Automatic crack density analysis method for metal micro-arc oxidation electron microscope scanning image

Through automated image analysis methods, using technical means such as CLAHE, Otsu algorithm and Canny edge detection, the problems of low efficiency and insufficient accuracy of traditional methods are solved, and efficient and accurate crack density analysis is achieved, suitable for metal microarc oxidation coatings. Crack defect evaluation on the surface of the layer.

CN119068005APending Publication Date: 2024-12-03LIAONING TECHNICAL UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411249833.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-06
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Traditional electron microscopy image analysis methods rely on manual operations, with low efficiency and limited accuracy and repetition of results. Especially in statistical analysis, there is a lack of high-efficiency and accurate methods, making it difficult to effectively evaluate the crack defects of the surface of metal microarc oxidation coatings.

Method used

Provide an automated method, including image preprocessing, edge detection, morphological operation, crack recognition and length calculation, and crack density estimation, through CLAHE, Otsu algorithm, Canny edge detection, Hough transform detection and 8-connectivity algorithm, etc. Technical means to realize automated crack density analysis.

Benefits of technology

It significantly improves the efficiency and accuracy of metal microarc oxidation electron microscopy scanning image analysis, reduces human errors, provides detailed quantitative data, supports material performance evaluation and industrial quality control, and has broad application potential.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119068005A_ABST
    Figure CN119068005A_ABST
Patent Text Reader

Abstract

The invention discloses an automatic crack density calculation method which is used for analyzing an electron microscope scanning image of a metal micro-arc oxidation coating. The method comprises the steps of image preprocessing, edge detection, morphological operation, crack identification and length calculation and crack density estimation. In the image preprocessing stage, an RGB image is converted into a grayscale image, and the contrast is enhanced by applying CLAHE. In the edge detection stage, a Canny algorithm is combined with a multi-scale method, and the crack edge of the connection fracture is processed through closed operation and expansion in morphological operation. In the crack identification and length calculation stage, a straight line segment is detected through Hough line transformation, and the length is calculated. In the crack density estimation, the total length is converted into a micrometer scale, and the density is calculated based on the actual area of the image. According to the method, the analysis efficiency and accuracy are remarkably improved, human errors are reduced, the method is suitable for various metal materials and micro-arc oxidation treatment conditions, and the time and cost of research and industrial application are effectively saved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of material science and electron microscope technology, and in particular relates to a crack density calculation method for automatically processing and analyzing metal micro-arc oxidation electron microscope scanning images. This method can automatically identify and measure the crack density in the image, and is suitable for evaluating the surface crack defects of the coating of the material after micro-arc oxidation treatment. Background Art

[0002] As an advanced surface modification method, metal micro-arc oxidation technology is widely used to improve the wear resistance, corrosion resistance and hardness of metals. This technology not only enhances the surface properties of the material but also significantly improves the service life by forming a dense ceramic layer on the metal surface. During the micro-arc oxidation process, there is a mismatch stress between the coating and the metal substrate, which induces cracks in the coating. Scanning electron microscopy (SEM) is a key analytical technology for evaluating the surface morphology and quality of the coating.

[0003] Although electron scanning microscopy provides extremely detailed surface images for materials science research, traditional image analysis methods mainly rely on manual operations, which is not only inefficient but also limits the accuracy and repeatability of the results. Manual processing of large amounts of image data is time-consuming and prone to errors, especially in the lack of efficient and accurate methods for statistical analysis.

[0004] Therefore, the present invention proposes an automated electron microscope image processing and analysis system for automatically processing and analyzing electron microscope scanning images of metal micro-arc oxidation coatings. The system significantly improves the efficiency and accuracy of analysis and reduces the possibility of human error by automatically performing image preprocessing, edge detection, morphological operations, crack identification and length calculation, and crack density estimation. The development of this system is of great significance for the quantitative analysis of crack defects in micro-arc oxidation coatings. Summary of the invention

[0005] Based on the shortcomings of the prior art, the purpose of the present invention is to provide an efficient automated method for processing and analyzing the surface images of metal micro-arc oxidation coatings obtained by electron microscope scanning. The method aims to improve the accuracy and efficiency of image analysis, especially for the statistical analysis of crack density.

[0006] In order to solve the above technical problems, the present invention provides an automatic crack density analysis method for electron microscope scanning images of metal micro-arc oxidation coatings, which is characterized by including image preprocessing, edge detection, morphological operation, crack identification and length calculation, and crack density estimation. Specifically, the following steps are included: image preprocessing, converting RGB images into grayscale images, and applying contrast-limited adaptive histogram equalization (CLAHE) to improve image contrast, so as to better highlight the visualization of cracks; image segmentation, thresholding processing is performed by Otsu algorithm to achieve effective separation of foreground and background, and Canny edge detection algorithm is used to further identify and highlight the edges of cracks; morphological operation, applying closed operation and expansion processing to remove small noise in the image and connect the broken edges of cracks to form continuous crack lines; crack detection and measurement, detecting and extracting straight line segments in the image through Hough transform, these straight line segments represent cracks, and calculating the total length of these line segments; crack density calculation, converting the total length of the detected cracks into actual micrometer length, and calculating the crack density according to the actual area of ​​the image.

[0007] The implementation steps of the present invention are as follows: S1, image preprocessing stage: Grayscale processing: Convert the color image to a grayscale image. The grayscale formula used is Y=0.299R+0.587G+0.114B. This step simplifies the subsequent image processing steps and effectively reduces the computational complexity. Contrast enhancement: Use contrast-limited adaptive histogram equalization (CLAHE) for image enhancement. Set clipLimit to 2.0 and tileGridSize to (8, 8) to improve the image quality and make subsequent edge detection more accurate.

[0008] S2, image segmentation and edge detection stage: Threshold processing: The Otsu algorithm is used to automatically determine the optimal threshold for image binarization, effectively separating the foreground from the background, and facilitating further identification of microscopic features. Edge detection: The Canny edge detection algorithm is used for edge identification, and dual thresholds (low threshold 30, high threshold 100) are used to accurately identify and highlight the edges of cracks.

[0009] S3, morphological operation and feature extraction stage: Morphological operations: Closing and dilation (iteration 2) were performed using a 3x3 structure element to remove small noise points and connect broken crack edges to improve edge continuity. Connected region analysis: Connected region analysis was used to identify and classify independent connected regions of cracks in the image, using an 8-connectivity algorithm to ensure accurate identification of all microstructures.

[0010] S4, crack length calculation and density estimation stage: Crack length calculation: Use Hough line transform to detect lines, with a threshold of 30, a minimum line length of 20, and a maximum line gap of 5 to identify and count crack line segments in the image. Crack density estimation: Calculate the total length of all crack line segments, convert the length from pixel units to micron units, use the actual size of the image (pixel length converted to 207 microns), and then calculate the crack density based on the total image area.

[0011] The automated crack density analysis method of the present invention can significantly improve the efficiency and accuracy of metal micro-arc oxidation electron microscope scanning image analysis, and has important application value in the fields of material science research and industrial quality control. The advantages of this method include: 1. Improve processing efficiency and accuracy: Through an integrated automated process, including image preprocessing, edge detection, morphological processing, and crack length calculation, the present invention greatly speeds up crack analysis while improving accuracy. Automated feature recognition and crack measurement avoid the tediousness and common errors of manual processing, ensuring consistency and repeatability of results.

[0012] 2. Enhanced data analysis capabilities: The present invention automatically counts crack length and calculates crack density, providing detailed quantitative data for understanding the microstructure of materials. These precise data not only help researchers better evaluate material properties, but also support engineers in making more accurate decisions during material selection and improvement.

[0013] 3. Ease of operation: The software based on Python and OpenCV supports batch processing. Users can automatically complete the analysis of a large number of images through simple interface operations, which significantly reduces the technical threshold and operation difficulty and reduces time costs.

[0014] 4. Wide application potential: Although the present invention is mainly aimed at crack analysis of metal micro-arc oxidation surface, its methods and techniques can also be extended to other types of material surface treatment analysis. Its flexibility and scalability enable it to meet the needs of different industrial and scientific research fields.

[0015] In summary, the automated crack density analysis method provided by the present invention not only improves the analysis efficiency and accuracy, but also optimizes the operation process. It has broad application prospects and practical value, and is particularly suitable for material science research and industrial applications that require fast and accurate crack data. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the drawings of the embodiment are briefly introduced below.

[0017] Figure 1 Flow chart of the present invention Figure 2 The grayscale image of the present invention

[0018] Figure 3 The enhanced image of the present invention

[0019] Figure 4 The binary image of the present invention

[0020] Figure 5 The closed operation edge image of the present invention

[0021] Figure 6 The final edge image of the present invention DETAILED DESCRIPTION

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments are briefly introduced below. Part of this specification illustrates the principles of the present invention through embodiments, and other aspects, features and advantages of the present invention will become clear through this detailed description. In the drawings referred to, the same or similar parts in different figures are represented by the same reference numerals.

[0023] like Figures 1 to 5 As shown, the automated metal micro-arc oxidation electron microscope scanning image processing method provided by the present invention aims to solve the problems of low efficiency and insufficient accuracy of existing manual image processing, and improves the efficiency and accuracy of analysis through an automated process. The following are the specific steps of using the method: S1. Image acquisition: Use an electron microscope to scan the sample after metal micro-arc oxidation treatment to obtain high-resolution image data.

[0024] S2. Image preprocessing ( Figure 1 ): Convert the acquired RGB image to a grayscale image using the formula Y = 0.299R +0.587G + 0.114B to simplify the complexity of the image. Use contrast-limited adaptive histogram equalization (CLAHE) to enhance the contrast of the image, set clipLimit to 2.0 and tileGridSize to (8, 8) to remove random noise in the image and provide clearer image quality for subsequent analysis.

[0025] S3. Image segmentation ( Figure 2 ): The Otsu algorithm automatically selects the optimal threshold for binarization processing, effectively separating the foreground and background in the image and highlighting important microstructures.

[0026] S4, edge detection ( Figure 3 ): Using the Canny edge detection algorithm, the dual thresholds are set to 50 and 150 to accurately identify and highlight the edges of the cracks.

[0027] S5. Morphological processing ( Figure 4 ): Perform morphological closing operation on the image after edge detection, that is, corrosion first and then expansion, and use 3x3 structure elements to improve the continuity and clarity of the cracks.

[0028] S6. Connected Region Analysis ( Figure 5 ): Connected region analysis is performed on the morphologically processed images, and the 8-connectivity algorithm is used to mark and classify independent cracks in the images to provide basic data for statistical analysis.

[0029] S7. Statistical analysis: Automatic counting, using image processing algorithms to count the number of independent cracks in the image. Crack length sum calculation, calculate the ratio of the sum of all crack lengths divided by the total image area, automatically output crack density, and provide key indicators for material performance evaluation.

[0030] S8. Result output: After the analysis is completed, the system automatically generates a report containing statistical data such as the number of cracks and crack density, and outputs the results in CSV file format for further data analysis and archiving.

[0031] The above steps and drawings show the detailed implementation of the present invention. Through these steps, the present invention not only improves the efficiency and accuracy of crack analysis, but also optimizes the operation process. It has broad application prospects and practical value, and is particularly suitable for material science research and industrial applications that require fast and accurate crack data. The above is only a preferred implementation of the present invention, and of course it cannot be used to limit the scope of rights of the present invention. It should be pointed out that for ordinary technicians in this technical field, without departing from the principle of the present invention, several improvements and changes can be made, and these improvements and changes are also regarded as the protection scope of the present invention.

Claims

1. A crack density calculation method for automatically processing and analyzing metal micro-arc oxidation electron microscope scanning images, characterized in that: It includes image preprocessing, edge detection, morphological operation, crack identification and length calculation, and crack density estimation; the image preprocessing converts the RGB image into a grayscale image and applies contrast-limited adaptive histogram equalization (CLAHE) to enhance the image contrast; The edge detection uses the Canny algorithm to perform edge detection and combines multi-scale edge detection to enhance the visualization of crack edges. The morphological operation includes using closing operations and dilation processing to remove small non-crack areas in the image and connect broken crack edges. The crack identification and length calculation uses Hough line transform to detect straight line segments in the image and calculate the total length of the cracks represented by these straight line segments. The crack density estimation converts the total length of the detected cracks into micrometer scale and calculates the crack density based on the actual area of ​​the image.

2. The method according to claim 1, characterized in that The clipLimit used in the CLAHE step in the image preprocessing is 2.0, and the tileGridSize is (8, 8); the dual thresholds used by the Canny algorithm in the edge detection step are 50 and 150 respectively; the kernel size used in the morphological operation is a 3x3 structural element, and two iterations are performed; the crack density estimation step further includes converting the pixel length into the actual micrometer length through the proportional relationship between the pixel in the image and the actual length.

3. A method for using the method for automatic processing of metal micro-arc oxidation electron microscope scanning images and crack density analysis as described in any one of claims 1 to 2, characterized in that: The following steps are involved: S1, preparation stage: load the electron microscope scanning image to be analyzed into the analysis system, and the system automatically performs image preprocessing, including converting the RGB image into a grayscale image and applying CLAHE to enhance the image contrast; S2. Edge detection and morphological operations: The system uses the Canny algorithm for edge detection and performs morphological operations to remove non-crack areas and crack edges that connect fractures; S3. Crack identification and length calculation: The system detects crack straight line segments in the image through Hough line transform and calculates the total length of these straight line segments; S4. Crack density estimation: The system converts the total length of the crack into micron scale, calculates the crack density based on the actual area of ​​the image, and generates analysis results; S5. Report generation and export: The system presents the analysis results in the form of a visual report and supports export to PDF or Excel format for further analysis and sharing; S6. Batch processing: The system supports batch loading and analysis of multiple image files, automatically processes each file and displays all processing results in a centralized manner.