Appearance detection system for automobile stamping and welding part

Abstract

The invention relates to the technical field of stamping and welding part appearance detection, in particular to an automobile stamping and welding part appearance detection system. According to the technical scheme, the automobile stamping and welding part appearance detection system comprises an image acquisition module, an image transmission module, an image preprocessing module, a feature extraction module, a defect identification module, a result evaluation and feedback module and a database management module, compared with a traditional automobile stamping and welding part appearance detection system, the automobile stamping and welding part appearance detection system generally depends on a traditional image processing technology to carry out defect recognition through a simple and regular algorithm, the accuracy and efficiency of the system are relatively low, and complex and changeable defect types are difficult to deal with; according to the appearance detection system for the automobile stamping and welding part, an advanced image processing technology and a deep learning algorithm are used for defect identification, image features can be automatically analyzed, defects on the stamping and welding part are accurately judged, and the defects are classified and positioned.

Description

Technical Field

[0001] The invention relates to the technical field of stamping and welding parts appearance detection, and in particular to an automobile stamping and welding parts appearance detection system. Background Art

[0002] The automotive stamping and welding parts appearance inspection system is an automated or semi-automated equipment or system specifically used to inspect and evaluate the appearance quality of automotive stamping and welding parts. It integrates multiple advanced technologies, such as image processing, machine vision, sensor technology, and data analysis, to conduct comprehensive, rapid, and accurate inspections of the appearance of automotive stamping and welding parts.

[0003] Existing automotive stamping and welding parts appearance inspection systems generally rely on traditional image processing technology or simple rule-based algorithms for defect identification. The accuracy and efficiency of the system are relatively low, and it is difficult to cope with complex and changeable defect types.

[0004] In view of the relatively low accuracy and efficiency of the existing automotive stamped and welded parts appearance inspection system, and its difficulty in dealing with complex and changeable defect types, this automotive stamped and welded parts appearance inspection system uses advanced image processing technology and deep learning algorithms for defect identification. It can automatically analyze image features, accurately judge defects on stamped and welded parts, and classify and locate them. Summary of the invention

[0005] In order to overcome the existing appearance inspection system of automobile stamping and welding parts, it generally relies on traditional image processing technology or a simple rule-based algorithm for defect identification. The accuracy and efficiency of the system are relatively low, and it is difficult to cope with the problem of complex and changeable defect types.

[0006] The technical solution of the present invention is: an appearance inspection system for automobile stamping and welding parts, comprising: Image acquisition module: used to collect the appearance image of the punch weldment; Image transmission module: used to transmit the appearance information collected by the image acquisition module; Image preprocessing module: used to preprocess the image information transmitted by the image transmission module; Feature extraction module: extracts feature points of the preprocessed image; Defect recognition module: used to identify defects based on feature points extracted by the feature extraction module; Result evaluation and feedback module: used to further evaluate the defect identification results; Database management module: used to store various data of weldment inspection.

[0007] Preferably, the image acquisition module includes multiple groups of high-resolution industrial cameras, which are responsible for capturing the image information of the stamped and welded parts in real time and clearly, through lenses of different angles and focal lengths to capture the overall appearance and details of the stamped and welded parts. At the same time, the image acquisition module also has a stable light source control system to ensure that high-quality images can be obtained under different lighting conditions.

[0008] Preferably, the image transmission module is used to realize wireless data transmission between nodes through wireless communication technologies such as Wi-Fi, Bluetooth, etc. The data transmission module is responsible for transmitting the image information of the stamped and welded parts collected by the image acquisition module between each module, and transmitting the information collected by the image acquisition module to the image preprocessing module.

[0009] Preferably, the image preprocessing module is used to preprocess the collected original image, using technical operations such as denoising, contrast enhancement, grayscale and edge detection to improve the accuracy and efficiency of subsequent image analysis. The image preprocessing module helps to highlight defect features and reduce background interference.

[0010] Preferably, the feature extraction module is used to extract information that can characterize the appearance characteristics of the stamped and welded parts from the preprocessed images, including shape, texture and color changes. The feature extraction module uses this information as the basis for subsequent defect identification to ensure that potential defect information can be accurately captured.

[0011] Preferably, the defect recognition module is used to identify defects in stamping weldments based on the extracted features using advanced technologies such as machine learning and deep learning. The defect recognition module can automatically analyze images, determine whether there are defects such as scratches, dents, cracks, etc., and classify and locate them.

[0012] Preferably, the defect identification module comprises the following steps when operating: S11: Using a feature extraction method based on a convolutional neural network or other advanced deep learning algorithms, extract information that can characterize the appearance characteristics of the stamped and welded parts from the preprocessed image; S12: Carefully design and select feature extraction algorithms based on the characteristics of stamping and welding parts and common defect types to ensure that potential defect information can be accurately captured; S13: Use a large amount of labeled stamping and welding parts image data to train the deep learning model, input the stamping and welding parts image to be detected into the trained model, automatically analyze the image and output the defect recognition result; S14: Using the positioning algorithm in the image processing technology, the identified defects are accurately located on the image, and the specific location and range of the defects are marked; S15: Generate a detailed test report based on the defect identification results to facilitate subsequent processing and analysis by the user; S16: Collect test data regularly to evaluate and optimize the performance of the model.

[0013] Preferably, by using advanced technologies such as machine learning and deep learning for defect identification, images can be automatically analyzed to accurately determine whether there are defects on the stamped and welded parts, and they can be classified and located, thus ensuring product quality while also ensuring production efficiency.

[0014] Preferably, the result evaluation and feedback module is used to further evaluate the results of defect identification and determine the severity and acceptability of the defects. At the same time, the result evaluation and feedback module is responsible for feeding back the test results to the production system, including real-time alarm, recording defect information and generating test reports, which is of great significance for timely adjusting the production process and improving product quality.

[0015] Preferably, the database management module is responsible for storing and managing the image data, defect information and inspection results of the stamped and welded parts. The database management module provides functions such as data query and statistical analysis to help users understand product quality distribution, identify common defect types and causes, and provide data support for product improvement and process optimization.

[0016] Preferably, the database management module comprises the following steps when operating: S21: The database management module collects data to be stored from each detection link; S22: Organize the collected data, including formatting, deduplication and classification, to ensure the consistency and accuracy of the data; S23: The sorted data is stored in the database, when the data in the punching and welding parts appearance inspection system changes, the data in the database is updated synchronously; S24: Back up the database regularly to prevent data loss or damage.

[0017] Beneficial effects of the present invention: 1. Compared with the traditional automobile stamping and welding parts appearance inspection system, which generally relies on traditional image processing technology or simple rule-based algorithms for defect identification, the accuracy and efficiency of the system are relatively low, and it is difficult to cope with complex and changeable defect types. This automobile stamping and welding parts appearance inspection system uses advanced image processing technology and deep learning algorithms for defect identification, which can automatically analyze image features, accurately determine defects on stamping and welding parts, and classify and locate them; 2. By using advanced technologies such as machine learning and deep learning for defect identification, the image can be automatically analyzed to accurately determine whether there are defects on the stamped and welded parts, and they can be classified and located, ensuring product quality while also ensuring production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 What is shown is a schematic diagram of the working process of a system for inspecting the appearance of automobile stamping and welding parts of the present invention; Figure 2 What is shown is a schematic diagram of the working process of a defect recognition module of an automobile stamping and welding parts appearance inspection system of the present invention; Figure 3 What is shown is a schematic diagram of the workflow of a database management module of an automobile stamping and welding parts appearance inspection system of the present invention. DETAILED DESCRIPTION

[0019] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0020] See also Figure 1-3 The present invention provides an embodiment: an automobile stamping and welding parts appearance inspection system, comprising: Image acquisition module: used to collect the appearance image of the punch weldment; Image transmission module: used to transmit the appearance information collected by the image acquisition module; Image preprocessing module: used to preprocess the image information transmitted by the image transmission module; Feature extraction module: extracts feature points of the preprocessed image; Defect recognition module: used to identify defects based on feature points extracted by the feature extraction module; Result evaluation and feedback module: used to further evaluate the defect identification results; Database management module: used to store various data of weldment inspection.

[0021] Preferably, the image acquisition module includes multiple groups of high-resolution industrial cameras, which are responsible for capturing the image information of the stamped and welded parts in real time and clearly, through lenses of different angles and focal lengths to capture the overall appearance and details of the stamped and welded parts. At the same time, the image acquisition module also has a stable light source control system to ensure that high-quality images can be obtained under different lighting conditions.

[0022] Preferably, the image transmission module is used to realize wireless data transmission between nodes through wireless communication technologies such as Wi-Fi, Bluetooth, etc. The data transmission module is responsible for transmitting the image information of the stamped and welded parts collected by the image acquisition module between each module, and transmitting the information collected by the image acquisition module to the image preprocessing module.

[0023] Preferably, the image preprocessing module is used to preprocess the collected original image, using technical operations such as denoising, contrast enhancement, grayscale and edge detection to improve the accuracy and efficiency of subsequent image analysis. The image preprocessing module helps to highlight defect features and reduce background interference.

[0024] Preferably, the feature extraction module is used to extract information that can characterize the appearance characteristics of the stamped and welded parts from the preprocessed images, including shape, texture and color changes. The feature extraction module uses this information as the basis for subsequent defect identification to ensure that potential defect information can be accurately captured.

[0025] Preferably, the defect recognition module is used to identify defects in stamping weldments based on the extracted features using advanced technologies such as machine learning and deep learning. The defect recognition module can automatically analyze images, determine whether there are defects such as scratches, dents, cracks, etc., and classify and locate them.

[0026] Preferably, the defect identification module comprises the following steps when operating: S11: Using a feature extraction method based on a convolutional neural network or other advanced deep learning algorithms, extract information that can characterize the appearance characteristics of the stamped and welded parts from the preprocessed image; S12: Carefully design and select feature extraction algorithms based on the characteristics of stamping and welding parts and common defect types to ensure that potential defect information can be accurately captured; S13: Use a large amount of labeled stamping and welding parts image data to train the deep learning model, input the stamping and welding parts image to be detected into the trained model, automatically analyze the image and output the defect recognition result; S14: Using the positioning algorithm in the image processing technology, the identified defects are accurately located on the image, and the specific location and range of the defects are marked; S15: Generate a detailed test report based on the defect identification results to facilitate subsequent processing and analysis by the user; S16: Collect test data regularly to evaluate and optimize the performance of the model.

[0027] Preferably, by using advanced technologies such as machine learning and deep learning for defect identification, images can be automatically analyzed to accurately determine whether there are defects on the stamped and welded parts, and they can be classified and located, thus ensuring product quality while also ensuring production efficiency.

[0028] Preferably, the result evaluation and feedback module is used to further evaluate the results of defect identification and determine the severity and acceptability of the defects. At the same time, the result evaluation and feedback module is responsible for feeding back the test results to the production system, including real-time alarm, recording defect information and generating test reports, which is of great significance for timely adjusting the production process and improving product quality.

[0029] Preferably, the database management module is responsible for storing and managing the image data, defect information and inspection results of the stamped and welded parts. The database management module provides functions such as data query and statistical analysis to help users understand product quality distribution, identify common defect types and causes, and provide data support for product improvement and process optimization.

[0030] Preferably, the database management module comprises the following steps when operating: S21: The database management module collects data to be stored from each detection link; S22: Organize the collected data, including formatting, deduplication and classification, to ensure the consistency and accuracy of the data; S23: The sorted data is stored in the database, when the data in the punching and welding parts appearance inspection system changes, the data in the database is updated synchronously; S24: Back up the database regularly to prevent data loss or damage.

[0031] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of those skilled in the art without departing from the spirit of the present invention.

Claims

1. An appearance inspection system for automobile stamping and welding parts; characterized by: Included are: Image acquisition module: used to collect the appearance image of the punch weldment; Image transmission module: used to transmit the appearance information collected by the image acquisition module; Image preprocessing module: used to preprocess the image information transmitted by the image transmission module; Feature extraction module: extracts feature points of the preprocessed image; Defect recognition module: used to identify defects based on feature points extracted by the feature extraction module; Result evaluation and feedback module: used to further evaluate the defect identification results; Database management module: used to store various data of weldment inspection.

2. The automotive stamping and welding parts appearance inspection system according to claim 1, characterized in that: The image acquisition module includes multiple groups of high-resolution industrial cameras. The high-resolution industrial cameras are responsible for capturing the image information of the stamped and welded parts in real time and clearly. Through lenses of different angles and focal lengths, the overall appearance and details of the stamped and welded parts can be captured. At the same time, the image acquisition module also has a stable light source control system to ensure that high-quality images can be obtained under different lighting conditions.

3. The automobile stamping and welding parts appearance inspection system according to claim 1 is characterized in that: The image transmission module is used to realize wireless data transmission between nodes through wireless communication technologies such as Wi-Fi, Bluetooth, etc. The data transmission module is responsible for transmitting the image information of the stamping and welding parts collected by the image acquisition module between modules, and transmitting the information collected by the image acquisition module to the image preprocessing module.

4. The automobile stamping and welding parts appearance inspection system according to claim 1 is characterized in that: The image preprocessing module is used to preprocess the collected original images, using technical operations such as denoising, contrast enhancement, grayscale and edge detection to improve the accuracy and efficiency of subsequent image analysis. The image preprocessing module helps to highlight defect features and reduce background interference.

5. The automobile stamping and welding parts appearance inspection system according to claim 1, characterized in that: The feature extraction module is used to extract information that can characterize the appearance characteristics of stamped and welded parts from the preprocessed images, including shape, texture and color changes. The feature extraction module uses this information as the basis for subsequent defect identification to ensure that potential defect information can be accurately captured.

6. The automobile stamping and welding parts appearance inspection system according to claim 1, characterized in that: The defect recognition module is used to identify defects in punch welds based on the extracted features using advanced technologies such as machine learning and deep learning. The defect recognition module can automatically analyze images to determine whether there are defects such as scratches, dents, cracks, etc., and classify and locate them.

7. The automobile stamping and welding parts appearance inspection system according to claim 1 is characterized in that: The defect identification module includes the following steps when working: S11: Using a feature extraction method based on a convolutional neural network or other advanced deep learning algorithms, extract information that can characterize the appearance characteristics of the stamped and welded parts from the preprocessed image; S12: Carefully design and select feature extraction algorithms based on the characteristics of stamping and welding parts and common defect types to ensure that potential defect information can be accurately captured; S13: Use a large amount of labeled stamping and welding parts image data to train the deep learning model, input the stamping and welding parts image to be detected into the trained model, automatically analyze the image and output the defect recognition result; S14: Using the positioning algorithm in the image processing technology, the identified defects are accurately located on the image, and the specific location and range of the defects are marked; S15: Generate a detailed test report based on the defect identification results to facilitate subsequent processing and analysis by the user; S16: Collect test data regularly to evaluate and optimize the performance of the model.

8. The automobile stamping and welding parts appearance inspection system according to claim 1 is characterized in that: The result evaluation and feedback module is used to further evaluate the results of defect identification and determine the severity and acceptability of the defects. At the same time, the result evaluation and feedback module is responsible for feeding back the test results to the production system, including real-time alarm, recording defect information and generating test reports, which is of great significance for timely adjusting the production process and improving product quality.

9. The automobile stamping and welding parts appearance inspection system according to claim 1, characterized in that: The database management module is responsible for storing and managing the image data, defect information and inspection results of stamping and welding parts. The database management module provides functions such as data query and statistical analysis to help users understand product quality distribution, identify common defect types and causes, and provide data support for product improvement and process optimization.

10. The automobile stamping and welding parts appearance inspection system according to claim 1, characterized in that: The database management module includes the following steps when working: S21: The database management module collects data to be stored from each detection link; S22: Organize the collected data, including formatting, deduplication and classification, to ensure the consistency and accuracy of the data; S23: The sorted data is stored in the database, when the data in the punching and welding parts appearance inspection system changes, the data in the database is updated synchronously; S24: Back up the database regularly to prevent data loss or damage.

Images