Galvanized strip steel surface quality detection system based on vision and vortex

Abstract

The invention provides a galvanized strip steel surface quality detection system based on vision and eddy current, and relates to the technical field of surface quality detection. The galvanized strip steel surface quality detection system based on vision and eddy current comprises a Tongxiang acquisition module, an image processing module, an eddy current detection module and a defect judgment module, and the image acquisition module is used for acquiring high-quality real-time surface image information; the image processing module is used for preprocessing the obtained image through an internal algorithm to obtain a defective ROI (Region of Interest); and the eddy current detection module is used for carrying out further defect detection and confirmation on the ROI. By combining a machine vision technology and an eddy current detection technology, the defect area is accurately locked by utilizing the advantages of high speed, high precision and high efficiency of the machine vision technology; and by utilizing the advantage that the eddy current detection technology has high detection sensitivity on defects on the metal surface and near the surface, eddy current signals of tiny defects and low-contrast defects are accurately described.

Description

Technical Field

[0001] The invention relates to the technical field of surface quality detection, and in particular to a galvanized steel strip surface quality detection system based on vision and eddy current. Background Art

[0002] Galvanized steel strip is an important industrial material. Its surface quality directly affects the performance and service life of the product. Therefore, the detection of its surface quality is particularly important. However, the traditional surface quality detection method mainly relies on manual visual inspection, which is inefficient and error-prone. In recent years, with the development of technology, machine vision technology and eddy current detection technology have been gradually applied to the detection of galvanized steel strip. However, when machine vision technology is applied in this field, it still has the problem of insufficient detection ability for tiny defects and low-contrast defects; and although eddy current detection technology has a high detection sensitivity for defects on the metal surface and near the surface, the types of defects that can be detected and the description parameters are limited, and it is impossible to comprehensively evaluate the surface quality of the product. Both technologies have their own defects.

[0003] Therefore, it is of great significance to combine the advantages of the two technologies and develop a surface quality inspection system that combines machine vision and eddy current inspection to avoid defects to a large extent and improve inspection efficiency and accuracy. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a galvanized strip surface quality inspection system based on vision and eddy current, which solves the problems that machine vision technology is insufficient in detecting tiny defects and low-contrast defects, and eddy current inspection technology has limited detectable defect types and description parameters, and cannot comprehensively evaluate the surface quality of the product.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a galvanized strip surface quality detection system based on vision and eddy current, including a Tongxiang acquisition module, an image processing module, an eddy current detection module and a defect judgment module, the image acquisition module is used to obtain high-quality real-time surface image information; the image processing module is used to pre-process the acquired image through an internal algorithm to obtain the ROI area of ​​the defect; the eddy current detection module is used to further detect and confirm the defects in the ROI area; the defect judgment module is used to classify the defects according to the specific parameters of the defects detected by the eddy current detection module.

[0006] Preferably, the image acquisition module includes a high-speed linear array CCD camera and a dedicated LED light source, which are used to continuously photograph the upper and lower surfaces of the galvanized steel strip to obtain high-quality real-time surface image information.

[0007] Preferably, the preprocessing includes exposure, grayscale processing, and cropping of locations where defects appear in the image.

[0008] Preferably, the eddy current detection module accurately detects defect parameters on the surface and near the surface of the galvanized steel strip by detecting changes in voltage and impedance.

[0009] Preferably, the defect judgment module compares the obtained data with the feature library of each defect category in the expert sample library built by the user, classifies similar defect data through an intelligent classifier and outputs the classification result.

[0010] Preferably, the high-speed linear array CCD camera can perform imaging at extremely high speed and resolution, and is used for quickly, continuously and accurately capturing images of galvanized steel strip.

[0011] Preferably, the dedicated LED light source is used to ensure image clarity and contrast.

[0012] Preferably, the working mode of the module can be summarized into the following steps:

[0013] S1. Receiving eddy current detection data: The defect determination module first receives data from the eddy current detection module;

[0014] S2. Data preprocessing: After receiving the raw data, the defect determination module will preprocess the data to remove noise and other interference factors and improve the signal-to-noise ratio of the data;

[0015] S3. Feature extraction: After preprocessing, the module extracts defect-related features from the data;

[0016] S4. Comparison with expert sample library: The defect determination module will compare the extracted features with the features in the pre-established expert sample library;

[0017] S5. Intelligent classification: After comparison with the expert database, the defect determination module can automatically classify defects using an intelligent classifier;

[0018] S6. Result output and optimization: Finally, the defect determination module will output the classification results, including the type, size, location and other information of the defect.

[0019] The present invention provides a galvanized steel strip surface quality detection system based on vision and eddy current.

[0020] Beneficial effects:

[0021] 1. The present invention combines machine vision technology and eddy current detection technology, and utilizes the advantages of high speed, high precision and high efficiency of machine vision technology to accurately lock the defect area; and then utilizes the advantage of high detection sensitivity of eddy current detection technology for defects on the metal surface and near the surface, and accurately describes the eddy current signals of tiny defects and low-contrast defects.

[0022] 2. The present invention can continuously photograph the upper and lower surfaces of galvanized steel strips by using a high-speed linear array CCD camera and a dedicated LED light source to obtain high-quality real-time surface image information. The image processing module pre-processes the acquired images through an internal algorithm, and can accurately determine the ROI area of ​​the suspected defect, so that the image of the defect area occupies a higher proportion in the algorithm. This method can effectively reduce the imbalance of positive and negative samples and improve the performance of detecting defects; the eddy current detection module uses the principle of eddy current detection to obtain more accurate surface and near-surface defect parameters of galvanized steel strips by detecting changes in voltage and impedance; the eddy current detection module transmits the obtained data, including eddy current signals, to the defect judgment module, so that the defect judgment module can judge and classify defects; the defect judgment module compares the obtained data with the defect category feature library in the expert sample library built by the user, and classifies similar defect data through an intelligent classifier and outputs the classification results; the user can confirm and optimize the classification results based on field experience.

[0023] 3. In the present invention, before the module works, the user will first build an expert sample library for use by the module. The user first uses the eddy current detection technology to detect known defects in the galvanized strip steel, and obtains different manifestations of eddy current signals at different defects, and then classifies these eddy current signals with different manifestations into known defect types and uploads them to the expert sample library; the user can also classify the eddy current signals obtained by using the eddy current detection technology in actual production, classify different eddy current signals into known defect types, expand the number of samples in the expert sample library, and improve the reliability of the defect judgment module. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a module block diagram of the present invention;

[0025] Figure 2 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments 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.

[0027] Embodiment 1:

[0028] like Figure 1 - Figure 2As shown, an embodiment of the present invention provides a galvanized strip surface quality detection system based on vision and eddy current, including a Tongxiang acquisition module, an image processing module, an eddy current detection module and a defect judgment module. The image acquisition module is used to obtain high-quality real-time surface image information; the image processing module is used to pre-process the acquired image through an internal algorithm to obtain the ROI area of ​​the defect; the eddy current detection module is used to perform further defect detection and confirmation on the ROI area; the defect judgment module is used to classify defects according to specific parameters of the defects detected by the eddy current detection module.

[0029] The specific image acquisition module uses a high-speed linear array CCD camera and a dedicated LED light source to continuously photograph the upper and lower surfaces of the galvanized steel strip and obtain high-quality real-time surface image information.

[0030] The image processing module uses internal algorithms to pre-process the acquired images, including exposure and grayscale processing, and crops the locations where defects appear in the images. In order to avoid unsatisfactory defect detection results caused by sample imbalance, which results in a large number of defects being missed, the module's internal algorithm uses a sample-balanced sampling strategy, which can accurately determine the ROI area of ​​suspected defects and increase the proportion of images in the defective area in the algorithm. This method can effectively reduce the imbalance of positive and negative samples and improve the performance of defect detection.

[0031] The eddy current detection module performs further defect detection and confirmation on the ROI area; the module uses the principle of eddy current detection. Since defects on the surface or near the surface of the galvanized steel strip can cause changes in the coil voltage and impedance in the eddy current detection, the defect parameters on the surface and near the surface of the galvanized steel strip can be detected more accurately by detecting changes in voltage and impedance; the module transmits the obtained data, including eddy current signals, to the defect judgment module to facilitate the defect judgment module to judge and classify the defects.

[0032] The defect determination module classifies defects according to the specific parameters of the defects detected by the eddy current detection module; the defect determination module compares the obtained data with the feature library of each defect category in the expert sample library built by the user, classifies similar defect data through the intelligent classifier and outputs the classification results; the user can confirm and optimize the classification results based on field experience.

[0033] Embodiment 2:

[0034] The embodiment of the present invention provides a method for detecting the surface quality of a galvanized steel strip based on vision and eddy current, using the system for detecting the surface quality of a galvanized steel strip based on vision and eddy current as described in any one of claims 1 to 7, comprising the following steps:

[0035] Specifically, before the module works, the user will first build an expert sample library for the module to use. The user first uses the eddy current detection technology to detect known defects in the galvanized strip steel, and obtains different manifestations of eddy current signals at different defects, and then classifies these eddy current signals with different manifestations into known defect types and uploads them to the expert sample library; at the same time, the user can also classify the eddy current signals obtained by using the eddy current detection technology in actual production, classify different eddy current signals into known defect types, expand the number of samples in the expert sample library, and improve the reliability of the defect judgment module; Figure 2 , which can be summarized into the following steps:

[0036] S1. Receiving eddy current detection data: The defect determination module first receives data from the eddy current detection module; these data include information on changes in coil voltage and impedance in eddy current detection caused by defects on the surface or near the surface of the galvanized steel strip.

[0037] S2. Data preprocessing: After receiving the raw data, the defect determination module will preprocess the data to remove noise and other interference factors and improve the signal-to-noise ratio of the data; this step may include filtering, smoothing, and data standardization to ensure the accuracy of subsequent analysis.

[0038] S3. Feature extraction: After preprocessing, the module will extract defect-related features from the data; these features may include parameters such as the amplitude, frequency, and phase of the eddy current signal. These parameters have different manifestations for different types of defects, and the defects can be classified into different categories based on the different manifestations of the parameters.

[0039] S4. Comparison with expert sample library: The defect judgment module will compare the extracted features with the features in the pre-established expert sample library; the expert sample library contains feature data of various known defect types. Since these data are obtained from a large number of experiments and actual production experience, the comparison and subsequent classification results are closer to production needs.

[0040] S5. Intelligent classification: After comparison with the expert database, the defect determination module can use the intelligent classifier to automatically classify the defects; the classifier will classify the current defect into known defect categories based on the similarity of the features.

[0041] S6. Result output and optimization: Finally, the defect determination module will output the classification results, including the type, size, location and other information of the defect; at the same time, users can confirm and optimize the classification results based on on-site experience and actual conditions to improve the accuracy and reliability of detection.

[0042] 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. The galvanized strip surface quality inspection system based on vision and eddy current includes a Tongxiang acquisition module, an image processing module, an eddy current detection module and a defect judgment module, and is characterized by: The image acquisition module is used to obtain high-quality real-time surface image information; the image processing module is used to pre-process the acquired image through an internal algorithm to obtain the ROI area of ​​the defect; the eddy current detection module is used to further detect and confirm the defects in the ROI area; the defect judgment module is used to classify the defects according to the specific parameters of the defects detected by the eddy current detection module.

2. The galvanized steel strip surface quality detection system based on vision and eddy current according to claim 1 is characterized in that: The image acquisition module includes a high-speed linear array CCD camera and a dedicated LED light source, which are used to continuously photograph the upper and lower surfaces of the galvanized steel strip to obtain high-quality real-time surface image information.

3. The galvanized steel strip surface quality detection system based on vision and eddy current according to claim 1 is characterized in that: The preprocessing includes exposure, grayscale processing, and cropping the position where defects appear in the image.

4. The galvanized steel strip surface quality detection system based on vision and eddy current according to claim 1 is characterized in that: The eddy current detection module accurately detects the defect parameters on the surface and near the surface of the galvanized steel strip by detecting changes in voltage and impedance.

5. The galvanized steel strip surface quality detection system based on vision and eddy current according to claim 1 is characterized in that: The defect judgment module compares the obtained data with the feature library of each defect category in the expert sample library built by the user, classifies similar defect data through an intelligent classifier and outputs the classification result.

6. The galvanized steel strip surface quality detection system based on vision and eddy current according to claim 2 is characterized in that: The high-speed linear array CCD camera can perform imaging at extremely high speed and resolution, and is used for quickly, continuously and accurately capturing images of galvanized steel strips.

7. The galvanized steel strip surface quality detection system based on vision and eddy current according to claim 2 is characterized in that: The dedicated LED light source is used to ensure image clarity and contrast.

8. A method for detecting the surface quality of a galvanized steel strip based on vision and eddy current, using a system for detecting the surface quality of a galvanized steel strip based on vision and eddy current as claimed in any one of claims 1 to 7, characterized in that: It includes the following steps: S1. Receiving eddy current detection data: The defect determination module first receives data from the eddy current detection module; S2. Data preprocessing: After receiving the raw data, the defect determination module will preprocess the data to remove noise and other interference factors and improve the signal-to-noise ratio of the data; S3. Feature extraction: After preprocessing, the module extracts defect-related features from the data; S4. Comparison with expert sample library: The defect determination module will compare the extracted features with the features in the pre-established expert sample library; S5. Intelligent classification: After comparison with the expert database, the defect determination module can automatically classify defects using an intelligent classifier; S6. Result output and optimization: Finally, the defect determination module will output the classification results, including the type, size, location and other information of the defect.

Images