论文部分内容阅读
结合数学形态学运算方法,对于板材缺陷图像的提取分割,提出了基于数学形态学的分割方法。利用图像各点间的像素值差异,对板材像素点求取无缺陷标准值,将该值与待测板材进行对应像素点对比,然后将差值求和取平均得到分割阈值;对图像中所有大于该阈值的点进行分割提取,得到分割图像。用数学形态学方法对噪声较大的图像进一步处理,最终得到较为理想的分割图像。
Combined with the mathematical morphology computing method, the segmentation method based on mathematical morphology is proposed for the extraction and segmentation of the sheet defect images. Using the pixel value difference between the points of the image to obtain the defect-free standard value of the pixel of the plate, comparing the value with the corresponding pixel of the plate to be tested, and then taking the average of the difference to obtain the segmentation threshold; A point larger than this threshold is subjected to segmentation and extraction to obtain a segmentation image. Using mathematical morphology method to further process the noisy image, finally obtain the ideal segmentation image.