Industrial product quality detection method and system based on artificial intelligence

Abstract

The invention relates to the technical field of quality detection, and discloses an industrial product quality detection method and system based on artificial intelligence. According to the industrial product quality detection method and system based on artificial intelligence, the overall image data of each industrial product is obtained through a high-resolution camera and an advanced image processing technology, and the quality of the image data is improved by applying automatic denoising, enhancing and cutting operations of an industrial product image preprocessing unit; by means of the method, subsequent calculation of parameters such as the size, the surface defect rate and the color standard deviation is more accurate and reliable, whether the product quality is qualified or not can be rapidly and accurately judged by conducting qualified numerical analysis on the parameters, the detection precision and consistency are remarkably improved, and the detection efficiency is improved. Meanwhile, the two-dimensional code technology is used for directly associating the detection report of the qualified product with the product package, the product traceability and the consumer trust degree are improved, and the detection efficiency is greatly improved when the method is integrated for large-scale industrial product production quality detection.

Description

Technical Field

[0001] The present invention relates to the technical field of quality inspection, and in particular to an industrial product quality inspection method and system based on artificial intelligence. Background Art

[0002] In modern manufacturing, the quality of industrial products is directly related to the competitiveness and market reputation of enterprises. High-quality products can not only meet the needs of consumers, but also enhance brand loyalty, reduce after-sales service costs and improve production efficiency. Therefore, ensuring that industrial products meet quality standards in all aspects of design, production and delivery is an important link that cannot be ignored in enterprise management and production processes. As consumers' requirements for product quality increase, enterprises need to continuously optimize quality inspection methods to adapt to market changes and technological advances.

[0003] Traditional manual inspection methods rely on the operator's experience and technical level and are easily affected by the inspector's subjective factors, resulting in insufficient accuracy in product quality inspection. In addition, when faced with large-scale industrial product production quality inspection, the inspection efficiency is particularly low. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the deficiencies in the prior art, the present invention provides an industrial product quality inspection method and system based on artificial intelligence. The overall image data of each industrial product is acquired by introducing a high-resolution camera and advanced image processing technology. The automatic denoising, enhancement and cropping operations of the industrial product image preprocessing unit are used to improve the quality of the image data, so that the subsequent calculation of parameters such as size, surface defect rate and color standard deviation is more accurate and reliable. By performing qualified numerical analysis on these parameters, it is possible to quickly and accurately determine whether the product quality is qualified, which significantly improves the accuracy and consistency of the inspection. At the same time, the inspection report of qualified products is directly linked to the product packaging by using QR code technology, which improves product traceability and consumer trust. By combining the above methods, the efficiency of inspection is greatly improved when facing large-scale industrial product production quality inspection, and the above problems are solved.

[0006] (II) Technical solution

[0007] To achieve the above object, the present invention provides the following technical solution: an industrial product quality detection method based on artificial intelligence, comprising the following steps:

[0008] S1. Obtain image data of industrial products through high-resolution cameras;

[0009] S2. De-noising, enhancing and cropping the image data of industrial products;

[0010] S3, calculating the average value of the industrial product size, the standard deviation of the industrial product size, the surface defect rate of the industrial product, the color standard deviation of the industrial product, and the industrial product strength according to the industrial product image after denoising, cropping, and enhancement operations;

[0011] S4. Conduct qualified numerical analysis on the average value of industrial product size, standard deviation of industrial product size, surface defect rate of industrial products, color standard deviation of industrial products and strength of industrial products, comprehensively judge whether the quality of industrial products is qualified, and generate industrial product quality inspection report;

[0012] S5. Eliminate unqualified industrial products and make the qualified industrial product quality inspection report into a QR code for later printing on the outer packaging of qualified industrial products.

[0013] Preferably, the image denoising formula of the industrial product is as follows:

[0014]

[0015] In the formula, I clean (x, y) represents the pixel value of the denoised industrial product image at the coordinate (x, y), I(x+i, y+j) represents the pixel value of the original industrial product image at the coordinate (x, y), k represents the size of the filter, represents a constant factor used to normalize the summation result. Represents the accumulation of all pixel values ​​I(x+i,y+j) within the k*k window.

[0016] Preferably, the image enhancement formula of the industrial product is as follows:

[0017] I enhanced (x,y)=α*I clean (x,y)+β

[0018] In the formula, I enhanced (x, y) represents the pixel value of the enhanced industrial product image at the coordinate (x, y), α represents the contrast value of the adjusted image, β represents the offset value of the adjusted image brightness, I clean (x,y) represents the pixel value of the denoised industrial product image at the coordinate (x,y).

[0019] Preferably, the image cropping formula of the industrial product is as follows:

[0020] I crop (x,y)=I enhanced (x',y')

[0021] x'=x start+(xx cut )

[0022] y′=y start +(yy cut )

[0023] In the formula, I crop (x, y) represents the pixel value of the cropped industrial product image at the new position (x, y), (x start ,y start ) represents the coordinates of the upper left corner of the cropping area, (x cut ,y cut ) represents the image cropping boundary of the enhanced industrial product, and (x', y') represents the new position coordinates of the image of the enhanced industrial product.

[0024] Preferably, the average value calculation formula of the industrial product size is as follows:

[0025]

[0026] In the formula, Size avg represents the average value of industrial product size, n represents the total number of industrial products measured, L p Indicates the size value of the pth product, p represents the counting subscript, It means that the average value of the industrial product size is obtained by adding up the sizes of all measured industrial products and then dividing it by the total number of measured industrial products.

[0027] Preferably, the standard deviation calculation formula of the industrial product size is as follows:

[0028]

[0029] In the formula, Size std represents the standard deviation of the size of industrial products, n represents the total number of industrial products measured, L p Indicates the size value of the pth product, Size avg Indicates the average value of industrial product size, represents the sum of the squares of the differences between each dimension value and the mean, Represents the sum of squares of the normalized standardized differences.

[0030] Preferably, the surface defect rate calculation formula of the industrial product is as follows:

[0031]

[0032] In the formula, Bmqx represents the surface defect rate of industrial products, Jcqx represents the number of detected industrial product defects, and Zjcm represents the total detection area of ​​the industrial product surface.

[0033] Preferably, the color standard deviation calculation formula of the industrial product is as follows:

[0034]

[0035] In the formula, Color std represents the color standard deviation of industrial products, n represents the total number of industrial products measured, c represents the number of color channels, C ij Indicates the measured value of the i-th product in the j-th color channel, Color avg,j represents the value of the jth color channel, It means to sum all color channels of industrial products. Represents the sum of squares of the normalized standardized differences.

[0036] Preferably, the industrial product strength calculation formula is as follows:

[0037]

[0038] In the formula, Gyqd represents the strength of industrial products, Clzy represents the maximum force acting on the industrial product material, and Jmcl represents the cross-sectional area of ​​the industrial product material.

[0039] An industrial product quality inspection system based on artificial intelligence, comprising an industrial product image acquisition unit, an industrial product image preprocessing unit, an industrial product image analysis unit, an industrial product quality inspection unit and an information display unit;

[0040] The industrial product image acquisition unit acquires image data of the industrial product through a high-resolution camera, and sends the image data of the industrial product to the industrial product image preprocessing unit;

[0041] The industrial product image preprocessing unit performs denoising, enhancement and cropping operations on the image data of the industrial product;

[0042] The industrial product image analysis unit calculates the average value of the industrial product size, the standard deviation of the industrial product size, the surface defect rate of the industrial product, the color standard deviation of the industrial product, and the industrial product strength according to the industrial product image after denoising, cropping, and enhancement operations;

[0043] The industrial product quality inspection unit performs qualified numerical analysis on the average value of industrial product size, the standard deviation of industrial product size, the surface defect rate of industrial products, the color standard deviation of industrial products, and the strength of industrial products, comprehensively determines whether the quality of industrial products is qualified, and generates an industrial product quality inspection report;

[0044] The information display unit removes unqualified industrial products and makes the quality inspection report of qualified industrial products into a QR code for later printing on the outer packaging of qualified industrial products for display.

[0045] Compared with the prior art, the present invention provides an industrial product quality detection method and system based on artificial intelligence, which has the following beneficial effects:

[0046] The present invention acquires the overall image data of each industrial product by introducing a high-resolution camera and advanced image processing technology, and uses the automated denoising, enhancement and cropping operations of the industrial product image preprocessing unit to improve the quality of the image data, making the subsequent calculation of parameters such as size, surface defect rate, color standard deviation more accurate and reliable. By performing qualified numerical analysis on these parameters, it is possible to quickly and accurately determine whether the product quality is qualified, which significantly improves the accuracy and consistency of detection. At the same time, the two-dimensional code technology is used to directly link the test report of qualified products with the product packaging, which improves product traceability and consumer trust. By combining the above methods, the efficiency of detection is greatly improved when facing large-scale industrial product production quality detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic diagram of the steps of the method of the present invention;

[0048] Figure 2 It is a schematic diagram of the system flow of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0050] In view of the problem that the traditional manual inspection method relies on the operator's experience and technical level and is easily affected by the subjective factors of the inspectors, resulting in insufficient product quality inspection accuracy and particularly low inspection efficiency when facing large-scale industrial product production quality inspection, a method for industrial product quality inspection based on artificial intelligence is proposed. Figure 1 , the method comprises the following steps:

[0051] S1. Obtain image data of industrial products through high-resolution cameras;

[0052] Acquiring image data of industrial products through high-resolution cameras is an indispensable technical means in modern manufacturing and quality inspection. High-resolution cameras can capture subtle product details and ensure that image clarity reaches the micron level, so as to enable accurate defect detection and analysis. In order to improve the efficiency and accuracy of image acquisition, advanced optical systems and intelligent lighting technologies are usually used, including ring lights, spotlights or LED lamps, to reduce the impact of shadows and reflections.

[0053] S2. De-noising, enhancing and cropping the image data of industrial products;

[0054] Denoising, enhancing and cropping the image data of industrial products are important steps to improve image quality and subsequent analysis accuracy. Through denoising, the application of mean filtering can effectively reduce random noise in the image and retain the main features, so as to more clearly identify subtle defects on the surface of the product. The enhancement operation uses contrast enhancement technology to make the key features in the image more prominent and improve visual recognizability, thereby helping the detection system to more accurately capture defect information or size data. Image cropping can effectively crop out key information and remove unnecessary background noise. This process not only improves computing efficiency, but also ensures accurate analysis of specific areas. The comprehensive use of these technical means can significantly improve the quality and practicality of industrial image data, and provide a more accurate basis for subsequent intelligent analysis and decision-making, including:

[0055] The image denoising formula for industrial products is as follows:

[0056]

[0057] De-noising technology can improve the clarity of an image by eliminating random noise in the image, making it easier to identify important details. This is very important for subsequent detection, especially for images taken in complex environments. In the formula, I clean (x, y) represents the pixel value of the denoised industrial product image at the coordinate (x, y), I(x+i, y+j) represents the pixel value of the original industrial product image at the coordinate (x, y), k represents the size of the filter, represents a constant factor used to normalize the summation result. Represents the accumulation of all pixel values ​​I(x+i,y+j) within the k*k window. Denoising helps filter out unnecessary signals, thereby reducing the risk of false positives, reducing the workload of manual review, and improving production efficiency.

[0058] The image enhancement formula for industrial products is as follows:

[0059] I enhanced (x,y)=α*I clean(x,y)+β

[0060] Image enhancement technology can make product defects and details more obvious, which helps to quickly identify and classify. The enhanced image is more intuitive and easy to understand, which is convenient for operators to judge and make decisions. In the formula, I enhanced (x, y) represents the pixel value of the enhanced industrial product image at the coordinate (x, y), α represents the contrast value of the adjusted image, β represents the offset value of the adjusted image brightness, I clean (x, y) represents the pixel value of the denoised industrial product image at the coordinate (x, y). Under different lighting conditions, the enhancement processing can improve the consistency of the image and ensure that the system can perform accurate detection in various environments;

[0061] The image cropping formula for industrial products is as follows:

[0062] I crop (x,y)=I enhanced (x',y')

[0063] x'=x start +(xx cut )

[0064] y′=y start +(yy cut )

[0065] The cropping operation can automatically remove irrelevant background information, ensuring that the focus of the analysis is on the product itself, making subsequent processing more efficient. In the formula, I crop (x, y) represents the pixel value of the cropped industrial product image at the new position (x, y), (x start ,y start ) represents the coordinates of the upper left corner of the cropping area, (x cut ,y cut ) represents the image cropping boundary of the enhanced industrial product, and (x', y') represents the new position coordinates of the image of the enhanced industrial product. By cropping, the amount of processed data is reduced, the computing pressure is reduced, and thus the processing speed and response time of the system are improved;

[0066] S3, calculating the average value of the industrial product size, the standard deviation of the industrial product size, the surface defect rate of the industrial product, the color standard deviation of the industrial product, and the industrial product strength according to the industrial product image after denoising, cropping, and enhancement operations;

[0067] Among them, the average value of industrial product size is calculated as follows:

[0068]

[0069] The average value provides a general level of product size, which can help companies analyze the consistency and stability of the production process and identify potential production problems. In the formula, Size avg represents the average value of industrial product size, n represents the total number of industrial products measured, L p Indicates the size value of the pth product, p represents the counting subscript, It means that the average value of the industrial product size is obtained by adding up the dimensions of all measured industrial products and then dividing it by the total number of measured industrial products. The average size index calculated by collecting data can be used as a benchmark for quality control and set a reasonable tolerance range for the production line, thereby promoting standardized management;

[0070] The standard deviation of industrial product size is calculated as follows:

[0071]

[0072] The standard deviation measures the discreteness of product size, helping production managers understand the consistency and variability of product size and discover potential quality problems. In the formula, Size std represents the standard deviation of the size of industrial products, n represents the total number of industrial products measured, L p Indicates the size value of the pth product, Size avg Indicates the average value of industrial product size, represents the sum of the squares of the differences between each dimension value and the mean, It represents the sum of squares of normalized standard deviations. By monitoring the trend of standard deviations, production parameters can be adjusted in a timely manner to reduce the generation of defective products and improve overall production efficiency.

[0073] The surface defect rate calculation formula of industrial products is as follows:

[0074]

[0075] The calculation of surface defect rate can directly reflect the quality of the product, and can help enterprises identify quality problems and formulate improvement measures. In the formula, Bmqx represents the surface defect rate of industrial products, Jcqx represents the number of industrial product defects detected, and Zjcm represents the total detection area of ​​the industrial product surface. By timely discovering surface defects, unqualified products can be prevented from entering the market, thereby reducing return and modification costs and improving customer satisfaction.

[0076] The color standard deviation calculation formula for industrial products is as follows:

[0077]

[0078] Color standard deviation can help production managers monitor the consistency of product color, ensure that the color of each batch of products meets expectations, and reduce the production of non-standard color products. In the formula, Color std represents the color standard deviation of industrial products, n represents the total number of industrial products measured, c represents the number of color channels, C ij Indicates the measured value of the i-th product in the j-th color channel, Color avg,j represents the value of the jth color channel, It means to sum all color channels of industrial products. represents the sum of squares of normalized standardized differences;

[0079] The formula for calculating industrial product intensity is as follows:

[0080]

[0081] By calculating and analyzing the strength data, we can ensure that the product meets the safety standards during use and reduce the occurrence of product failures. In the formula, Gyqd represents the strength of industrial products, Clzy represents the maximum force acting on the industrial product material, and Jmcl represents the cross-sectional area of ​​the industrial product material. Strength analysis can help engineers evaluate and select the most suitable materials, and achieve a balance between cost and performance by improving material efficiency and reducing waste.

[0082] S4. Conduct qualified numerical analysis on the average value of industrial product size, standard deviation of industrial product size, surface defect rate of industrial products, color standard deviation of industrial products and strength of industrial products, comprehensively judge whether the quality of industrial products is qualified, and generate industrial product quality inspection report;

[0083] In order to conduct quality inspection on the industrial products produced, it is necessary to give the conditions for qualified inspection, among which:

[0084] The qualified size should be within the design tolerance range, set to 100±1, and the surface defect rate should be controlled below 1%. By sampling and inspecting the products, if the number of defects accounts for more than 1% of the total number of samples, the quality of the batch of products is unqualified. Color consistency is an important indicator to ensure the appearance quality of the product. The color difference analysis of the product is carried out, and the color standard deviation is set not to exceed 2ΔE (CIE Lab color difference unit). Products beyond this range will be considered unqualified. In the strength test, the product should meet or exceed the specified strength standard of 5N / mm 2 , otherwise the product is considered unqualified. If any one of the above conditions is not met, it means that the quality of the industrial product is unqualified, and an industrial product quality inspection report is generated;

[0085] S5. Eliminate unqualified industrial products and make the qualified industrial product quality inspection report into a QR code for later printing on the outer packaging of qualified industrial products;

[0086] After quality inspection of industrial products, the information display unit is used to automatically eliminate unqualified products, and the quality inspection report of qualified industrial products is displayed. The report includes indicators such as product size, strength, and surface defect rate. To ensure convenient dissemination and traceability of information, the unit uses QR code technology to convert the quality inspection report of qualified products into a QR code format. This QR code will be printed on the outer packaging of qualified products. Consumers and related parties only need to scan the QR code with a smartphone to quickly obtain detailed quality information and test results of the product. In addition, the QR code system will be connected to the product traceability platform to ensure that each qualified product can be accurately tracked, improve product transparency and customer trust, and provide data support for subsequent quality analysis and improvement.

[0087] See also Figure 2 , an industrial product quality inspection system based on artificial intelligence, comprising an industrial product image acquisition unit, an industrial product image preprocessing unit, an industrial product image analysis unit, an industrial product quality inspection unit and an information display unit;

[0088] The industrial product image acquisition unit acquires image data of the industrial product through a high-resolution camera, and sends the image data of the industrial product to the industrial product image preprocessing unit;

[0089] The industrial product image preprocessing unit performs denoising, enhancement and cropping operations on the image data of the industrial product;

[0090] The industrial product image analysis unit calculates the average value of the industrial product size, the standard deviation of the industrial product size, the surface defect rate of the industrial product, the color standard deviation of the industrial product, and the industrial product strength according to the industrial product image after denoising, cropping, and enhancement operations;

[0091] The industrial product quality inspection unit conducts qualified numerical analysis on the average value of industrial product size, standard deviation of industrial product size, surface defect rate of industrial products, color standard deviation of industrial products, and strength of industrial products, comprehensively judges whether the quality of industrial products is qualified, and generates industrial product quality inspection reports;

[0092] The information display unit removes unqualified industrial products and makes the qualified industrial product quality inspection report into a QR code for later printing and display on the outer packaging of the qualified industrial products.

[0093] Through the comprehensive application of the above methods and systems, it is possible to quickly and accurately determine whether the product quality is qualified, which significantly improves the accuracy and consistency of detection. At the same time, when facing large-scale industrial product production quality detection, it greatly improves the efficiency of detection.

[0094] Embodiment 1:

[0095] In this experiment, the industrial product image acquisition unit is used to obtain relevant image data of industrial products, and the quality of the industrial products is inspected. The following five sets of industrial product size data are collected:

[0096] Where L1 = 100.4, L2 = 100.6, L3 = 100.5, L4 = 100.2, L5 = 100.3;

[0097] Calculation formula based on the average value of industrial product size:

[0098] Since 100.40 is within the allowable range [99-101], the average value of the size is acceptable;

[0099] According to the standard deviation calculation formula of industrial product size:

[0100] Since the value is less than 1, the size standard deviation test is qualified;

[0101] According to the above calculated values, the average size of the industrial product is qualified, and the size standard deviation test is qualified, so the quality test of the industrial product is qualified;

[0102] Embodiment 2:

[0103] In this experiment, the industrial product image acquisition unit is used to obtain relevant image data of industrial products, and the quality of the industrial products is inspected. The following five sets of industrial product size data are collected:

[0104] Where L1 = 100.5, L2 = 100.7, L3 = 100.6, L4 = 100.4, L5 = 102.5;

[0105] Calculation formula based on the average value of industrial product size:

[0106] Since 101.14 is outside the allowable range [99-101], the average value of the dimension fails;

[0107] According to the standard deviation calculation formula of industrial product size:

[0108] Since the value is less than 1, the size standard deviation test is qualified;

[0109] According to the above calculated values, the average size of the industrial product is unqualified, and the size standard deviation test is qualified. However, due to the existence of one unqualified item, the quality test of the industrial product is unqualified.

[0110] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An industrial product quality detection method based on artificial intelligence, characterized in that: The following steps are involved: S1. Obtain image data of industrial products through high-resolution cameras; S2. De-noising, enhancing and cropping the image data of industrial products; S3, calculating the average value of the industrial product size, the standard deviation of the industrial product size, the surface defect rate of the industrial product, the color standard deviation of the industrial product, and the industrial product strength according to the industrial product image after denoising, cropping, and enhancement operations; S4. Conduct qualified numerical analysis on the average value of industrial product size, standard deviation of industrial product size, surface defect rate of industrial products, color standard deviation of industrial products and strength of industrial products, comprehensively judge whether the quality of industrial products is qualified, and generate industrial product quality inspection report; S5. Eliminate unqualified industrial products and make the qualified industrial product quality inspection report into a QR code for later printing on the outer packaging of qualified industrial products.

2. The method for detecting industrial product quality based on artificial intelligence according to claim 1, characterized in that: The image denoising formula of the industrial product is as follows: In the formula, I clean (x, y) represents the pixel value of the denoised industrial product image at the coordinate (x, y), I(x+i, y+j) represents the pixel value of the original industrial product image at the coordinate (x, y), k represents the size of the filter, represents a constant factor used to normalize the summation result. Represents the accumulation of all pixel values ​​I(x+i,y+j) within the k*k window.

3. The method for detecting industrial product quality based on artificial intelligence according to claim 2, characterized in that: The image enhancement formula of the industrial product is as follows: I enhanced (x,y)=α*I clean (x,y)+β In the formula, I enhanced (x, y) represents the pixel value of the enhanced industrial product image at the coordinate (x, y), α represents the contrast value of the adjusted image, β represents the offset value of the adjusted image brightness, I clean (x,y) represents the pixel value of the denoised industrial product image at the coordinate (x,y).

4. The method for detecting industrial product quality based on artificial intelligence according to claim 3, characterized in that: The image cropping formula of the industrial product is as follows: I crop (x,y)=I enhanced (x’,y’) x’=x start +(x-x cut ) and′=y start +(yy cut ) In the formula, I crop (x, y) represents the pixel value of the cropped industrial product image at the new position (x, y), (x start ,y start ) represents the coordinates of the upper left corner of the cropping area, (x cut ,y cut ) represents the image cropping boundary of the enhanced industrial product, and (x', y') represents the new position coordinates of the image of the enhanced industrial product.

5. The method for detecting industrial product quality based on artificial intelligence according to claim 4, characterized in that: The average value calculation formula of the industrial product size is as follows: In the formula, Size avg represents the average value of industrial product size, n represents the total number of industrial products measured, L p Indicates the size value of the pth product, p represents the counting subscript, It means that the average value of the industrial product size is obtained by adding up the sizes of all measured industrial products and then dividing it by the total number of measured industrial products.

6. The method for detecting industrial product quality based on artificial intelligence according to claim 5, characterized in that: The standard deviation calculation formula of the industrial product size is as follows: In the formula, Size std represents the standard deviation of the size of industrial products, n represents the total number of industrial products measured, L p Indicates the size value of the pth product, Size avg Indicates the average value of industrial product size, represents the sum of the squares of the differences between each dimension value and the mean, Represents the sum of squares of the normalized standardized differences.

7. The method for detecting industrial product quality based on artificial intelligence according to claim 6, characterized in that: The surface defect rate calculation formula of the industrial product is as follows: In the formula, Bmqx represents the surface defect rate of industrial products, Jcqx represents the number of detected industrial product defects, and Zjcm represents the total detection area of ​​the industrial product surface.

8. The method for detecting industrial product quality based on artificial intelligence according to claim 7, characterized in that: The color standard deviation calculation formula of the industrial product is as follows: In the formula, Color std represents the color standard deviation of industrial products, n represents the total number of industrial products measured, c represents the number of color channels, C ij Indicates the measured value of the i-th product in the j-th color channel, Color avg,j represents the value of the jth color channel, It means to sum all color channels of industrial products. Represents the sum of squares of the normalized standardized differences.

9. The method for detecting industrial product quality based on artificial intelligence according to claim 8, characterized in that: The industrial product intensity calculation formula is as follows: In the formula, Gyqd represents the strength of industrial products, Clzy represents the maximum force acting on the industrial product material, and Jmcl represents the cross-sectional area of ​​the industrial product material.

10. An industrial product quality inspection system based on artificial intelligence, characterized in that: It includes an industrial product image acquisition unit, an industrial product image preprocessing unit, an industrial product image analysis unit, an industrial product quality detection unit and an information display unit; The industrial product image acquisition unit acquires image data of the industrial product through a high-resolution camera, and sends the image data of the industrial product to the industrial product image preprocessing unit; The industrial product image preprocessing unit performs denoising, enhancement and cropping operations on the image data of the industrial product; The industrial product image analysis unit calculates the average value of the industrial product size, the standard deviation of the industrial product size, the surface defect rate of the industrial product, the color standard deviation of the industrial product, and the industrial product strength according to the industrial product image after denoising, cropping, and enhancement operations; The industrial product quality inspection unit performs qualified numerical analysis on the average value of industrial product size, the standard deviation of industrial product size, the surface defect rate of industrial products, the color standard deviation of industrial products, and the strength of industrial products, comprehensively determines whether the quality of industrial products is qualified, and generates an industrial product quality inspection report; The information display unit removes unqualified industrial products and makes the quality inspection report of qualified industrial products into a QR code for later printing on the outer packaging of qualified industrial products for display.

Images