel gray value of the mth pixel detection region of the sub-region bias detection image, δ1 is a preset weight coefficient and δ1>0;

[0035] calculating the standard deviation coefficient, represents the standard deviation coefficient of the sub-region bias detection image, X e represents the standard pixel gray value of the sub-region bias detection image, δ2 is a preset weight coefficient and δ2>0.

[0036] As a further description of the above technical solution:

[0037] The detection contour line is constructed as follows: based on the machining information of the target machining node, a machining region standard contour line is obtained, and a detection contour line is constructed with the standard contour line as a reference line and extending 0.5Le length to both sides, where Le is a current machining region image contour line error compensation parameter.

[0038] As a further description of the above technical solution:

[0039] The machining condition includes qualified products, substandard products, and waste products.

[0040] As a further description of the above technical solution:

[0041] The product machining defect coefficient is integrated and calculated, and the current machining condition of the product is evaluated based on the product machining defect coefficient, specifically:

[0042] The contour precision bias coefficient and the region precision bias coefficient are called to calculate the product machining defect coefficient, P Q = P L * β1+ P Y * β2, where P Q represents the product machining defect coefficient, P L represents the contour bias defect coefficient, P Y represents the region precision bias coefficient, β1 and β2 represent weight coefficients, and β1 and β2 are both greater than 0.

[0043] A preset product machining defect coefficient evaluation index P g1 , P g2 , where P g1 > P g2 > 0.

[0044] When P Q > P g1 , it indicates that the current product machining condition is waste.

[0045] When P Q < Pg2 When P

[0046] When P g1 ≥P Q ≥P g2 , indicating that the current product processing condition is a defective product.

[0047] As a further description of the above technical solution:

[0048] The rejection processing unit is used to call the current product processing condition, and performs rejection operation on the product based on the current product processing condition, specifically:

[0049] Collecting the image of the processing area of the target processing node to obtain the current product processing condition,

[0050] When the current product processing condition is a defective product, a first control instruction is generated and controlled to the execution mechanism, and the execution mechanism rejects the currently processed product and places it in the defective product recycling frame;

[0051] When the current product processing condition is a defective product, a first control instruction is generated and controlled to the execution mechanism, and the execution mechanism rejects the currently processed product and places it in the defective product recycling frame;

[0052] When the current product processing condition is a qualified product, no rejection operation is performed on the currently processed product.

[0053] The present application provides a production line intelligent control system. It has the following beneficial effects:

[0054] The production line intelligent control system calculates the contour precision deviation coefficient and the area precision deviation coefficient of the processing area by distributing the contour of the processing area and the processing area independently, and integrates and calculates to obtain the product processing defect coefficient. By evaluating the current processing state of the product based on the product processing defect coefficient, accurate analysis and evaluation of the product processing condition from two dimensions are realized, so as to accurately identify defective products and waste products, and automatically control the rejection of defective products and waste products to prevent them from entering the next processing node, significantly improving the processing efficiency of the production line and reducing resource waste. BRIEF DESCRIPTION OF DRAWINGS

[0055] Fig. 1 A module schematic diagram of the production line intelligent control system according to the present application is provided.

[0056] Fig. 2 A processing flow schematic diagram of the production line intelligent control system according to the present application is provided. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application but not all the embodiments of the present application.

[0058] With reference to Figs. 1-2 An intelligent control system of a production line, wherein the production line comprises a plurality of processing nodes, the processing nodes are connected with each other through conveying lines, a product to be processed is conveyed to each processing node through the conveying lines in sequence, and the product is processed through each processing node in a pipeline manner. The intelligent control system of the production line comprises a monitoring and collecting terminal. Optionally, the monitoring and collecting terminal is an industrial camera, which is arranged at each processing node of the production line and is used to collect a product image of the product processed by the processing node. The product image comprises region image information of the product processed by a target processing node. Based on processing information of the target processing node, the processing information comprises processing drawings, process files or other data sources to determine the position, size and shape of a processing region of the target processing node, and then the processing region of the target processing node is marked on the product image to obtain a processing region image. The image features of the product image itself are removed by sliding the processing region on the product image and obtaining the processing region image, the complexity of subsequent image analysis is reduced, and the efficiency of image analysis and processing is improved.

[0059] An image multi-dimensional processing module is configured to perform contour detection processing on the processing region image to obtain a contour deviation detection image, and perform analysis based on the contour deviation detection image to obtain a contour precision deviation coefficient. The contour precision deviation coefficient is used to evaluate the processing precision of the processing region corresponding to the target processing node. The greater the contour precision deviation coefficient is, the worse the processing precision of the contour is. The image multi-dimensional processing module is further configured to perform region distribution processing on the processing region image to obtain a plurality of sub-region deviation detection images, calculate a sub-region deviation defect coefficient based on each sub-region deviation detection image, and integrate and calculate to obtain a region precision deviation coefficient. The region precision deviation coefficient is used to evaluate the processing precision of the processing region corresponding to the target processing node. The greater the region precision deviation coefficient is, the worse the processing precision of the processing region is.

[0060] A processing condition analysis unit is configured to integrate and calculate a product processing defect coefficient, and evaluate a current processing condition of the product based on the product processing defect coefficient. The processing condition comprises a qualified product, a substandard product and a waste product. The qualified product means that the product is currently processed and qualified to enter the next processing node for processing. The substandard product means that the product is currently processed and unqualified, and needs to be reworked. The waste product means that the product is currently processed and unqualified, and cannot be reworked.

[0061] The rejection processing unit is arranged in the conveying system between each processing node of the production line, and the rejection processing unit comprises a controller and an execution mechanism, the controller is used to call a current processing condition of a product, generate a control instruction based on the current processing condition of the product, and transmit the control instruction to the execution mechanism, and the execution mechanism is used to perform a rejection operation on the product, and the execution mechanism is a mechanical hand, that is, when the controller calls the current processing condition of the product, the control instruction is generated to control the mechanical hand when the processing condition is a defective product or a waste product, and the mechanical hand is used to reject the processing product of the defective product or the waste product from the production line, so that the defective product and the waste product cannot enter the next processing node of the production line.

[0062] The embodiment provides a production line intelligent control system, which calculates a contour precision deviation coefficient and a region precision deviation coefficient of a processing area independently through a profile of the processing area, integrates and calculates a product processing defect coefficient, evaluates a current processing state of a product based on the product processing defect coefficient, accurately analyzes and evaluates the processing state of the product from two dimensions, accurately identifies defective products and waste products, and automatically controls the rejection of the defective products and the waste products to prevent the defective products and the waste products from entering a next processing node, thereby significantly improving the processing efficiency of the production line and reducing resource waste.

[0063] In still another embodiment of the present application, the profile deviation detection image obtained by performing contour detection processing on the processing area image is specifically as follows:

[0064] An edge image is obtained by performing edge detection on the processing area image, and a contour line is constructed by tracking continuous edge pixels based on the edge image to obtain a contour line of the current processing area image;

[0065] The contour line of the current processing area image is mapped on a blank image background layer;

[0066] A detection contour line is constructed, and the construction of the detection contour line is specifically as follows: a standard contour line of the processing area is obtained based on processing information of the target processing node, the standard contour line is taken as a reference line and is extended to both sides by 0.5Le length to construct a detection contour line, wherein Le is a current processing area image contour line error compensation parameter, the detection contour line is mapped on the blank image background layer and covers the contour line of the current processing area image, and the contour deviation monitoring image is obtained by removing the detection contour line region by cutting. Thus, the standard contour line region meeting the error requirement is removed, and the contour deviation monitoring image only retains the contour line with processing deviation.

[0067] Further, the contour precision deviation coefficient is obtained based on analysis of the contour deviation monitoring image, and the analysis of the contour deviation monitoring image is specifically as follows:

[0068] The standard contour line of the processing area is marked on the contour deviation monitoring image;

[0069] Extract the pixels located outside the standard contour line in the contour deviation monitoring image, and calculate the gray value of these pixels, marked as the external contour deviation value Pw.

[0070] Extract the pixels located inside the standard contour line in the contour deviation monitoring image, and calculate the gray value of these pixels, marked as the internal contour deviation value Pn.

[0071] The external contour deviation value and the internal contour deviation value are integrated and calculated to obtain the contour precision deviation coefficient PL. The calculation logic of the contour precision deviation coefficient PL is that the external contour deviation value weight coefficient S1 and the internal contour deviation value weight coefficient S2 are preset, and S1>S2>0. Optionally, S1=0.62 and S2=0.38. The calculation logic of the contour precision deviation coefficient PL is PL=Pw*S1+Pn*S2.

[0072] The embodiment provides a production line intelligent control system. When the contour precision deviation coefficient is calculated, the current processing area image contour line extracted by constructing a detection contour line is processed, only the contour line with processing deviation is retained, the subsequent calculation amount is significantly reduced, the external contour deviation value and the internal contour deviation value are calculated based on the standard contour line distribution, the weight of the external contour deviation value and the internal contour deviation value is set by the staff based on the influence degree, and the contour precision deviation coefficient is calculated by weighted summation, the accuracy and comprehensiveness of the contour precision deviation analysis are significantly improved, and the accuracy of the product processing state evaluation is improved.

[0073] In still another embodiment of the present application, the processing area image is subjected to region distribution processing to obtain a plurality of sub-region deviation detection images, and a sub-region deviation defect coefficient of each sub-region deviation detection image is calculated based on the sub-region deviation detection image to obtain a region precision deviation coefficient by integrated calculation.

[0074] The processing area image is called, the current processing area image contour line is mapped on the processing area image, and the standard region detection image is obtained by cutting and removing the current processing area image contour line region; the contour on the processing area image is removed, and only the standard region detection image inside the processing region is retained;

[0075] Based on the processing information of the target processing node, the regions with the same processing standard are divided on the standard region detection image to obtain a plurality of sub-region deviation detection images. In an ideal state, the pixel gray values of the sub-region deviation detection images of the regions with the same processing standard are the same, the deviation defect coefficients of the sub-region deviation detection images are calculated in a distributed manner, and then the deviation defect coefficients of the sub-region deviation detection images are summed to obtain a region precision deviation coefficient.

[0076] Further, the distributed calculation of the deviation defect coefficients of the sub-region deviation detection images is specifically as follows:

[0077] m pixel detection regions are constructed on each sub-region deviation detection image, and pixel gray values of each pixel detection region are calculated to obtain a gray value data set U of the sub-region deviation detection image, wherein U∈(X1, X2, X3...X m ), X m represents a pixel gray value of the mth pixel detection region of the sub-region deviation detection image.

[0078] Self-deviation coefficients and standard deviation coefficients are calculated respectively by the gray value data set of the sub-region deviation detection image, wherein the self-deviation coefficient represents the machining precision deviation between different pixel detection regions on the same sub-region deviation detection image, and the standard deviation coefficient represents the deviation between the machining precision of the sub-region deviation detection image and the precision in the standard precision machining state.

[0079] The self-deviation coefficients and the standard deviation coefficients are summed to obtain a deviation defect coefficient of the sub-region deviation detection image.

[0080] In still another embodiment of the present application, the self-deviation coefficients and the standard deviation coefficients are calculated respectively by the gray value data set of the sub-region deviation detection image, specifically as follows:

[0081] The gray value data set U of the sub-region deviation detection image is called, wherein U∈(X1, X2, X3...X m );

[0082] The self-deviation coefficient is calculated as follows: P z represents the self-deviation coefficient of the sub-region deviation detection image, X m represents a pixel gray value of the mth pixel detection region of the sub-region deviation detection image, and δ1 is a preset weight coefficient and δ1>0.

[0083] The standard deviation coefficient is calculated as follows: represents the standard deviation coefficient of the sub-region deviation detection image, X e represents a standard pixel gray value of the sub-region deviation detection image, and δ2 is a preset weight coefficient and δ2>0.

[0084] The present embodiment provides a production line intelligent control system. When calculating the region precision deviation coefficient, the same standard region on a standard region detection image is divided into independent sub-region deviation detection images, and the machining precision deviation between different pixel detection regions on the same sub-region deviation detection image and the deviation between the machining precision of the sub-region deviation detection image and the precision in the standard precision machining state are calculated to obtain the region precision deviation coefficient, thereby significantly enhancing the accuracy and comprehensiveness of the region precision deviation analysis and further improving the accuracy of the product machining state evaluation.

[0085] In still another embodiment of the present application, the integrated product processing defect coefficient is calculated, and the current product processing condition is evaluated based on the product processing defect coefficient, specifically:

[0086] The profile precision deviation coefficient and the area precision deviation coefficient are called to calculate the product processing defect coefficient, P Q = P L *β1+P Y *β2, wherein P Q represents the product processing defect coefficient, P L represents the profile deviation defect coefficient, P Y represents the area precision deviation coefficient, β1 and β2 represent weight coefficients, and β1 and β2 are both greater than 0.

[0087] The preset product processing defect coefficient evaluation index P g1 , P g2 , wherein P g1 >P g2 >0.

[0088] When P Q >P g1 , it indicates that the current product processing condition is a waste product.

[0089] When P Q P g2 , it indicates that the current product processing condition is a qualified product.

[0090] When P g1 P Q P g2 , it indicates that the current product processing condition is a substandard product.

[0091] In still another embodiment of the present application, the rejection processing unit is used to call the current product processing condition, and the product is rejected based on the current product processing condition, specifically:

[0092] The target processing node processing area image product current processing condition is collected,

[0093] When the product current processing condition is a substandard product, a first control instruction is generated and controlled to an execution mechanism, the execution mechanism rejects the currently processed product and places it in a substandard product recycling frame;

[0094] When the product current processing condition is a waste product, a second control instruction is generated and controlled to an execution mechanism, the execution mechanism rejects the currently processed product and places it in a waste product recycling frame;

[0095] By setting the substandard product recycling frame and the waste product recycling frame, the rejection unit can classify and store the substandard products and the waste products when rejecting the substandard products and the waste products, which facilitates subsequent staff to classify and process the substandard products and the waste products.

[0096] When the current processing state of the product is a qualified product, the current processing product is not rejected at this time. That is, the qualified product is conveyed into the next processing node for flow processing through the conveying line between the processing nodes, wherein the monitoring and collecting terminal, the image processing module, the processing state analysis unit and the rejection processing unit are arranged at each processing node position of the production line, so as to realize the collection, analysis and evaluation of the processing state of the product after processing of each processing node, and realize the rejection of the substandard product and the waste product, so that the substandard product and the waste product cannot flow into the adjacent next processing node, avoiding the continuous processing of the substandard product and the waste product by the subsequent processing node, improving the processing efficiency of the production line and reducing the resource waste.

[0097] In the description of the present specification, the description of the terms "one embodiment", "example", "specific example" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are contained in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0098] The above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can make equivalent replacement or change according to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. An intelligent control system for a production line, characterized in that: include: The monitoring and acquisition terminal is used to collect images of products processed by processing nodes on the production line, delineate processing areas on the product images based on the processing information of the target processing nodes, and obtain images of the processing areas of the target processing nodes; The image multi-dimensional processing module performs contour detection on the processing area image to obtain a contour deviation detection image, obtains the contour accuracy deviation coefficient based on the contour deviation monitoring image analysis, and marks the standard contour line of the processing area on the contour deviation monitoring image; extracts pixels outside the standard contour line in the contour deviation monitoring image, calculates the grayscale values ​​of these pixels, and marks them as external contour deviation values ​​Pw; extracts pixels inside the standard contour line in the contour deviation monitoring image, calculates the grayscale values ​​of these pixels, and marks them as internal contour deviation values ​​Pn; integrates the external contour deviation value and the internal contour deviation value to obtain the contour accuracy deviation coefficient PL; Performing regional distribution processing on the processing area image to obtain multiple sub-region deviation detection images, calculating and obtaining the deviation defect coefficient of each sub-region based on the deviation detection images of each sub-region, and integrating and calculating to obtain the regional accuracy deviation coefficient; retrieving the processing area image, mapping the contour line of the current processing area image onto the processing area image, and cropping and removing the contour line area of ​​the current processing area image to obtain the standard area detection image; Based on the processing information of the target processing node, the area with the same processing standard is divided on the standard area detection image to obtain multiple sub-area deviation detection images, and the deviation defect coefficient of each sub-area deviation detection image is distributedly calculated. On each sub-area deviation detection image, m pixel detection areas of the same size are constructed, and the pixel grayscale value of each pixel detection area is calculated to obtain the grayscale value dataset U∈(X1, X2, X3...X m ), X m Representing the pixel grayscale value of the mth pixel detection area of ​​the sub-region deviation detection image; calculating the self-deviation coefficient and the standard deviation coefficient respectively using the grayscale value data set of the sub-region deviation detection image; summing the self-deviation coefficient and the standard deviation coefficient to obtain the deviation defect coefficient of the sub-region deviation detection image; then summing the deviation defect coefficients of each sub-region deviation detection image to obtain the regional accuracy deviation coefficient; A processing status analysis unit is used to integrate and calculate the product processing defect coefficient and evaluate the current processing status of the product based on the product processing defect coefficient; The rejection processing unit is used to retrieve the current processing status of the product and perform rejection operations on the product based on the current processing status of the product.

2. The intelligent control system for a production line according to claim 1, characterized in that: The contour detection process of the processing area image is performed to obtain the contour deviation detection image as follows: Performing image edge detection and extraction on the processing area image to obtain an edge image, and obtaining the image contour line of the current processing area based on the edge image; Map the image outline of the current processing area to the blank image background layer; Construct a detection contour line, map the detection contour line on the blank image background layer and cover the image contour line of the current processing area, and crop and remove the detection contour line area to obtain the contour deviation monitoring image.

3. The intelligent control system for a production line according to claim 1, characterized in that: The gray value data set of the sub-region deviation detection image is used to calculate the self-deviation coefficient and the standard deviation coefficient respectively as follows: Retrieve the grayscale value dataset U∈(X1, X2, X3...X m ); Calculate the coefficient of self-bias, P z Denotes the self-bias coefficient of the sub-region deviation detection image, X m Represents the pixel grayscale value of the mth pixel detection area of ​​the sub-region deviation detection image, δ1 is a preset weight coefficient and δ1>0; Calculate the coefficient of standard deviation, represents the standard deviation coefficient of the sub-region deviation detection image, X e Represents the standard pixel grayscale value of the sub-region deviation detection image, δ2 is the preset weight coefficient and δ2>0.

4. An intelligent control system for a production line according to claim 1 or 2, characterized in that: The specific steps of constructing the detection contour line are as follows: based on the processing information of the target processing node, the standard contour line of the processing area is obtained, and the detection contour line is constructed by extending the standard contour line to both sides by a length of 0.5Le, where Le is the error compensation parameter of the image contour line of the current processing area.

5. The intelligent control system for a production line according to claim 1, characterized in that: The processing status includes qualified products, defective products and waste products.

6. The intelligent control system for a production line according to claim 1, characterized in that: The integrated calculation of the product processing defect coefficient and the evaluation of the current processing status of the product based on the product processing defect coefficient are specifically as follows: Retrieve the contour accuracy deviation coefficient and the regional accuracy deviation coefficient to calculate the product processing defect coefficient, P Q =P L *β1+P Y *β2, where P Q Indicates the product processing defect coefficient, P L Indicates contour deviation defect, P Y represents the regional accuracy deviation coefficient, β1 and β2 represent weight coefficients, and both β1 and β2 are greater than 0; Preset product processing defect coefficient evaluation index P g1 、P g2 , where P g1 >P g2 >0; When P Q >P g1 When , it means the current product processing status is waste; When P Q <P g2 When , it means the current product processing status is qualified; When P g1 ≥P Q ≥P g2 , it indicates that the current product processing status is defective.

7. The intelligent control system for a production line according to claim 1, characterized in that: The rejection processing unit is used to retrieve the current processing status of the product and perform rejection operations on the product based on the current processing status of the product. Specifically: Collect the current processing status of the target processing node processing area image product, When the current processing status of the product is defective, a first control instruction is generated and sent to the execution mechanism, which removes the currently processed product and places it in a defective product recycling box; When the current processing status of the product is waste, a second control instruction is generated and sent to the execution mechanism, and the execution mechanism removes the currently processed product and places it in a waste recycling box; When the current processing status of the product is qualified, the currently processed product will not be rejected.

Images