Optimization processing method and system for industrial product image

Abstract

The invention provides an optimization processing method and system for an industrial product image, and the method comprises an image preprocessing step: collecting and preprocessing an original industrial product image, and outputting the preprocessed image; the preprocessing comprises the steps of detecting the image quality of the original industrial product image, and if the image quality is lower than a preset value, performing intelligent enhancement; an image analyzing and processing step: analyzing the preprocessed image, further processing according to an analysis result, and outputting the processed image; the analysis result comprises whether a shielding object exists in the image or not and whether an image main body proportion exceeds a preset value or not. By combining advanced computer vision and intelligent image processing technologies, the method can automatically identify and remove image backgrounds, accurately identify key features of industrial products, and perform intelligent completion and restoration of the images, thereby optimizing the processing flow of the industrial product images, and improving the quality and consistency of the images.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a method and system for optimizing processing of industrial product images, and in particular to a method and system for intelligently enhancing, removing backgrounds, and completing industrial product images. Background Art

[0002] With the rapid development of e-commerce procurement and digital marketing, high-quality product images are becoming increasingly important in industrial product display and user decision-making. However, industrial product image processing generally faces the following challenges:

[0003] First, industrial products vary in shape, material, and color. Metal products, in particular, have significant reflective properties, often resulting in the reflection and background colors being identical. This makes it difficult for automated image processing tools on the market to meet these needs. Second, due to limitations in photography technology and equipment at many industrial companies, images are low-resolution and of poor quality. Requiring suppliers to provide higher-quality images is sometimes difficult to achieve, leading to an urgent need for intelligent image quality improvements within the industrial products industry. Third, industrial products are typically large and difficult to transport, requiring photography to be conducted in complex environments such as workshops. This results in cluttered backgrounds and places higher demands on background cutouts. Finally, industrial products are often surrounded by obstructions, which complicates image processing. These obstructions may have similar colors and textures to the product itself, making it difficult for traditional methods to remove these obstructions without compromising the integrity of the product image. Furthermore, completing the product image requires manual processing by more skilled designers, which is time-consuming.

[0004] These factors make traditional manual image processing methods inefficient and subject to subjective factors such as the operator's experience, making consistent processing results difficult. This is especially true when processing images with occlusions. Traditional methods often require manufacturers to provide additional materials or rely on other reference images for manual synthesis, which not only increases processing difficulty but also reduces efficiency.

[0005] Patent document CN111415339A discloses a method for detecting defects in images of industrial products with complex textures. The method can match the image of the industrial product to be inspected with the normal industrial product images in a preset database based on a mean-aware hashing algorithm, input the obtained target image into a preset convolutional neural network, and output the detection results of the image of the industrial product to be inspected. However, the method is essentially different from the technical method adopted in this application.

[0006] Therefore, the market urgently needs a system that can automatically remove complex backgrounds, accurately identify specific elements, and intelligently complete industrial product images to improve the efficiency and quality of image processing.

[0007] This invention aims to provide an efficient and automated solution by combining advanced computer vision and intelligent image processing technology to significantly improve the processing effect of industrial product images, ensure image clarity and consistency, and thus meet market demand. Summary of the Invention

[0008] In view of the defects in the prior art, the object of the present invention is to provide a method and system for optimizing the processing of industrial product images.

[0009] According to the present invention, a method for optimizing and processing industrial product images is provided, comprising:

[0010] Image preprocessing step: collect original industrial product images and preprocess them, and output the preprocessed images;

[0011] The pre-processing includes detecting the image quality of the original industrial product image, and performing intelligent enhancement if the image quality is lower than a preset value;

[0012] Image analysis and processing step: analyzing the pre-processed image, further processing it according to the analysis results, and outputting the processed image; the analysis results include whether there are any obstructions in the image and whether the proportion of the image subject exceeds a preset value.

[0013] Preferably, it also includes:

[0014] Machine learning steps: Store the output images in a learning database, train and optimize the model based on the machine learning algorithm and database, and use the optimized model to automatically analyze and process the original industrial product images.

[0015] Preferably, the intelligent enhancement includes using an intelligent image enhancement algorithm to adaptively adjust image contrast, brightness and sharpness, combined with machine learning technology to learn enhancement patterns from other images, and improve the details and texture performance of original industrial product images whose image quality is lower than a preset value.

[0016] Preferably, the image analysis and processing step includes:

[0017] Occlusion detection and processing: Use image recognition algorithms to assist users in automatically detecting and identifying occlusions in images. After user confirmation, an intelligent algorithm is used to automatically remove the occlusions. The trained intelligent generation model is used to complete the image restoration by completing the area where the occlusions were removed.

[0018] Background detection and removal: Use automated image segmentation technology to automatically segment the outline of industrial products. Use the original image provided by the manufacturer to train the model, and the model automatically removes background elements that do not meet the preset values. The automated image segmentation technology includes edge detection and region growing algorithms.

[0019] Preferably, the automated image segmentation technology separates the target product by analyzing the edge and color features of the original industrial product image, and is capable of identifying the shapes and materials of a variety of industrial products.

[0020] According to the present invention, a system for optimizing and processing industrial product images is provided, comprising:

[0021] Image preprocessing module: collects original industrial product images, performs preprocessing, and outputs preprocessed images;

[0022] The pre-processing includes detecting the image quality of the original industrial product image, and performing intelligent enhancement if the image quality is lower than a preset value;

[0023] Image analysis and processing module: analyzes the pre-processed image, further processes it according to the analysis results, and outputs the processed image; the analysis results include whether there are any obstructions in the image and whether the proportion of the image subject exceeds a preset value.

[0024] Preferably, it also includes:

[0025] Machine learning module: The output images are stored in a learning database, the model is trained and optimized based on the machine learning algorithm and database, and the optimized model is used to automatically analyze and process the original industrial product images.

[0026] Preferably, the intelligent enhancement includes using an intelligent image enhancement algorithm to adaptively adjust image contrast, brightness and sharpness, combined with machine learning technology to learn enhancement patterns from other images, and improve the details and texture performance of original industrial product images whose image quality is lower than a preset value.

[0027] Preferably, the image analysis and processing module includes:

[0028] Occlusion detection and processing: Use image recognition algorithms to assist users in automatically detecting and identifying occlusions in images. After user confirmation, an intelligent algorithm is used to automatically remove the occlusions. The trained intelligent generation model is used to complete the image restoration by completing the area where the occlusions were removed.

[0029] Background detection and removal: Use automated image segmentation technology to automatically segment the outline of industrial products. Use the original image provided by the manufacturer to train the model, and the model automatically removes background elements that do not meet the preset values. The automated image segmentation technology includes edge detection and region growing algorithms.

[0030] Preferably, the automated image segmentation technology separates the target product by analyzing the edge and color features of the original industrial product image, and is capable of identifying the shapes and materials of a variety of industrial products.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. By combining advanced computer vision and intelligent image processing technologies, the present invention can automatically identify and remove image backgrounds, accurately identify key features of industrial products, and intelligently complete and repair images, thereby optimizing the processing flow of industrial product images and improving image quality and consistency. The provided system is suitable for e-commerce platforms and product displays, providing an efficient and automated solution for image processing for industrial product companies and has good practicality.

[0033] 2. The present invention solves the problems of low pixels and low quality of industrial product images caused by equipment limitations and environmental conditions during the shooting process. It can significantly improve the clarity and detail of the image, and meet the market demand for high-quality industrial product images.

[0034] 3. The present invention provides a customized data template. In order to meet the special needs of different suppliers, the data acquisition module allows suppliers to customize the data template according to their own attribute descriptions and data requirements, making data import more flexible.

[0035] 4. The present invention uses an intelligent image enhancement algorithm to enhance the quality of industrial product images, providing enterprises with more competitive product display effects in e-commerce platforms and digital marketing; by automatically extracting images and removing background parts that do not meet the requirements, it solves the problem of complex backgrounds interfering with product images, simplifies the image processing process, and reduces manual time and costs; uses image processing methods to eliminate specific objects in the image, effectively solving the common occlusion problem in industrial product images taken in complex environments, and ensuring the integrity and professionalism of product display.

[0036] This method can not only effectively improve the quality of industrial product images, but also provide enterprises with more competitive product display effects in e-commerce platforms and digital marketing.

[0037] Other beneficial effects of the present invention will be explained through the introduction of specific technical features and technical solutions in the specific implementation methods. Those skilled in the art should be able to understand the beneficial technical effects brought about by the introduction of these technical features and technical solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0039] Figure 1 Flow chart of the method of the present invention.

[0040] Figure 2 Flowchart in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the scope of the present invention. These all fall within the scope of protection of the present invention.

[0042] Reference Figure 1 As shown, a method for optimizing industrial product images includes:

[0043] Image preprocessing step: collect original industrial product images and preprocess them, and output the preprocessed images;

[0044] The pre-processing includes detecting the image quality of the original industrial product image, and performing intelligent enhancement if the image quality is lower than a preset value.

[0045] This intelligent industrial product image quality enhancement method aims to address the low pixel count and quality issues often associated with capturing industrial product images due to equipment limitations and environmental conditions. This method significantly improves image clarity and detail, meeting market demand for high-quality industrial product images. It also utilizes customized data templates. To accommodate the unique needs of different suppliers, the data acquisition module allows suppliers to customize data templates based on their own attribute descriptions and data requirements, making data import more flexible.

[0046] in:

[0047] The intelligent image enhancement algorithm automatically analyzes and identifies low-quality areas in images. It adaptively adjusts image contrast, brightness, and sharpness to enhance the overall visual quality. Combined with machine learning, it learns enhancement patterns from a large number of high-quality images, intelligently improving the details and textures of low-quality images.

[0048] The image enhancement techniques used here include:

[0049] (1) Super-resolution reconstruction: Use deep learning models to perform super-resolution reconstruction on low-resolution images to improve the pixel quality of the image. The process is as follows:

[0050] Goal: Generate high-resolution images from low-resolution images and improve the pixel quality of the images.

[0051] Basic formula: I HR =f(I LR ,s)

[0052] Where:

[0053] f is the super-resolution model

[0054] S is the scaling factor

[0055] Loss function:

[0056] Mean Square Error:

[0057]

[0058] Perceptual loss:

[0059]

[0060] Training and Evaluation:

[0061] 1. Data preparation: Generate low-resolution images from high-resolution images to form a training dataset.

[0062] 2. Model training: Use optimization algorithms to minimize the loss function and thus adjust the model parameters.

[0063] 3. Model evaluation: Use the test dataset to evaluate the model performance. Common indicators include peak signal-to-noise ratio and structural similarity index.

[0064] (2) Noise elimination: Use adaptive filtering technology to remove image noise and enhance image clarity.

[0065] Super-resolution processing utilizes deep learning models for super-resolution reconstruction, converting low-pixel images into high-resolution images. This process improves image clarity by reconstructing edges and details, and automatically adapts to output requirements of varying resolutions, ensuring the output image is no less than 1280*1280 and less than 1MB in size.

[0066] This method can not only effectively improve the quality of industrial product images, but also provide enterprises with more competitive product display effects in e-commerce platforms and digital marketing.

[0067] Image analysis and processing step: analyzing the pre-processed image, further processing it according to the analysis results, and outputting the processed image; the analysis results include whether there are any obstructions in the image and whether the proportion of the image subject exceeds a preset value.

[0068] The computer vision techniques used here include:

[0069] (1) Image segmentation technology: Using edge detection and region growing algorithms, the outline of industrial products is automatically segmented to ensure accurate removal of complex backgrounds.

[0070] (2) Object detection technology: Utilize feature extraction and classification algorithms to identify specific elements in an image and improve recognition accuracy.

[0071] This method for extracting and automatically removing backgrounds from industrial product images aims to address the impact of complex backgrounds on product display. This method efficiently and accurately extracts industrial product images and removes backgrounds, enhancing the professionalism of product displays.

[0072] Automated image segmentation technology can automatically identify and segment the outlines of industrial products. By analyzing the edge and color features of the image, this technology can accurately separate the target product from a complex background. It supports a variety of industrial product shapes and materials, adapts to different product characteristics, and achieves high-precision image segmentation. Specifically, it includes:

[0073] Preprocessing:

[0074] Grayscale conversion: Convert color images to grayscale images to simplify image processing.

[0075] Noise removal: Use methods such as Gaussian blur to remove noise from the image and improve the accuracy of edge detection.

[0076] Edge Detection:

[0077] Identify edges in an image using algorithms such as edge detection. These algorithms detect edges by calculating the gradient of the image.

[0078] Image Segmentation:

[0079] Threshold segmentation: Set a threshold based on pixel intensity to segment the image into foreground (industrial products) and background.

[0080] Watershed algorithm: Use the gradient information of the image to divide the image into different areas.

[0081] Contour detection: Contour detection extracts the contours of industrial products.

[0082] Background removal and optimization technology is trained on a large number of original images provided by manufacturers, especially strengthening learning and identifying various production scenes such as workshops, factories, machine tools, etc., to automatically remove background elements outside the product.

[0083] This method solves the problem of complex background interference on product images, which not only simplifies the image processing process but also reduces manual time and cost.

[0084] This method, called "specific object removal" and "intelligent completion" for industrial product images, aims to address the problem of obstructions and unnecessary elements affecting product display in industrial product images. This method efficiently removes specific objects from images and intelligently completes obscured or missing portions, improving image integrity.

[0085] in:

[0086] Occlusion processing technology uses advanced image recognition algorithms to automatically detect and identify specific objects in an image. Users manually select and confirm these objects, and then intelligent removal technology removes these unnecessary objects, ensuring a clear and focused product image.

[0087] Intelligent completion technology learns from a large amount of industrial product image data and intelligently generates occluded or missing parts, restoring the integrity of the image and ensuring that the completed image is natural and consistent.

[0088] This method effectively solves the common occlusion and background interference problems in industrial product images taken in complex environments, ensuring the integrity and professionalism of product display.

[0089] The deep learning model used is able to:

[0090] (1) Feature extraction network: used to extract and classify image features to achieve accurate recognition of industrial product features.

[0091] (2) Intelligent generation model: used for image completion, ensuring the integrity of the image by intelligently generating occluded areas.

[0092] At the same time, an automated image processing pipeline was used, including:

[0093] (1) Batch processing system: Design an automated processing pipeline to support the rapid processing of large batches of images and improve overall efficiency.

[0094] (2) Cloud computing services: Utilize cloud computing resources to provide scalable processing capabilities and support large-scale image processing needs.

[0095] Machine learning steps: Store the output images in a learning database, train and optimize the model based on the machine learning algorithm and database, and use the optimized model to automatically analyze and process the original industrial product images.

[0096] Based on intelligent algorithm optimization technology, machine learning optimization improves the system's adaptability to different types of industrial product images by continuously training and optimizing algorithms.

[0097] At the same time, the user interface and interactive design technology of the present invention provide visualization tools and a user-friendly interface; users can change the image position according to their own requirements based on the processed image, and support real-time preview, batch processing and saving.

[0098] Through the comprehensive application of these technologies, the present invention can effectively solve various complex problems in industrial product image processing, significantly improve image quality and processing efficiency, and meet the market demand for high-quality industrial product images.

[0099] The above is a basic embodiment of the present invention. The technical solution of the present invention is further described below through a preferred embodiment.

[0100] Example 1

[0101] Reference Figure 2 As shown in FIG, a method for intelligent enhancement, background removal, and completion of industrial product images is provided. The specific steps are as follows:

[0102] 1. Original image upload / synchronization: The system supports users to upload or synchronize original industrial product images in multiple formats.

[0103] 2. Image Quality Assessment and Enhancement: First, the quality of the uploaded image is assessed. If the image quality falls below the preset standard (700*700), it will be determined to require intelligent enhancement. The intelligent enhancement module applies an image enhancement algorithm to improve the image's resolution and clarity (1280*1280) to meet display requirements.

[0104] 3. Obstruction Detection and Processing: After image enhancement, the system determines whether any obstructions are present on the product. If an obstruction is detected, the system prompts the user to manually verify and confirm the obstruction needs to be removed. After user confirmation, the system automatically removes the obstruction using intelligent algorithms, ensuring the integrity and clarity of the product image.

[0105] 4. Background Detection and Removal: The system further checks whether the main image portion exceeds 40% and determines whether the image background is white. If the background does not meet the standard, the system will determine that the image needs background removal. The background removal module can accurately cut out the product image.

[0106] Background detection and removal process

[0107] Subject ratio detection:

[0108] Calculate the subject ratio: Use image segmentation technology to determine the pixel area of ​​the subject in the image.

[0109] Judgment criteria: If the subject accounts for less than 40%, the background needs to be removed.

[0110] Background color detection:

[0111] Detect background color: Analyze the background color to determine whether it is close to white (RGB close to 255,255,255).

[0112] Judgment criteria: If the background is not white, it needs to be removed or replaced.

[0113] Background removal:

[0114] Apply Mask: Generate a mask using the segmentation result to keep only the subject.

[0115] Remove Background: Make the background transparent or replace it with standard white.

[0116] Output optimization:

[0117] Edge Smoothing: Smooth edges to avoid jagged edges.

[0118] Color Correction: Adjust the color and brightness of your subject so it blends naturally into its new background.

[0119] Judgment results

[0120] The subject accounts for less than 40%: the background needs to be removed to highlight the subject.

[0121] Background is not white: Needs to be removed or replaced with a white background to comply with standards.

[0122] 5. Finished product image generation and storage: After a series of processing, the generated image will be stored in the designated learning database. In this step, manual adjustments to the product position in the image are allowed to generate a more appropriate finished product image.

[0123] Machine Learning and Model Optimization: After each processing step, the system feeds the results back to the machine learning module for continuous learning and optimization. By continuously accumulating data and experience, the system can gradually improve processing accuracy and efficiency, adapting to more diverse image processing needs.

[0124] By combining advanced computer vision and intelligent image processing technologies, the present invention can automatically identify and remove image backgrounds, accurately identify key features of industrial products, and intelligently complete and repair images, thereby optimizing the processing flow of industrial product images and improving image quality and consistency. The provided system is suitable for e-commerce platforms and product displays, and provides an efficient and automated solution for image processing for industrial product companies, with good practicality.

[0125] The present invention also provides an optimization processing system for industrial product images. The optimization processing system for industrial product images can be implemented by executing the process steps of the optimization processing method for industrial product images, that is, those skilled in the art can understand the optimization processing method for industrial product images as a preferred implementation of the optimization processing system for industrial product images.

[0126] Specifically, a system for optimizing and processing industrial product images includes:

[0127] Image preprocessing module: collects original industrial product images, performs preprocessing, and outputs preprocessed images;

[0128] The pre-processing includes detecting the image quality of the original industrial product image, and performing intelligent enhancement if the image quality is lower than a preset value;

[0129] Image analysis and processing module: analyzes the pre-processed image, further processes it according to the analysis results, and outputs the processed image; the analysis results include whether there are any obstructions in the image and whether the proportion of the image subject exceeds a preset value.

[0130] Also includes:

[0131] Machine learning module: The output images are stored in a learning database, the model is trained and optimized based on the machine learning algorithm and database, and the optimized model is used to automatically analyze and process the original industrial product images.

[0132] The intelligent enhancement includes using an intelligent image enhancement algorithm to adaptively adjust image contrast, brightness and sharpness, and combining machine learning technology to learn enhancement patterns from other images to improve the details and texture performance of original industrial product images whose image quality is lower than a preset value.

[0133] The image analysis and processing module includes:

[0134] Occlusion detection and processing: Use image recognition algorithms to assist users in automatically detecting and identifying occlusions in images. After user confirmation, an intelligent algorithm is used to automatically remove the occlusions. The trained intelligent generation model is used to complete the image restoration by completing the area where the occlusions were removed.

[0135] Background detection and removal: Use automated image segmentation technology to automatically segment the outline of industrial products. Use the original image provided by the manufacturer to train the model, and the model automatically removes background elements that do not meet the preset values. The automated image segmentation technology includes edge detection and region growing algorithms.

[0136] Automated image segmentation technology separates the target product by analyzing the edge and color features of the original industrial product image, and can recognize the shape and material of a variety of industrial products.

[0137] Those skilled in the art will appreciate that, in addition to implementing the system and its various devices, modules, and units provided by the present invention in purely computer-readable program code, it is entirely possible to implement the same functions of the system and its various devices, modules, and units provided by the present invention in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; the devices, modules, and units for implementing various functions can also be considered as both software modules implementing the method and structures within the hardware component.

[0138] The above describes specific embodiments of the present invention. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art may make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. The embodiments of this application and the features in the embodiments may be combined with each other in any manner unless there is a conflict.

Claims

1. A method for optimizing industrial product images, characterized in that: include: Image preprocessing step: collect original industrial product images and preprocess them, and output the preprocessed images; The pre-processing includes detecting the image quality of the original industrial product image, and performing intelligent enhancement if the image quality is lower than a preset value; Image analysis and processing step: analyzing the pre-processed image, further processing it according to the analysis results, and outputting the processed image; the analysis results include whether there are any obstructions in the image and whether the proportion of the image subject exceeds a preset value.

2. The method for optimizing industrial product images according to claim 1, characterized in that: Also includes: Machine learning steps: Store the output images in a learning database, train and optimize the model based on the machine learning algorithm and database, and use the optimized model to automatically analyze and process the original industrial product images.

3. The method for optimizing industrial product images according to claim 1, wherein: The intelligent enhancement includes using an intelligent image enhancement algorithm to adaptively adjust image contrast, brightness and sharpness, and combining machine learning technology to learn enhancement patterns from other images to improve the details and texture performance of original industrial product images whose image quality is lower than a preset value.

4. The method for optimizing industrial product images according to claim 1, wherein: The image analysis and processing step comprises: Occlusion detection and processing: Use image recognition algorithms to assist users in automatically detecting and identifying occlusions in images. After user confirmation, an intelligent algorithm is used to automatically remove the occlusions. The trained intelligent generation model is used to complete the image restoration by completing the area where the occlusions were removed. Background detection and removal: Use automated image segmentation technology to automatically segment the outline of industrial products. Use the original image provided by the manufacturer to train the model, and the model automatically removes background elements that do not meet the preset values. The automated image segmentation technology includes edge detection and region growing algorithms.

5. The method for optimizing industrial product images according to claim 4, characterized in that: Automated image segmentation technology separates the target product by analyzing the edge and color features of the original industrial product image, and can recognize the shape and material of a variety of industrial products.

6. A system for optimizing and processing industrial product images, characterized in that: include: Image preprocessing module: collects original industrial product images, performs preprocessing, and outputs preprocessed images; The pre-processing includes detecting the image quality of the original industrial product image, and performing intelligent enhancement if the image quality is lower than a preset value; Image analysis and processing module: analyzes the pre-processed image, further processes it according to the analysis results, and outputs the processed image; the analysis results include whether there are any obstructions in the image and whether the proportion of the image subject exceeds a preset value.

7. The system for optimizing and processing industrial product images according to claim 6, characterized in that: Also includes: Machine learning module: The output images are stored in a learning database, the model is trained and optimized based on the machine learning algorithm and database, and the optimized model is used to automatically analyze and process the original industrial product images.

8. The system for optimizing and processing industrial product images according to claim 6, characterized in that: The intelligent enhancement includes using an intelligent image enhancement algorithm to adaptively adjust image contrast, brightness and sharpness, and combining machine learning technology to learn enhancement patterns from other images to improve the details and texture performance of original industrial product images whose image quality is lower than a preset value.

9. The system for optimizing and processing industrial product images according to claim 6, characterized in that: The image analysis and processing module includes: Occlusion detection and processing: Use image recognition algorithms to assist users in automatically detecting and identifying occlusions in images. After user confirmation, an intelligent algorithm is used to automatically remove the occlusions. The trained intelligent generation model is used to complete the image restoration by completing the area where the occlusions were removed. Background detection and removal: Use automated image segmentation technology to automatically segment the outline of industrial products. Use the original image provided by the manufacturer to train the model, and the model automatically removes background elements that do not meet the preset values. The automated image segmentation technology includes edge detection and region growing algorithms.

10. The system for optimizing and processing industrial product images according to claim 9, characterized in that: Automated image segmentation technology separates the target product by analyzing the edge and color features of the original industrial product image, and can recognize the shape and material of a variety of industrial products.

Images