Plastic masterbatch detection system based on image recognition

Abstract

The invention discloses a plastic masterbatch detection system based on image recognition, and relates to the technical field of image processing. The system collects current plastic masterbatch image data, analyzes the brightness distribution of the current plastic masterbatch image data, calculates current light source angle data, collects a number of adjusted plastic masterbatch image data according to the current light source angle data and a shadow area, aligns and fuses the adjusted plastic masterbatch image data, generates plastic masterbatch image data with uniform brightness, performs color difference detection on the plastic masterbatch image data with uniform brightness, and generates color difference detection results. The invention aims at the problem of uneven brightness caused by uneven illumination of plastic masterbatch images, accurately calculates the light source angle and adjusts the light source position through light source analysis and adjustment, collects a number of adjusted images, aligns and fuses the images, generates plastic masterbatch images with uniform brightness, and is fully automated, efficient and accurate, and provides strong support for color difference detection of plastic masterbatch.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to a plastic masterbatch detection system based on image recognition. Background Art

[0002] In the production process of plastic masterbatch, quality control is a crucial link. As the raw material of plastic products, the color uniformity and quality of plastic masterbatch directly affect the appearance and performance of the final product. Traditional plastic masterbatch quality inspection mainly relies on manual visual inspection, which is not only inefficient but also easily affected by human factors, resulting in instability and unreliability of the test results.

[0003] At present, a Chinese invention patent with application number CN202311615599.5 discloses a plastic masterbatch detection method based on image feature analysis, including: obtaining a grayscale image and a brightness image of a plastic masterbatch image; obtaining a reflective area based on the brightness image; obtaining a number of candidate grayscale values ​​in the grayscale image, calculating the possibility that each candidate grayscale value is an upper reference value, and then obtaining the best upper reference value; obtaining the best lower reference value; taking the average of the best upper reference value and the best lower reference value as an initial segmentation threshold; obtaining a number of grayscale segmentation areas based on the initial segmentation threshold, and obtaining an abnormal masterbatch area based on the number of grayscale segmentation areas. Although the invention can improve segmentation efficiency and segmentation accuracy by setting a suitable initial segmentation threshold, it is difficult to compensate for the uneven brightness caused by uneven illumination of the plastic masterbatch image and fuse it to generate an image with uniform brightness, which affects the detection effect. Summary of the invention

[0004] The technical problem solved by the present invention is that it is difficult to compensate for the uneven brightness caused by the uneven illumination of the plastic masterbatch image and fuse it to generate an image with uniform brightness, which affects the detection effect.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] The plastic masterbatch detection system based on image recognition includes an image acquisition module, a light source analysis module, a light source adjustment module, an image fusion module and a color difference recognition module:

[0007] The image acquisition module is used to acquire current plastic masterbatch image data;

[0008] The light source analysis module is used to analyze the brightness distribution of the current plastic masterbatch image data, divide the current plastic masterbatch image data into a shadow area and a light source area according to the brightness distribution, and calculate the current light source angle data;

[0009] The light source adjustment module is used to generate a plurality of adjusted light source position data according to the current light source angle data and the shadow area, input the adjusted light source position data into the servo motor control end, and collect a plurality of adjusted plastic masterbatch image data;

[0010] The image fusion module is used to register and fuse the adjusted plastic masterbatch image data to generate plastic masterbatch image data with uniform brightness;

[0011] The color difference recognition module is used to perform color difference detection on uniform brightness plastic masterbatch image data to generate a color difference detection result.

[0012] Preferably, the image acquisition module includes an image capturing unit and an image storage unit:

[0013] The image primary shooting unit is used to shoot the current plastic masterbatch image and generate first current plastic masterbatch image data;

[0014] The image storage unit is used to store the first current plastic masterbatch image data.

[0015] Preferably, the light source analysis module includes an image processing unit, a shadow detection unit and a light source detection unit:

[0016] The image processing unit is used to convert the current plastic masterbatch image data into a grayscale image and remove noise to generate second current plastic masterbatch image data;

[0017] The shadow detection unit is used to count the number of pixels of each gray level in the second current plastic masterbatch image data, generate a gray histogram, and divide the image into a highlight area, a light source area and a shadow area according to a preset gray threshold.

[0018] Preferably, the light source detection unit is used to calculate the centroid position of the highlight area, and the mathematical expression is: ,

[0019] in, is the horizontal coordinate of the center of gravity of the highlighted area, is the ordinate of the center of gravity of the highlighted area, is the coordinate set of the pixels in the highlighted area, is the pixel value after binarization;

[0020] The horizontal and vertical angles of the light source are calculated based on the center of gravity of the highlight area. The mathematical expression is: ,

[0021] in, is the horizontal angle of the light source, is the vertical angle of the light source, is the vertical distance from the plastic masterbatch to the shooting position, is the pixel horizontal coordinate of the center point of the second current plastic masterbatch image data, The pixel ordinate of the center point of the second current plastic masterbatch image data;

[0022] Calculate the incident angle of the light source based on the horizontal and vertical angles of the light source , the mathematical expression is: ,

[0023] Output the incident angle of the light source as the current light source angle data.

[0024] Preferably, the light source adjustment module includes a position calculation unit, a light source control unit and an image secondary shooting unit:

[0025] The position calculation unit is used to calculate the light source height adjustment amount , the mathematical expression is: ,

[0026] in, is the preset unit height adjustment, is the area of ​​the shaded region, is the preset standard shadow area;

[0027] According to the current light source angle data, an adjusted light source angle data is generated every 5°, and an adjusted light source angle data group is output. The light source height adjustment amount and the adjusted light source angle data are combined and output as adjusted light source position data.

[0028] Preferably, the light source control unit is used to input the light source position adjustment data into the servo motor control end;

[0029] The image secondary shooting unit is used to shoot the image after each light source position adjustment, and output it as adjusted plastic masterbatch image data.

[0030] Preferably, the image fusion module includes an image registration unit and an image fusion unit:

[0031] The image registration unit is used to grayscale the adjusted plastic masterbatch image data, establish a coordinate system, place the second current plastic masterbatch image data in the coordinate system, and transform the edge point pixels of the adjusted plastic masterbatch image data to the edge point pixels of the second current plastic masterbatch image data based on the edge point pixels of the second current plastic masterbatch image data, thereby generating a registered plastic masterbatch image data set;

[0032] The image fusion unit is used to extract the light source area in the plastic masterbatch image data set according to the grayscale threshold and set the brightness weight value , the mathematical expression is: ,

[0033] in, is the pixel brightness value, is the preset width of the control weight distribution;

[0034] Normalize the weight of the light source area, the mathematical expression is: ,

[0035] in, is the normalized weight, For the The image is at pixel location The weight value of

[0036] The light source area is weighted averaged and fused according to the normalized weight of each pixel, and the edge of the light source area is retained using a guided filter to generate plastic masterbatch image data with uniform brightness.

[0037] Preferably, the color difference recognition module includes a distribution analysis unit and a color difference detection unit:

[0038] The distribution analysis unit is used to identify the boundary of each masterbatch in the uniform brightness plastic masterbatch image data using an image segmentation algorithm, generate independent plastic masterbatch image data, and extract the Lab value of the independent plastic masterbatch image data in the Lab color space;

[0039] The color difference detection unit is used to calculate the color difference according to the Lab value of the preset standard color , the mathematical expression is: ,

[0040] in, is the Lab value of the preset standard color, It is the Lab value of independent plastic masterbatch image data;

[0041] Determine the color difference based on the preset standard color difference:

[0042] like If it is greater than the standard color difference, the result is judged as having color difference;

[0043] like If the color difference is less than the standard color difference, the result is judged as no color difference;

[0044] The judgment results are combined and output as the color difference detection result.

[0045] An electronic device comprises a storage device, a processor and a computer program stored in the storage device and executable on the processor, wherein the processor executes the plastic masterbatch detection system based on image recognition.

[0046] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the plastic masterbatch detection system based on image recognition is implemented.

[0047] Beneficial effects of the present invention: The present invention aims at the problem of uneven brightness caused by uneven illumination of plastic masterbatch images. By analyzing and adjusting the light source, the light source angle is accurately calculated and the light source position is adjusted. A plurality of adjusted images are collected, and the images are registered and fused to generate a plastic masterbatch image with uniform brightness. The whole process is automated, efficient and accurate, and provides strong support for the color difference detection of plastic masterbatch. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of the basic flow of a plastic masterbatch detection system based on image recognition provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0050] Example, see Figure 1 , provides a plastic masterbatch detection system based on image recognition, including image acquisition module, light source analysis module, light source adjustment module, image fusion module and color difference recognition module:

[0051] The image acquisition module is used to acquire the current plastic masterbatch image data.

[0052] The light source analysis module is used to analyze the brightness distribution of the current plastic masterbatch image data, divide the current plastic masterbatch image data into a shadow area and a light source area according to the brightness distribution, and calculate the current light source angle data.

[0053] The light source adjustment module is used to generate a number of adjusted light source position data according to the current light source angle data and the shadow area, input the adjusted light source position data to the servo motor control end, and collect a number of adjusted plastic masterbatch image data.

[0054] The image fusion module is used to register and fuse the adjusted plastic masterbatch image data to generate plastic masterbatch image data with uniform brightness.

[0055] The color difference recognition module is used to perform color difference detection on uniform brightness plastic masterbatch image data and generate color difference detection results.

[0056] The image acquisition module includes an image capture unit and an image storage unit:

[0057] The image primary shooting unit is used to shoot the current plastic masterbatch image and generate first current plastic masterbatch image data.

[0058] The image one-shot unit can quickly respond and capture the real-time image of the plastic masterbatch. The generated first current plastic masterbatch image data is stored clearly and rich in details, providing a high-quality foundation for subsequent processing.

[0059] The image storage unit is used to store the first current plastic masterbatch image data.

[0060] The image storage unit can stably and safely store the first current plastic masterbatch image data to prevent loss or damage, facilitate subsequent search, analysis and comparison, and improve work efficiency and accuracy.

[0061] The image acquisition module can efficiently and accurately capture the image information of the current plastic masterbatch and convert it into the first current plastic masterbatch image data that can be used for subsequent processing. Through the coordinated work of the image one-shot unit and the image storage unit, the integrity and traceability of the first current plastic masterbatch image data are ensured.

[0062] The light source analysis module includes an image processing unit, a shadow detection unit, and a light source detection unit:

[0063] The image processing unit is used to convert the current plastic masterbatch image data into a grayscale image and remove noise to generate second current plastic masterbatch image data.

[0064] The image processing unit converts complex color image data into simple grayscale images, reduces processing complexity, removes noise in the image, improves image quality, and provides more accurate second current plastic masterbatch image data for subsequent processing.

[0065] The shadow detection unit is used to count the number of pixels of each gray level in the second current plastic masterbatch image data, generate a gray histogram, and divide the image into a highlight area, a light source area, and a shadow area according to a preset gray threshold.

[0066] The shadow detection unit accurately identifies the highlight area, light source area and shadow area in the image by statistically analyzing the grayscale histogram, and performs segmentation according to the preset grayscale threshold, providing a clear area division for the light source detection unit.

[0067] The light source detection unit is used to calculate the center of gravity of the highlight area. The mathematical expression is: ,

[0068] in, is the horizontal coordinate of the center of gravity of the highlighted area, is the ordinate of the center of gravity of the highlighted area, is the coordinate set of the pixels in the highlighted area, is the pixel value after binarization.

[0069] The horizontal and vertical angles of the light source are calculated based on the center of gravity of the highlight area. The mathematical expression is: ,

[0070] in, is the horizontal angle of the light source, is the vertical angle of the light source, is the vertical distance from the plastic masterbatch to the shooting position, is the pixel horizontal coordinate of the center point of the second current plastic masterbatch image data, It is the pixel ordinate of the center point of the second current plastic masterbatch image data.

[0071] Calculate the incident angle of the light source based on the horizontal and vertical angles of the light source , the mathematical expression is: ,

[0072] Output the incident angle of the light source as the current light source angle data.

[0073] The light source detection unit accurately calculates the center of gravity of the highlight area, providing an accurate reference point for the calculation of the light source angle. By calculating the horizontal angle, vertical angle and incident angle, it outputs accurate current light source angle data, providing strong support for subsequent light source adjustments.

[0074] The light source analysis module can deeply analyze the light source conditions in the plastic masterbatch image data, accurately calculate the light source angle, and provide key data for subsequent light source adjustment. The coordinated work of the image processing unit, shadow detection unit, and light source detection unit provides a reliable foundation for subsequent image processing and color difference detection.

[0075] The light source adjustment module includes a position calculation unit, a light source control unit and an image secondary shooting unit:

[0076] The position calculation unit is used to calculate the light source height adjustment amount , the mathematical expression is: ,

[0077] in, is the preset unit height adjustment, is the area of ​​the shaded region, It is the preset standard shaded area.

[0078] According to the current light source angle data, an adjusted light source angle data is generated every 5°, and an adjusted light source angle data group is output. The light source height adjustment amount and the adjusted light source angle data are combined and output as adjusted light source position data.

[0079] The position calculation unit accurately calculates the adjustment amount of the light source height based on the current light source angle data, and generates an adjustment light source angle data every 5° to ensure the meticulousness and accuracy of the light source adjustment. The light source height adjustment amount and the adjustment light source angle data are combined and output as complete adjustment light source position data, providing clear guidance for subsequent light source control.

[0080] The light source control unit is used to input the light source position adjustment data to the servo motor control end.

[0081] The light source control unit accurately inputs the light source position adjustment data to the servo motor control end, realizing precise adjustment of the light source position and angle, with fast response speed and accurate adjustment, ensuring the real-time and effectiveness of the light source adjustment.

[0082] The image secondary shooting unit is used to shoot the image after each light source position adjustment, and outputs the image data of the adjusted plastic masterbatch.

[0083] The image secondary shooting unit will take a new plastic masterbatch image in time after each light source position adjustment. The output adjusted plastic masterbatch image data is of high quality and clear details, providing strong support for subsequent image fusion and color difference detection.

[0084] The light source adjustment module intelligently adjusts the light source position and angle according to the data provided by the light source analysis module to improve the brightness distribution of the plastic masterbatch image, realizes the precise adjustment of the light source and the re-collection of image data, and provides high-quality image data for subsequent image fusion and color difference detection.

[0085] The image fusion module includes an image registration unit and an image fusion unit:

[0086] The image registration unit is used to grayscale the adjusted plastic masterbatch image data, establish a coordinate system, place the second current plastic masterbatch image data in the coordinate system, and use the edge point pixels of the second current plastic masterbatch image data as a reference to transform the edge point pixels of the adjusted plastic masterbatch image data to the edge point pixels of the second current plastic masterbatch image data to generate a registered plastic masterbatch image data set.

[0087] The image registration unit performs grayscale processing on the adjusted plastic masterbatch image data, reduces the processing complexity, improves the registration efficiency, establishes an accurate coordinate system, accurately aligns the second current plastic masterbatch image data with the adjusted plastic masterbatch image data, ensures the accuracy of the registration, and transforms the edge point pixels of the adjusted image based on the edge point pixels of the second current plastic masterbatch image data, thereby achieving accurate registration between images and providing a reliable basis for subsequent fusion.

[0088] The image fusion unit is used to extract the light source area in the plastic masterbatch image dataset according to the grayscale threshold and set the brightness weight value. , the mathematical expression is: ,

[0089] in, is the pixel brightness value, The width of the preset control weight distribution.

[0090] Normalize the weight of the light source area, the mathematical expression is: ,

[0091] in, is the normalized weight, For the The image is at pixel location The weight value of .

[0092] The light source area is weighted averaged and fused according to the normalized weight of each pixel, and the edge of the light source area is retained using a guided filter to generate plastic masterbatch image data with uniform brightness.

[0093] The image fusion unit accurately extracts the light source area in the image according to the grayscale threshold, which provides a basis for the subsequent weight setting. The brightness weight value is set reasonably to ensure that the brightness of the fused image is uniform, avoiding visual problems caused by uneven brightness. The weight of the light source area is normalized to ensure the rationality and effectiveness of the weight. The weighted average fusion method is used in combination with the guided filter to retain the edge details of the light source area, and uniform brightness plastic masterbatch image data with uniform brightness and clear details is generated, providing a high-quality image foundation for subsequent color difference detection.

[0094] The image fusion module achieves accurate registration and high-quality fusion of multiple adjusted plastic masterbatch image data through the coordinated work of image registration and image fusion, generating plastic masterbatch image data with uniform brightness and clear details, effectively solving the brightness difference problem caused by uneven light source in the image, and providing a reliable image basis for subsequent color difference detection.

[0095] The color difference recognition module includes a distribution analysis unit and a color difference detection unit:

[0096] The distribution analysis unit is used to identify the boundary of each masterbatch in the uniform brightness plastic masterbatch image data by using an image segmentation algorithm, generate independent plastic masterbatch image data, and extract the Lab value of the independent plastic masterbatch image data in the Lab color space.

[0097] The distribution analysis unit uses advanced image segmentation algorithms to accurately identify the boundaries of each masterbatch in the image of uniform brightness plastic masterbatch, ensuring the accuracy of subsequent analysis. After successfully segmenting the image, the unit extracts the image data of independent plastic masterbatch and converts it to Lab color space for subsequent color analysis, which helps to speed up the entire color difference identification process. By extracting Lab values, the distribution analysis unit provides a clear color feature description for each plastic masterbatch, providing a reliable data basis for subsequent color difference detection.

[0098] The color difference detection unit is used to calculate the color difference based on the Lab value of the preset standard color , the mathematical expression is: ,

[0099] in, is the Lab value of the preset standard color, It is the Lab value of independent plastic masterbatch image data.

[0100] Determine the color difference based on the preset standard color difference:

[0101] like If it is greater than the standard color difference, the result is judged as having color difference.

[0102] like If the color difference is smaller than the standard color difference, the result is judged as no color difference.

[0103] The judgment results are combined and output as the color difference detection result.

[0104] The color difference detection unit can accurately calculate the color difference of each plastic masterbatch based on the preset Lab value of the standard color, ensuring the objectivity and accuracy of color difference detection. Through the preset standard color difference threshold, it can automatically determine whether each plastic masterbatch has color difference, which not only improves the detection efficiency, but also reduces the errors caused by human factors. The standard color and color difference threshold are adjusted according to different application requirements to adapt to different color difference detection scenarios, and can be effective in a variety of application occasions.

[0105] The color difference recognition module can identify the color difference in the image of plastic masterbatch with uniform brightness. By comprehensively using image segmentation algorithm and color space conversion technology, it analyzes the color characteristics of each plastic masterbatch and compares it with the preset standard color to determine whether there is color difference. This not only improves the detection efficiency, but also ensures the objectivity and accuracy of the detection results.

[0106] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. Among them, the storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0107] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. The plastic masterbatch detection system based on image recognition is characterized by: Including image acquisition module, light source analysis module, light source adjustment module, image fusion module and color difference recognition module: The image acquisition module is used to acquire current plastic masterbatch image data; The light source analysis module is used to analyze the brightness distribution of the current plastic masterbatch image data, divide the current plastic masterbatch image data into a shadow area and a light source area according to the brightness distribution, and calculate the current light source angle data; The light source adjustment module is used to generate a plurality of adjusted light source position data according to the current light source angle data and the shadow area, input the adjusted light source position data into the servo motor control end, and collect a plurality of adjusted plastic masterbatch image data; The image fusion module is used to register and fuse the adjusted plastic masterbatch image data to generate plastic masterbatch image data with uniform brightness; The color difference recognition module is used to perform color difference detection on the uniform brightness plastic masterbatch image data and generate a color difference detection result; The light source analysis module includes an image processing unit, a shadow detection unit and a light source detection unit: The image processing unit is used to convert the current plastic masterbatch image data into a grayscale image and remove noise to generate second current plastic masterbatch image data; The shadow detection unit is used to count the number of pixels of each gray level in the second current plastic masterbatch image data, generate a gray histogram, and divide the image into a highlight area, a light source area, and a shadow area according to a preset gray threshold; The light source detection unit is used to calculate the center of gravity position of the highlight area, and the mathematical expression is: ， in, is the horizontal coordinate of the center of gravity of the highlighted area, is the ordinate of the center of gravity of the highlighted area, is the coordinate set of the pixels in the highlighted area, is the pixel value after binarization; The horizontal and vertical angles of the light source are calculated based on the center of gravity of the highlight area. The mathematical expression is: ， in, is the horizontal angle of the light source, is the vertical angle of the light source, is the vertical distance from the plastic masterbatch to the shooting position, is the pixel horizontal coordinate of the center point of the second current plastic masterbatch image data, The pixel ordinate of the center point of the second current plastic masterbatch image data; Calculate the incident angle of the light source based on the horizontal and vertical angles of the light source , the mathematical expression is: ， Output the incident angle of the light source as the current light source angle data, The light source adjustment module includes a position calculation unit, a light source control unit and an image secondary shooting unit: The position calculation unit is used to calculate the light source height adjustment amount , the mathematical expression is: ， in, is the preset unit height adjustment, is the area of ​​the shaded region, is the preset standard shadow area; According to the current light source angle data, an adjusted light source angle data is generated every 5°, and an adjusted light source angle data group is output, and the light source height adjustment amount and the adjusted light source angle data are combined and output as adjusted light source position data; The light source control unit is used to input the light source position adjustment data into the servo motor control end; The image secondary shooting unit is used to shoot the image after each light source position adjustment, and output it as the adjusted plastic masterbatch image data; The image fusion module includes an image registration unit and an image fusion unit: The image registration unit is used to grayscale the adjusted plastic masterbatch image data, establish a coordinate system, place the second current plastic masterbatch image data in the coordinate system, and transform the edge point pixels of the adjusted plastic masterbatch image data to the edge point pixels of the second current plastic masterbatch image data based on the edge point pixels of the second current plastic masterbatch image data, thereby generating a registered plastic masterbatch image data set; The image fusion unit is used to extract the light source area in the plastic masterbatch image data set according to the grayscale threshold and set the brightness weight value , the mathematical expression is: ， in, is the pixel brightness value, is the preset width of the control weight distribution; Normalize the weight of the light source area, the mathematical expression is: ， in, is the normalized weight, For the The image is at pixel location The weight value of The light source area is weighted averaged and fused according to the normalized weight of each pixel, and the edge of the light source area is retained using a guided filter to generate plastic masterbatch image data with uniform brightness.

2. The plastic masterbatch detection system based on image recognition according to claim 1, characterized in that: The image acquisition module includes an image capturing unit and an image storage unit: The image primary shooting unit is used to shoot the current plastic masterbatch image and generate first current plastic masterbatch image data; The image storage unit is used to store the first current plastic masterbatch image data.

3. The plastic masterbatch detection system based on image recognition as claimed in claim 2, characterized in that: The color difference recognition module includes a distribution analysis unit and a color difference detection unit: The distribution analysis unit is used to identify the boundary of each masterbatch in the uniform brightness plastic masterbatch image data using an image segmentation algorithm, generate independent plastic masterbatch image data, and extract the Lab value of the independent plastic masterbatch image data in the Lab color space; The color difference detection unit is used to calculate the color difference according to the Lab value of the preset standard color , the mathematical expression is: ， in, is the Lab value of the preset standard color, It is the Lab value of independent plastic masterbatch image data; Determine the color difference based on the preset standard color difference: like If it is greater than the standard color difference, the result is judged as having color difference; like If the color difference is less than the standard color difference, the result is judged as no color difference; The judgment results are combined and output as the color difference detection result.

4. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the plastic masterbatch detection system based on image recognition according to any one of claims 1 to 3 is implemented.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the plastic masterbatch detection system based on image recognition according to any one of claims 1 to 3 is implemented.

Images