A method and system for detecting defects of disassembled components of a waste server

Abstract

The application discloses a kind of waste server disassembly component defect detection method and system, including obtaining the overall image to be identified after server disassembly component;The initial detection block is obtained by quadrature tree division to the image to be identified, and the color distribution of several initial detection blocks is obtained;Color histogram is constructed based on the color distribution of initial detection block, and whether the component in the image to be identified exists defect is determined according to color histogram;If there is defect, the initial detection block of the image to be identified is positioned, the position information of defect feature is determined, and the defect feature type in the initial detection block with defect is extracted.The application uses computer to simulate the visual function of human eye, automatically extracts information from disassembly component image, analyzes and processes with database information, replaces the traditional sorting component method relying on the visual function of human eye, improves waste server terminal component sorting efficiency, reduces labor intensity, realizes the reuse of disassembly component.

Description

Technical Field

[0001] This invention relates to the field of component defect detection technology, and in particular to a method and system for detecting defects in disassembled components from used servers. Background Technology

[0002] With the development of information technology and industrial technology, service terminals are increasingly used in banking services, charging piles, and other fields. Due to technological upgrades and functional enhancements, the rate of equipment obsolescence is accelerating. Inspection is one of the steps in the high-value reuse of parts. Currently, the inspection of damaged or intact parts of discarded service terminals usually relies on manual judgment based on experience and visual observation to determine whether the parts are defective. Moreover, the structure of discarded service terminals is often complex or their internal parts are difficult to observe. The resolution of the human eye is limited and subject to subjective bias, which not only increases sorting time and labor costs but also increases the labor intensity of personnel.

[0003] Therefore, by using machine vision recognition and inspection technology, without damaging the structure and performance of parts, machine vision can automatically complete data collection, processing, feature extraction and recognition, as well as various analyses and calculations, thereby identifying part defects, improving the utilization rate of parts, and realizing the reuse of waste service terminal parts. Summary of the Invention

[0004] In order to overcome the above-mentioned shortcomings of the prior art, the purpose of this invention is to provide a method and system for detecting defects in disassembled components of waste servers.

[0005] The technical solution adopted by this invention to solve its technical problem is: a method and system for detecting defects in disassembled components of a used server, comprising the following steps:

[0006] Obtain the overall image of the components after disassembly on the server side;

[0007] The image to be identified is divided into several initial detection blocks by a quadtree, and the color distribution of the several initial detection blocks is obtained.

[0008] A color histogram is constructed based on the color distribution of the initial detection block, and the presence of defects in the components of the image to be identified is determined based on the color histogram.

[0009] If a defect exists, locate the initial detection block with the defect in the image to be identified, determine the location information of the defect features, and extract the defect feature type from the initial detection block with the defect.

[0010] As a further improvement of the present invention: the step of dividing the image to be identified into a quadtree to obtain a number of initial detection blocks and obtaining the color distribution of the number of initial detection blocks includes: preprocessing the image to be identified, and using an X-ray detector to detect the color distribution of each initial detection block in the preprocessed image to be identified.

[0011] As a further improvement of the present invention: the step of constructing a color histogram based on the color distribution of the initial detection block, and determining whether there are defects in the components in the image to be identified based on the color histogram specifically includes:

[0012] Obtain the comparison results between the color histogram and the standard image histogram in the component image feature database;

[0013] If the comparison result is greater than a preset threshold, it is determined that the component in the image to be identified has a defect.

[0014] As a further improvement of the present invention: if a defect exists, locating the initial detection block containing the defect in the image to be identified and determining the location information of the defect features includes:

[0015] The initial detection block with defects is partitioned into a quadtree to form several blocks to be verified;

[0016] By comparing the color histogram with the standard image histogram in the component image feature database, the block to be verified where the defect exists and the corresponding component are identified.

[0017] As a further improvement of the present invention: the extraction of defect feature types in the initial detection block with defects specifically includes: after determining the block to be verified with defects, extracting the defect features and comparing them with known defect type features in the component image feature database to determine the defect type in the block to be verified, wherein the defect type includes cracks, scratches, holes and deformation.

[0018] As a further improvement of the present invention: the step of dividing the image to be identified into a quadtree to obtain a number of initial detection blocks includes: dividing the images in the component image feature database into quadtree blocks, so that the images in the component image feature database form the same number of blocks as the initial detection blocks.

[0019] As a further improvement of the present invention, it also includes: if no defect exists, deleting the image to be identified and obtaining the next image to be identified for defect detection.

[0020] As a further improvement of the present invention: applied to a component defect detection device, the component defect detection device includes a conveyor belt, a light source, an industrial camera, a stop switch, an X-ray detector, and a signal processing unit. The conveyor belt transports components disassembled from a waste service terminal. The stop switch and the industrial camera are installed in the shooting area. After the disassembled components are transported to the shooting area, the stop switch blocks the components, and the industrial camera simultaneously takes pictures of the disassembled components. The industrial camera is signal-connected to the X-ray detector, and the X-ray detector is signal-connected to the signal processing unit.

[0021] Based on the same inventive concept, this invention also provides a defect detection system for disassembled components of used servers, comprising:

[0022] The acquisition module is used to acquire the overall image of the components to be identified after disassembly on the server side;

[0023] The first processing module is used to divide the image to be identified into a quadtree to obtain a number of initial detection blocks and to obtain the color distribution of the number of initial detection blocks.

[0024] The processing module is used to construct a color histogram based on the color distribution of the initial detection block, and to determine whether there are defects in the components in the image to be identified based on the color histogram;

[0025] The confirmation module, if a defect exists, locates the initial detection block with the defect in the image to be identified, determines the location information of the defect features, and extracts the defect feature type from the initial detection block with the defect.

[0026] As a further improvement of the present invention: the processing module is also used to perform quadtree partitioning on the initial detection block with defects to form several blocks to be verified; and then, based on the comparison between the color histogram and the standard image histogram in the component image feature database, to determine the blocks to be verified where defects exist, and to determine the components corresponding to the defects.

[0027] Compared with the prior art, the beneficial effects of the present invention are:

[0028] This invention acquires an overall image of the components after disassembly at the server end, divides the image into quadtree blocks, uses color histograms to determine if there are defects in the components, confirms the location and type of defect features, and automatically extracts information from the disassembled component images by simulating human visual function using a computer. This information is then analyzed and processed with database information, replacing the traditional method of sorting components based on human visual function. This improves the sorting efficiency of waste service terminal components, reduces labor intensity, and enables the reuse of disassembled components. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating a method for detecting defects in disassembled components of a used server in Example 1.

[0030] Figure 2 This is a schematic diagram of the component defect detection device in Example 1.

[0031] Figure descriptions: 1-Belt conveyor, 2-Disassembled parts, 3-Light source, 4-Stop switch, 5-X-ray detector, 6-Industrial camera, 7-Signal processing unit. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0033] To address the problems of high labor intensity, limited human visual resolution, subjective bias, and low component utilization rates associated with the current method of identifying defects in dismantled parts at waste service terminals, this invention will now be further explained in conjunction with the accompanying drawings and embodiments:

[0034] Example 1

[0035] like Figures 1-2 As shown in Embodiment 1, this invention discloses a method for detecting defects in disassembled components of a used server, comprising the following steps:

[0036] S1: Obtain the overall image of the components to be identified after disassembly on the server side;

[0037] The image to be identified is obtained by taking pictures with an industrial camera. After the waste service terminal is disassembled, all the components are photographed and uploaded. In other words, the image to be identified includes all the components that make up the waste service terminal.

[0038] Since taking photos of individual parts for zero-defect inspection is time-consuming and inefficient, it is better to take photos of all parts as a whole before performing component defect inspection.

[0039] S2: The image to be identified is preprocessed, and the preprocessed image is divided into several initial detection blocks by quadtree partitioning. The preprocessing includes denoising, enhancement, size conversion, orientation conversion, etc., to obtain a clearer image. The image size and orientation conversion is mainly to keep the image size consistent with the image size in the pre-set component image feature database, so as to facilitate quick image comparison and accurate detection of the component defect features corresponding to the image to be identified.

[0040] Since the image to be identified contains images of all the components after the service terminal has been disassembled, in order to specifically detect which component has a defect, the preprocessed image to be identified is divided into quadtree blocks to form initial detection blocks for easy detection. For example, there are 4 initial detection blocks. If the size of the image to be identified is 50*50, then the size of each initial detection block is 25*25.

[0041] More preferably, the image to be identified can be divided into quadtree blocks according to the distribution location and quantity of key components or components prone to defects in the service terminal, forming corresponding initial detection blocks. The initial detection blocks are then compared one by one with standard images in the component image feature database to accurately identify whether there are defects in the components within the initial detection blocks. For example, after the image to be identified is divided into four equal initial detection blocks by quadtree blocks, a second quadtree block is performed on the initial detection blocks containing key components or components prone to defects to perform preliminary defect screening.

[0042] S3: The color distribution of each initial detection block is detected using an X-ray detector on the preprocessed image to be identified;

[0043] Preferably, the images in the component image feature database are also divided into quadtree blocks, so that the images in the component image feature database form the same number of blocks as the initial detection blocks, thereby improving the efficiency of comparing the color distribution of the image to be identified with the images in the component image feature database.

[0044] S4: Construct a color histogram based on the color distribution of the initial detection block, and determine whether the component has defects based on the color histogram;

[0045] Preferably, the step of constructing a color histogram based on the color distribution of the initial detection block, and determining whether the component has defects based on the color histogram, specifically includes:

[0046] Obtain the comparison result between the color histogram and the standard image histogram in the component image feature database; multiple standard image histograms are preset for each component, and the standard image histograms are the color histograms of the component under normal and defect-free conditions.

[0047] The comparison result is compared with a preset threshold to determine whether the component has a defect;

[0048] If the comparison result is greater than the preset threshold, it is determined that the component in the image to be identified has a defect; if the comparison result is not greater than the preset threshold, it is determined that the component in the image to be identified does not have a defect.

[0049] In this implementation, the comparison results include the results of comparing the color histogram with one or more standard image histograms.

[0050] S5: If a defect exists, locate the initial detection block with the defect in the image to be identified, determine the location information of the defect features, and extract the defect feature type from the initial detection block with the defect.

[0051] Preferably, the initial detection block with defects is divided into quadtrees to form several blocks to be verified, and the blocks to be verified with defects are further determined according to the color histogram;

[0052] First, the image to be identified is divided into several initial detection blocks to check whether there are defects in the parts. If there are defects, the specific initial detection block is locked. The recognition area is then gradually narrowed down by using a quadtree block. If there are no defects, the image to be identified is directly deleted.

[0053] S51: After identifying the defective block to be verified, compare the defect features in the block with known component images to locate the position of the defect features within the block. At the same time, extract the defect features and compare them with known defect type features to determine the defect type within the block.

[0054] The component image feature database pre-stores different types of defect features and their corresponding color distribution maps. Each type of defect feature is associated with a cluster color distribution map. These defect features include cracks, scratches, holes, deformations, and other types. Due to the different types and degrees of defects, the component image feature database pre-creates images of various defect degrees for different defect features to facilitate rapid matching during comparison.

[0055] S6: If no defects are found, delete the image to be identified and obtain the next image to be identified for defect detection.

[0056] To further improve the efficiency of acquiring images of the parts to be inspected, this embodiment incorporates a conveyor belt 1, a light source 3, an industrial camera 6, a stop switch 4, and a signal processing unit 7. The light source provides constant illumination to the industrial camera. The stop switch blocks the component 2 as it is being transported. The signal processing unit acquires, processes, and analyzes the component images, and saves the results. The disassembled components of the used service terminal are placed on the conveyor belt, which transports them to the industrial camera's shooting area. The stop switch then stops the component, securing it in place, making it easier for the industrial camera to take pictures.

[0057] The system captures images of component defects using an industrial camera and converts them to a uniform size. The information processing unit then compares the color distribution of the images with a component image feature database to determine the location of defects in the disassembled parts. Simultaneously, it uses image processing methods to establish a component defect recognition network to determine the type of component defect.

[0058] Example 2

[0059] This embodiment provides a defect detection system for disassembled components from waste servers, including:

[0060] The acquisition module is used to acquire the overall image of the components to be identified after disassembly on the server side;

[0061] The first processing module is used to divide the image to be identified into a quadtree to obtain a number of initial detection blocks and to obtain the color distribution of the number of initial detection blocks.

[0062] The processing module is used to construct a color histogram based on the color distribution of the initial detection block, and to determine whether there are defects in the components in the image to be identified based on the color histogram;

[0063] The confirmation module, if a defect exists, locates the initial detection block with the defect in the image to be identified, determines the location information of the defect features, and extracts the defect feature type from the initial detection block with the defect.

[0064] Preferably, the processing module is further configured to perform quadtree partitioning on the initial detection block with defects to form several blocks to be verified; and then, based on the comparison between the color histogram and the standard image histogram in the component image feature database, determine the blocks to be verified where defects exist, and determine the components corresponding to the defects.

[0065] The main functions of this invention are: to rapidly match image information with database information by simulating the visual recognition function of the human eye; and to improve the accuracy of defect judgment by comparing multiple images to identify component defects, ultimately drawing conclusions and providing information. By replacing the traditional method of sorting components relying on human visual function, this invention improves the sorting efficiency of waste service terminals, reduces labor intensity, and enables the reuse of disassembled components.

[0066] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for detecting defects in disassembled components of a used server, characterized in that, Includes the following steps: Obtain the overall image of the components after disassembly on the server side; The image to be identified is divided into several initial detection blocks by a quadtree, and the color distribution of the several initial detection blocks is obtained. A color histogram is constructed based on the color distribution of the initial detection block, and the presence of defects in the components of the image to be identified is determined based on the color histogram. If a defect exists, locate the initial detection block with the defect in the image to be identified, determine the location information of the defect features, and extract the defect feature type from the initial detection block with the defect. If a defect exists, locating the initial detection block containing the defect in the image to be identified, and determining the location information of the defect features, includes: The initial detection block with defects is divided into quadtrees to form several blocks to be verified. Then, based on the comparison between the color histogram and the standard image histogram in the component image feature database, the blocks to be verified with defects and the corresponding components are determined. The extraction of defect feature types from the initial detection block containing defects specifically includes: after determining the block to be verified containing defects, extracting the defect features and comparing them with known defect type features in the component image feature database to determine the defect type within the block to be verified. The defect type includes cracks, scratches, holes, and deformation.

2. The method for defect detection of disassembled components of a used server according to claim 1, characterized in that, The step of dividing the image to be identified into several initial detection blocks by quadtree partitioning and obtaining the color distribution of the several initial detection blocks includes: preprocessing the image to be identified, and using an X-ray detector to detect the color distribution of each initial detection block in the preprocessed image to be identified.

3. The method for defect detection of disassembled components of a waste server according to claim 2, characterized in that, The step of constructing a color histogram based on the color distribution of the initial detection block, and determining whether there are defects in the components of the image to be identified based on the color histogram, specifically includes: Obtain the comparison results between the color histogram and the standard image histogram in the component image feature database; If the comparison result is greater than a preset threshold, it is determined that the component in the image to be identified has a defect.

4. The method for defect detection of disassembled components of a used server according to claim 3, characterized in that, The step of dividing the image to be identified into a quadtree to obtain several initial detection blocks includes: dividing the images in the component image feature database into quadtree blocks, so that the images in the component image feature database form the same number of blocks as the initial detection blocks.

5. The method for defect detection of disassembled components of a used server according to claim 1, characterized in that, It also includes, if no defects are found, deleting the image to be identified and obtaining the next image to be identified for defect detection.

6. A method for detecting defects in disassembled components of a used server according to any one of claims 1-5, characterized in that, An application is made in a component defect detection device, which includes a conveyor belt, a light source, an industrial camera, a stop switch, an X-ray detector, and a signal processing unit. The conveyor belt transports components disassembled from a waste service terminal. The stop switch and the industrial camera are installed in the imaging area. After the disassembled components are transported to the imaging area, the stop switch blocks the components, and the industrial camera simultaneously images the disassembled components. The industrial camera is signal-connected to the X-ray detector, and the X-ray detector is signal-connected to the signal processing unit.

7. A defect detection system for disassembled components of a used server, characterized in that, include: The acquisition module is used to acquire the overall image of the components to be identified after disassembly on the server side; The first processing module is used to divide the image to be identified into a quadtree to obtain a number of initial detection blocks and to obtain the color distribution of the number of initial detection blocks. The processing module is used to construct a color histogram based on the color distribution of the initial detection block, and to determine whether there are defects in the components in the image to be identified based on the color histogram; The confirmation module, if a defect exists, locates the initial detection block with the defect in the image to be identified, determines the location information of the defect features, and extracts the defect feature type from the initial detection block with the defect: The processing module is also used to perform quadtree partitioning on the initial detection block with defects to form several blocks to be verified; and then, based on the comparison between the color histogram and the standard image histogram in the component image feature database, to determine the blocks to be verified where defects exist, and to determine the components corresponding to the defects.

Images