Robot and system for detecting defective rate of parts

By designing a part defect rate detection robot system, and using image acquisition and defect detection algorithms to achieve automated detection, the problems of low efficiency and poor accuracy of traditional manual detection are solved, the detection accuracy and production efficiency are improved, and product quality is ensured.

CN119000692AInactive Publication Date: 2024-11-22SUZHOU VOCATIONAL UNIVERSITY (SUZHOU OPEN UNIVERSITY)

Patent Information

Application Number
CN202411215701.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-02
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional quality inspection methods rely on manual visual inspection, which has problems such as low detection efficiency, poor accuracy, poor repeatability and susceptible to human factors, making it difficult to meet the needs of efficient and high-precision quality inspection of modern large-scale and fast production lines.

Method used

A part defect rate detection robot and detection system were designed, including image acquisition module, image processing module, defect detection algorithm module, data analysis module, user interface module, alarm system module and data storage and management module. Through image acquisition, Preprocessing, feature extraction and defect detection algorithms identify defective products, and conduct real-time data analysis and alarm to provide high-precision and automated detection results.

Benefits of technology

It realizes automated inspection, improves the accuracy and consistency of inspection, can detect parts quality in real time, reduce labor costs, improve production efficiency and product quality, and ensure that each part meets quality standards.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119000692A_ABST
    Figure CN119000692A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of part detection, and discloses a part defective rate detection robot and a detection system.The part defective rate detection robot comprises an image acquisition module, an image processing module, a defect detection algorithm module, a data analysis module, a user interface module, an alarm system module and a data storage and management module; the image processing module is used for carrying out preprocessing and feature extraction on the image and then transmitting the processed data to the defect detection algorithm module, and the defect detection algorithm module is used for analyzing the image, identifying defective products and sending a detection result to the data analysis module and the alarm system module. By means of the automatic detection system, errors of human visual fatigue and subjective judgment are avoided, a high-resolution image processing industrial camera and an image processing algorithm can detect tiny defects which are difficult to find by naked eyes, and therefore the effect of improving detection precision and consistency can be achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of parts detection, and in particular to a parts defective rate detection robot and a detection system. Background Art

[0002] With the development of industrial automation and intelligent manufacturing, quality control has become increasingly important in the production process. Traditional quality inspection methods mainly rely on manual visual inspection.

[0003] Manual visual inspection has many shortcomings, such as low detection efficiency, poor accuracy, poor repeatability, and susceptibility to human factors. These defects make it difficult for traditional methods to meet the needs of efficient and high-precision quality inspection when facing modern large-scale and fast production lines. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a parts defective rate detection robot and detection system, which solves the problem that traditional quality detection methods are difficult to meet the needs of efficient and high-precision quality detection.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a robot detection system for detecting defective parts rate, comprising an image acquisition module, an image processing module, a defect detection algorithm module, a data analysis module, a user interface module, an alarm system module, and a data storage and management module. The image acquisition module acquires part images and transmits them to the image processing module. The image processing module preprocesses and extracts features of the images, and then transmits the processed data to the defect detection algorithm module. The defect detection algorithm module analyzes the images, identifies defective products, and sends the detection results to the data analysis module and the alarm system module. The data analysis module performs statistical and trend analysis on the detection results, generates reports, and displays the results on the user interface. The user interface displays the detection results and statistical analysis in real time, provides operation control functions, and displays alarm information. The data storage and management module stores all detection results, image data, statistical data and alarm logs, supports data query and backup, and when the alarm system module detects defective products, it triggers an audible and visual alarm or sends a notification, and records an alarm log for storage and subsequent query by the data storage and management module.

[0006] Preferably, the image acquisition module includes a shell, an inner wall of the shell is rotatably connected to an opening and closing door, a lower surface of the shell is fixedly connected to a leg, an upper right inner wall of the shell is fixedly connected to a control center, an upper left inner wall of the shell is fixedly connected to a lighting panel, a base is fixedly connected to the interior of the rear side of the shell, a mechanical arm is fixedly connected to the upper surface of the base, a side panel is fixedly connected to the interior of the front side of the shell, an outer wall of the side panel is rotatably connected to a rotating shaft, and an outer wall of the side panel is fixedly connected to a splint.

[0007] Preferably, the image processing module includes an image preprocessing unit and a feature extraction unit; The image preprocessing unit includes denoising: using filters, the filters including Gaussian filtering and median filtering to remove image noise and improve image quality, using adaptive filtering technology to dynamically adjust filtering parameters according to image content; Grayscale: Convert color images to grayscale images to simplify subsequent processing, using weighted average method or single channel extraction method for grayscale; Image enhancement: adjust contrast and brightness to highlight part features, use histogram equalization technology, and adopt adaptive contrast enhancement CLAHE to improve local contrast; The feature extraction unit includes edge detection: using Sobel and Canny operators to extract part edges, and combining edge refinement technology to maintain edge continuity and clarity; Morphological operations: The morphological operations include dilation, erosion, opening and closing operations for removing small noises and disconnected edges, and selecting appropriate structural elements and operation times according to part features; Contour detection: Use the findContours function in OpenCV to extract the contour and analyze the geometric features of the contour, including area, perimeter, and shape, for defect determination.

[0008] Preferably, the defect detection algorithm module includes a machine learning model unit and a traditional algorithm unit; The machine learning model unit includes data set preparation: collecting a large number of defective and non-defective part images, annotating them, and using data enhancement technology to generate diversified training data; Model training: Use convolutional neural network (CNN) for feature learning. The framework uses TensorFlow, Keras or PyTorch. Use rotation, flipping and scaling operations to increase the diversity of training data. Split the data set into training set, validation set and test set to enable the model to have generalization capabilities. Model deployment: Integrate the trained model into the inspection system to perform defect detection in real time, and use GPU acceleration or edge computing devices to optimize the model inference speed; The conventional algorithm unit includes template matching: detecting part defects by matching predefined templates, evaluating the matching degree using the normalized correlation coefficient NCC, and rotating and scaling the template to adapt to different part postures; Threshold segmentation: A segmentation method based on the gray value of the image to identify obvious defect areas and use adaptive threshold segmentation technology to improve adaptability to different lighting conditions; Morphological analysis: Defects are determined by analyzing the morphological features of parts, including size and shape, and using morphological filtering and morphological reconstruction techniques to improve detection accuracy.

[0009] Preferably, the data analysis module includes a statistical analysis unit and a trend analysis unit; The statistical analysis unit includes defective rate calculation: counting the number of defective products detected within a certain period of time and the total number of detections, calculating the defective rate, and providing defective rate analysis by shift, date, and product type; Data visualization: using charts to show the change in defective rate, the charts include line charts and bar charts, and the data visualization uses interactive data visualization tools, including Tableau and PowerBI; The trend analysis unit includes time series analysis: analyzing the change of defective rate over time and identifying periodic or trend changes, and the time series analysis uses the time series analysis method of ARIMA model and exponential smoothing method; Anomaly detection: Identify the time points when the defective rate suddenly increases, indicating possible production problems, and use machine learning algorithms for anomaly detection, including isolation forest and LOF.

[0010] Preferably, the user interface module includes a detection result display unit and an operation control panel unit; The detection result display unit includes real-time display: the image of the currently detected part and the detection result are displayed in real time on the display, and the image zooming, shrinking and panning functions are provided to facilitate detailed viewing; History Record: Provides query function to view historical inspection data and results, and supports retrieval by time, shift and part type; The operation control panel unit includes parameter settings: users can adjust detection parameters through the interface, the detection parameters include exposure time and threshold settings, and provide preset parameter schemes to switch different detection modes with one button; System maintenance: Provides system maintenance and troubleshooting tools, displays device status, error logs, and supports remote maintenance and diagnosis.

[0011] Preferably, the alarm system module includes a real-time alarm unit and an alarm recording unit; The real-time alarm unit includes an audible and visual alarm: when a defective product is detected, an alarm is sounded through a buzzer and a warning light, and different levels of alarm prompts are set according to the severity of the defective product; Information push: Notify relevant personnel via SMS and email, and integrate into the company's MES system for automated processing; The alarm recording unit records the time, type and cause of each alarm for subsequent analysis, and provides alarm statistical reports to analyze the alarm frequency and distribution.

[0012] Preferably, the data storage and management module includes a database unit and a cloud storage unit; The database unit includes a structured database: using relational databases MySQL and PostgreSQL to store test results and statistical data, and designing a standardized database structure; Unstructured database: Use NoSQL databases MongoDB and CouchDB to store image data and metadata, adopt a distributed storage architecture, and support large-scale data storage and fast access; The cloud storage unit includes cloud services: using AWS, Azure or Google Cloud cloud services, uploading data to the cloud for easy remote access and management, and providing an API interface to achieve data interaction and integration with other systems; Data backup: Back up data regularly to prevent data loss, and use offsite backup strategies to make data safe and reliable.

[0013] The present invention provides a parts defect rate detection robot and detection system. It has the following beneficial effects: 1. The present invention avoids human visual fatigue and subjective judgment errors through an automated detection system, improves the accuracy and consistency of detection, and high-resolution image processing, industrial cameras and image processing algorithms can detect tiny defects that are difficult to detect with the naked eye, thereby achieving the effect of improving detection accuracy and consistency.

[0014] 2. The present invention can detect the quality of parts in real time during the operation of the production line through the parts defective rate detection robot detection system. There is no need to stop the machine for quality inspection. The detection results are fed back in real time, and defective products can be quickly discovered and processed, reducing the flow of defective products into subsequent production links, thereby achieving the effect of improving production efficiency.

[0015] 3. The present invention reduces the reliance on manual quality inspectors and reduces labor costs through the automated detection system. It also reduces the impact of defective products on subsequent processing and assembly through early detection and elimination of defective products, thereby achieving the effect of reducing production costs.

[0016] 4. The present invention can detect each produced part through the parts defective rate detection robot detection system to ensure that each part leaving the factory meets the quality standards. Through the data analysis module, the system can identify common defects and production problems and help improve the process flow, thereby achieving the effect of improving the overall product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A three-dimensional diagram of a robot for detecting defective parts according to the present invention; Figure 2 A schematic diagram of a mechanical arm of the present invention; Figure 3 This is a schematic diagram of the architecture of a robot detection system for detecting defective parts of the present invention; Figure 4 It is a schematic diagram of the image processing module architecture of the present invention; Figure 5 It is a schematic diagram of the defect detection algorithm module architecture of the present invention; Figure 6 This is a schematic diagram of the data analysis module architecture of the present invention; Figure 7 It is a schematic diagram of the user interface module architecture of the present invention; Figure 8 It is a schematic diagram of the module architecture of the alarm system of the present invention; Fig. 9 It is a schematic diagram of the data storage and management module architecture of the present invention.

[0018] Among them, 1. Shell; 2. Opening and closing door; 3. Legs; 4. Control center; 5. Lighting panel; 6. Base; 7. Mechanical arm; 8. Side panel; 9. Rotating shaft; 10. Clamp. DETAILED DESCRIPTION

[0019] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Please see attached Figure 1 - Attachment Fig. 9The embodiment of the present invention provides a robot detection system for detecting defective parts, including an image acquisition module, an image processing module, a defect detection algorithm module, a data analysis module, a user interface module, an alarm system module, and a data storage and management module. The image acquisition module acquires part images and transmits them to the image processing module. The image processing module preprocesses and extracts features of the images, and then transmits the processed data to the defect detection algorithm module. The defect detection algorithm module analyzes the images, identifies defective products, and sends the detection results to the data analysis module and the alarm system module. The data analysis module performs statistical and trend analysis on the detection results, generates reports, and displays the results on the user interface. The user interface displays the detection results and statistical analysis in real time, provides operation control functions, and displays alarm information. The data storage and management module stores all detection results, image data, statistical data, and alarm logs, and supports data query and backup. When a defective product is detected, the alarm system module triggers an audible and visual alarm or sends a notification, and records an alarm log for storage and subsequent query by the data storage and management module. Specifically, the image acquisition module is responsible for acquiring images of parts for subsequent processing and analysis; the image processing module pre-processes the acquired images and extracts useful features for defect detection; the defect detection algorithm module analyzes part images through machine learning models or traditional algorithms, identifies and classifies defects; the data analysis module performs statistical and trend analysis on the detection results, and provides visual reports and anomaly detection; the user interface module provides an operation and control interface, displays real-time detection results and historical data, and allows users to adjust parameters and perform system maintenance; the alarm system module provides real-time alarms when defective products are detected, and notifies relevant personnel to handle the situation; the data storage and management module stores detection results, statistical data, alarm logs and image data, and provides data query and backup functions.

[0021] The image acquisition module includes a shell 1, an opening and closing door 2 is rotatably connected to the inner wall of the shell 1, a leg 3 is fixedly connected to the lower surface of the shell 1, a control center 4 is fixedly connected to the inner wall of the upper right side of the shell 1, a lighting panel 5 is fixedly connected to the inner wall of the upper left side of the shell 1, a base 6 is fixedly connected to the rear side of the shell 1, a mechanical arm 7 is fixedly connected to the upper surface of the base 6, a side panel 8 is fixedly connected to the front side of the shell 1, a rotating shaft 9 is rotatably connected to the outer wall of the side panel 8, and a clamping plate 10 is fixedly connected to the outer wall of the side panel 8; Specifically, the opening and closing door 2 is used to open the interior of the shell 1 to assist the robotic arm 7 in picking up the parts to be inspected, the legs 3 are used to support the stability of 1, the control center 4 is used to connect the image acquisition module and other modules and store some other module systems, the lighting panel 5 is used to illuminate the interior of the shell 1 for image acquisition, the base 6 is used to support the robotic arm 7 and store the buzzer and warning light, the robotic arm 7 is equipped with industrial cameras and various sensors for collecting information on the parts, the side panels 8 are used to support the rotating shaft 9, and the splint 10 is used to limit the parts to be inspected.

[0022] The image processing module includes an image preprocessing unit and a feature extraction unit; The image preprocessing unit includes denoising: using filters, including Gaussian filtering and median filtering to remove image noise and improve image quality, using adaptive filtering technology to dynamically adjust filtering parameters according to image content; Grayscale: Convert color images to grayscale images to simplify subsequent processing, using weighted average method or single channel extraction method for grayscale; Image enhancement: adjust contrast and brightness to highlight part features, use histogram equalization technology, and adopt adaptive contrast enhancement CLAHE to improve local contrast; The feature extraction unit includes edge detection: using Sobel and Canny operators to extract part edges, combined with edge refinement technology to maintain edge continuity and clarity; Morphological operations: Morphological operations include dilation, erosion, opening and closing operations to remove small noise and disconnected edges. Appropriate structural elements and operation times are selected according to part features. Contour detection: Use the findContours function in OpenCV to extract the contour and analyze the geometric features of the contour, including area, perimeter, and shape, to determine defects; The defect detection algorithm module includes a machine learning model unit and a traditional algorithm unit; The machine learning model unit includes data set preparation: collecting a large number of images of parts with and without defects, annotating them, and using data enhancement techniques to generate diverse training data; Model training: Use convolutional neural network (CNN) for feature learning. The framework uses TensorFlow, Keras or PyTorch. Use rotation, flipping and scaling operations to increase the diversity of training data. Split the data set into training set, validation set and test set to enable the model to have generalization capabilities. Model deployment: Integrate the trained model into the inspection system to perform defect detection in real time, and use GPU acceleration or edge computing devices to optimize the model inference speed; The traditional algorithm unit includes template matching: detecting part defects by matching predefined templates, using the normalized correlation coefficient NCC to evaluate the matching degree, and rotating and scaling the template to adapt to different part postures; Threshold segmentation: A segmentation method based on the gray value of the image to identify obvious defect areas and use adaptive threshold segmentation technology to improve adaptability to different lighting conditions; Morphological analysis: Defects are determined by analyzing the morphological features of parts, including size and shape. Morphological filtering and morphological reconstruction techniques are used to improve detection accuracy.

[0023] The data analysis module includes a statistical analysis unit and a trend analysis unit; The statistical analysis unit includes defective rate calculation: counting the number of defective products detected and the total number of tests within a certain period of time, calculating the defective rate, and providing defective rate analysis by shift, date, and product type; Data visualization: Use charts to show the change of defective rate. The charts include line charts and bar charts. Data visualization uses interactive data visualization tools, including Tableau and PowerBI. The trend analysis unit includes time series analysis: analyzing the change of defective rate over time, identifying periodic or trend changes, and time series analysis using ARIMA model and exponential smoothing method; Anomaly detection: Identify the time points when the defective rate suddenly increases, indicating possible production problems, and use machine learning algorithms for anomaly detection. Machine learning algorithms include isolation forest and LOF.

[0024] The user interface module includes a detection result display unit and an operation control panel unit; The unit for displaying the test results includes real-time display: the image of the part currently being tested and the test results are displayed in real time on the display, and the functions of image zooming in, zooming out and panning are provided for easy detailed viewing; History Record: Provides query function to view historical inspection data and results, and supports retrieval by time, shift and part type; The operation control panel unit includes parameter settings: users can adjust the detection parameters through the interface, including exposure time and threshold settings, and provide preset parameter solutions, and switch different detection modes with one button; System maintenance: Provides system maintenance and troubleshooting tools, displays device status, error logs, and supports remote maintenance and diagnosis.

[0025] The alarm system module includes a real-time alarm unit and an alarm recording unit; The real-time alarm unit includes sound and light alarms: when defective products are detected, the unit will sound an alarm through a buzzer and a warning light, and set different levels of alarm prompts according to the severity of the defective products; Information push: Notify relevant personnel via SMS and email, and integrate into the company's MES system for automated processing; The alarm recording unit records the time, type, and cause of each alarm for subsequent analysis, and provides alarm statistical reports to analyze the alarm frequency and distribution.

[0026] The data storage and management module includes a database unit and a cloud storage unit; The database unit includes structured database: using relational databases MySQL and PostgreSQL to store test results and statistical data, and designing standardized database structures; Unstructured database: Use NoSQL databases MongoDB and CouchDB to store image data and metadata, adopt a distributed storage architecture, and support large-scale data storage and fast access; The cloud storage unit includes cloud services: using AWS, Azure or Google Cloud cloud services, uploading data to the cloud facilitates remote access and management, and provides an API interface to achieve data interaction and integration with other systems; Data backup: Back up data regularly to prevent data loss, and use offsite backup strategies to make data safe and reliable.

[0027] Workflow Parts enter the inspection area: The parts are sent to the inspection area by a conveyor belt or a robotic arm, which triggers the sensor to start the inspection process.

[0028] Image acquisition: Industrial cameras capture images of parts, and light sources provide uniform illumination to ensure clear images.

[0029] Image processing and analysis: Preprocess and extract features from images, and use trained models or traditional algorithms for defect detection.

[0030] Result judgment: Determine whether the part is defective based on the test results, record the results and display them in the user interface.

[0031] Defective product handling: Defective products are removed from the production line by robotic arms or other equipment to ensure production quality.

[0032] Data Analysis and Reporting: Statistics and analysis of test data, generating reports, and providing them to quality control personnel and management for reference.

[0033] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A robot inspection system for inspecting defective parts, characterized in that: It includes an image acquisition module, an image processing module, a defect detection algorithm module, a data analysis module, a user interface module, an alarm system module, and a data storage and management module. The image acquisition module acquires part images and transmits them to the image processing module. The image processing module preprocesses and extracts features of the images, and then transmits the processed data to the defect detection algorithm module. The defect detection algorithm module analyzes the images, identifies defective products, and sends the detection results to the data analysis module and the alarm system module. The data analysis module performs statistical and trend analysis on the detection results, generates reports, and displays the results on the user interface. The user interface displays the detection results and statistical analysis in real time, provides operation control functions, and displays alarm information. The data storage and management module stores all detection results, image data, statistical data and alarm logs, and supports data query and backup. When the alarm system module detects defective products, it triggers an audible and visual alarm or sends a notification, and records an alarm log for the data storage and management module to store and query later.

2. A robot detection system for detecting defective parts according to claim 1, characterized in that: The image acquisition module comprises a shell (1), the inner wall of the shell (1) is rotatably connected to an opening and closing door (2), the lower surface of the shell (1) is fixedly connected to a support leg (3), the upper right inner wall of the shell (1) is fixedly connected to a control center (4), the upper left inner wall of the shell (1) is fixedly connected to a lighting panel (5), the rear interior of the shell (1) is fixedly connected to a base (6), the upper surface of the base (6) is fixedly connected to a mechanical arm (7), the front interior of the shell (1) is fixedly connected to a side panel (8), the outer wall of the side panel (8) is rotatably connected to a rotating shaft (9), and the outer wall of the side panel (8) is fixedly connected to a clamping plate (10).

3. A robot detection system for detecting defective parts according to claim 1, characterized in that: The image processing module includes an image preprocessing unit and a feature extraction unit; The image preprocessing unit includes denoising: using filters, the filters including Gaussian filtering and median filtering to remove image noise and improve image quality, using adaptive filtering technology to dynamically adjust filtering parameters according to image content; Grayscale: Convert color images to grayscale images to simplify subsequent processing, using weighted average method or single channel extraction method for grayscale; Image enhancement: adjust contrast and brightness to highlight part features, use histogram equalization technology, and adopt adaptive contrast enhancement CLAHE to improve local contrast; The feature extraction unit includes edge detection: using Sobel and Canny operators to extract part edges, and combining edge refinement technology to maintain edge continuity and clarity; Morphological operations: The morphological operations include dilation, erosion, opening and closing operations for removing small noises and disconnected edges, and selecting appropriate structural elements and operation times according to part features; Contour detection: Use the findContours function in OpenCV to extract the contour and analyze the geometric features of the contour, including area, perimeter, and shape, for defect determination.

4. A robot detection system for detecting defective parts according to claim 1, characterized in that: The defect detection algorithm module includes a machine learning model unit and a traditional algorithm unit; The machine learning model unit includes data set preparation: collecting a large number of defective and non-defective part images, annotating them, and using data enhancement technology to generate diversified training data; Model training: Use convolutional neural network (CNN) for feature learning. The framework uses TensorFlow, Keras or PyTorch. Use rotation, flipping and scaling operations to increase the diversity of training data. Split the data set into training set, validation set and test set to enable the model to have generalization capabilities. Model deployment: Integrate the trained model into the inspection system to perform defect detection in real time, and use GPU acceleration or edge computing devices to optimize the model inference speed; The conventional algorithm unit includes template matching: detecting part defects by matching predefined templates, evaluating the matching degree using the normalized correlation coefficient NCC, and rotating and scaling the template to adapt to different part postures; Threshold segmentation: A segmentation method based on the gray value of the image to identify obvious defect areas and use adaptive threshold segmentation technology to improve adaptability to different lighting conditions; Morphological analysis: Defects are determined by analyzing the morphological features of parts, including size and shape, and using morphological filtering and morphological reconstruction techniques to improve detection accuracy.

5. A robot detection system for detecting defective parts according to claim 1, characterized in that: The data analysis module includes a statistical analysis unit and a trend analysis unit; The statistical analysis unit includes defective rate calculation: counting the number of defective products detected within a certain period of time and the total number of detections, calculating the defective rate, and providing defective rate analysis by shift, date, and product type; Data visualization: using charts to show the change in defective rate, the charts include line charts and bar charts, and the data visualization uses interactive data visualization tools, including Tableau and PowerBI; The trend analysis unit includes time series analysis: analyzing the change of defective rate over time and identifying periodic or trend changes, and the time series analysis uses the time series analysis method of ARIMA model and exponential smoothing method; Anomaly detection: Identify the time points when the defective rate suddenly increases, indicating possible production problems, and use machine learning algorithms for anomaly detection, including isolation forest and LOF.

6. A robot inspection system for inspecting defective parts according to claim 1, characterized in that: The user interface module includes a detection result display unit and an operation control panel unit; The detection result display unit includes real-time display: the image of the currently detected part and the detection result are displayed in real time on the display, and the image zooming, shrinking and panning functions are provided to facilitate detailed viewing; History Record: Provides query function to view historical inspection data and results, and supports retrieval by time, shift and part type; The operation control panel unit includes parameter settings: users can adjust detection parameters through the interface, the detection parameters include exposure time and threshold settings, and provide preset parameter schemes to switch different detection modes with one button; System maintenance: Provides system maintenance and troubleshooting tools, displays device status, error logs, and supports remote maintenance and diagnosis.

7. A robot detection system for detecting defective parts according to claim 1, characterized in that: The alarm system module includes a real-time alarm unit and an alarm recording unit; The real-time alarm unit includes an audible and visual alarm: when a defective product is detected, an alarm is sounded through a buzzer and a warning light, and different levels of alarm prompts are set according to the severity of the defective product; Information push: Notify relevant personnel via SMS and email, and integrate into the company's MES system for automated processing; The alarm recording unit records the time, type and cause of each alarm for subsequent analysis, and provides alarm statistical reports to analyze the alarm frequency and distribution.

8. A robot inspection system for inspecting defective parts according to claim 1, characterized in that: The data storage and management module includes a database unit and a cloud storage unit; The database unit includes a structured database: using relational databases MySQL and PostgreSQL to store test results and statistical data, and designing a standardized database structure; Unstructured database: Use NoSQL databases MongoDB and CouchDB to store image data and metadata, adopt a distributed storage architecture, and support large-scale data storage and fast access; The cloud storage unit includes cloud services: using AWS, Azure or Google Cloud cloud services, uploading data to the cloud for easy remote access and management, and providing an API interface to achieve data interaction and integration with other systems; Data backup: Back up data regularly to prevent data loss, and use offsite backup strategies to make data safe and reliable.

Citation Information

Patent Citations

  • Workpiece processing detection system based on three-dimensional image technology and using method of workpiece processing detection system

    CN109884068A

  • Intelligent robot processing production detection system

    CN112598651A

  • Quick identification system of refabrication bearing defect

    CN207832745U

  • Online size crack detection device

    CN214407325U

  • Automatic machine of vision on detection device for workpiece with complex curved surface

    EP4016058A1

Cited By

  • Galvanizing quality control system for automobile part machining

    CN120293991A

  • Latex skinning defect online detection method and system based on machine vision

    CN120726008A

  • Device for detecting position defects of aircraft assembly parts

    CN120846205A

  • Self-adaptive appearance defect detection device for machined part

    CN120870135A