在不同型号的光学显微镜上采集了多种熔喷材料图像,利用大律法全局阈值、LOG和CANNY算子边缘检测及边缘检测基础上的阈值分割对其进行处理,研究表明光学显微镜下熔喷材料的图像自动分析难以实现.针对扫描电镜(SEM)图片,提出一种基于边缘检测、图像分割、细化、线段连接后配对的熔喷超细纤维直径分布测试算法,验证试验表明该方法能准确地测量熔喷超细纤维的主体直径.
A large number of images of different area densities of melt-blown webs are captured in optical microscope, and processed with Otsu's global threshold, LOG, CANNY edge detection and threshold segmentation based on edge detection operator. The experiment indicates that the idea of automatic analysis using optical microscope images is difficult to achieve. Based on edge detection, image segmentation, thinning, line linking and matching, a series of algorithms for measuring microfiber diameter in melt-blown web is proposed for scanning electron microscope (SEM) images. Experiment indicates that the proposed method can get the main diameter of fiber accurately.