Visual inspection method for surface defects of industrial products based on gray level co-occurrence matrix and ransac
A grayscale co-occurrence matrix and visual detection technology, applied in image analysis, image enhancement, instruments, etc., can solve the problems of low detection accuracy, complex calculation, narrow application range, etc., and achieve high detection efficiency, high detection accuracy, and stability. Good results
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment approach 1
[0023] Specific implementation mode one: as figure 1 As shown, the visual detection method for surface defects of industrial products based on gray level co-occurrence matrix and RANSAC includes the following steps:
[0024] Step 1: The image collected by the industrial camera (such as figure 2 shown) to perform grayscale and median filtering operations;
[0025] Step 2: For the image after grayscale and median filter operation in step 1, use the pre-stored image template to match and locate the surface block to be detected on the image after median filter operation, and perform Rotate so that the orientation of the surface block to be tested is consistent with the surface block to be detected in the template image;
[0026] The pre-stored image template is obtained by collecting the same detection surface of a standard industrial product;
[0027] Step 3: Equally segment the image of the surface block to be detected obtained in step 2 to obtain N partial image regions, th...
specific Embodiment approach 2
[0032] Embodiment 2: This embodiment differs from Embodiment 1 in that the value of H×W in step 3 is less than 1 / 2 of the smallest defect area in the surface block to be tested.
[0033] When the values of H and W are small, the detection accuracy of the detection algorithm for the area of the defect area on the surface block to be detected is relatively high; on the contrary, when the values of H and W are large, the detection accuracy of the area of the defect area on the surface block to be detected is relatively high. The detection accuracy is low.
[0034] When the values of H and W are small, it means that the subdivision degree of the detection algorithm for the surface block to be detected increases, and the result will increase the calculation amount of the algorithm; The subdivision degree of the detection surface block is reduced, which will reduce the calculation amount of the algorithm.
[0035] Other steps and parameters are the same as those in Embodi...
specific Embodiment approach 3
[0036] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is: the solution method of four feature quantities in the described step four is:
[0037] Contrast C:
[0038]
[0039] In the formula, i and j are the row and column coordinates of the normalized gray-level co-occurrence matrix; p(i, j) is the value of the i-th row and j-column in the normalized gray-level co-occurrence matrix;
[0040] Energy E:
[0041]
[0042] Relevance R:
[0043]
[0044] where u i is the average value of the row coordinates in the normalized gray-level co-occurrence matrix; u j is the average value of the column coordinates in the normalized gray level co-occurrence matrix; σ i is the standard deviation of the row coordinates in the normalized gray-level co-occurrence matrix; σ j is the standard deviation of the column coordinates in the normalized gray level co-occurrence matrix;
[0045] Homogeneity B:
[0046]
[0047] Other ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More