【摘要】 针对复杂表面轮廓不规则、边缘部分灰度骤变等特征,在对图像预处理后,先采用迭代法从具有多层次灰度区域的待检测图像中自动获取待检测区域。然后对其选择合适形式的映射函数,拟合图像低频部分,通过差分法获取可能包含缺陷信息的高频部分。采用边缘检测、开运算与闭运算等方法,去除图像差分结果中轮廓信息与微小瑕疵,突出缺陷。实验结果表明,该方法可有效检测出复杂表面中裂纹缺陷,避免由微小瑕疵引起的误检,检测效果良好。更多还原

【Abstract】 According to the characteristics of irregular contour on complex surface and great change of the edge part gray level,a surface defect detection method based on surface fitting is proposed in this paper.After the preprocessing of image,the iterative method is used to get the area to be detected automatically from the raw image with multi-layer gray level.Then an appropriate mapping function is selected for the detection region to fit the low frequency part which may contain the defect information of the image,and the high frequency part is obtained by image difference technology.Edge detection,open operation,close operation and other methods are used to remove the contour and minor fault from the image difference and highlight the defects.The experimental results show that this method can effectively detect the crack defects in the complex surface,also avoid the false detection caused by minor fault,and the detection effect is reliable.更多还原