高能闪光照相中Sobel算子的边缘检测方法

摘要

针对高能闪光照相中投影图像的特点,对数字图像处理中具有代表性的Sobel边缘检测算法进行了分析,发现该算法存在检测出的边缘粗和人为进行阈值的选取会造成边缘点的误判的缺点;为此提出了一种阈值确定方法和基于最大梯度的定位准则算法.改进的算法实测结果表明:该方法有效地解决了Sobel算子检测出的边缘过粗的问题,得到的边缘较细,定位精确,提高了边缘检测精度,可用于高能辐射照相中的边缘提取.%Edge detection of flash X-ray radiography image is a principal means to extract quantitative boundary information of primary implosion material. The precision of edge detection is an important index for judging flash X-ray radiography diagnostic level . In terms of feature of high-energy flash X-ray radiography image, typical Sobel edge detection algorithm in digital image processing is analyzed in this paper. A simple algorithm for threshold alternative and localization criteria based on maximum gradient is proposed to overcome typical Sobel operater's disadvantages of the obtained image edges being thick and causing edge point misjudgement. The simulated result shows that the obtained edges of the image are thin and the edge position extracted by the algorithm is accurate.