基于自适应模糊阈值分割的韧窝图像处理

为了能准确研究韧窝直径与常规力学性能之间的关系，尽量减小在测量中人为因素的影响，本文采用计算机图像处理技术对韧性断裂产生的等轴韧窝图像进行辅助分析，应用自适应模糊阈值分割法对预处理后的韧窝图像进行二值分割，很好的实现了韧窝与背景的分离。并且根据韧窝直径的大小对韧窝进行了分类。为了能快速准确的获得韧窝直径，对基于Freeman链码的多区域面积的算法进行了改进，改进后的方法实现简单，计算量小，测得韧窝直径的结果准确。

Image Processing of dimples Base on Image Adaptive Threshooding by Index of Fuzziness

In order to minimize the impact of man making and to study the relationships of diameter of dimple and some mechanical properties, in this paper we make use of computer image processing technology to research image of dimple which was formed by the ductile fracture.. And we divorce the dimple from the background very well, by using the adaptive thresholding by index of fuzziness to make the image binarization. Then we classify the dimple by its diameter. To be able to quickly and accurately access dimples diameter, we improve the method for calculating areas of multiple regions in image based on freeman chain code. Compared with the old algorithm, its required options are reduced and its result is good.