摘要
针对目前轴承荧光磁粉探伤人工观察方法存在的问题,介绍一种轴承荧光磁粉探伤结果的自动识别方法,详细讲述了裂纹图像识别处理方案的原理及流程。采用选定色彩范围的方法获取图像可疑成分,主要使用数学形态学的方法对可疑图像进行处理,通过计算图像连通区域的大小和圆形度,并与设定值比较,以确认裂纹。算法在Matlab中进行测试,证实可有效识别出裂纹图像。
Aim at the existing problems of the manual inspecting in bearing fluorescent magnetic detection,an automatic identification method on the result of bearing fluorescent magnetic detection is introduced, the principle and flow of the crack pattern recognition processing plan are narrated in detail, image suspicious ingredient by adopting the color scope, mainly using mathematical morphology method to carry on the image processing,calculate the regional image＇s size and roundness then confirm the crack by comparing them with the specified value. The algorithm is tested in Matlab,eonfirming that may distinguish the crack image effectively.
出处
《现代电子技术》
2009年第7期107-110,共4页
Modern Electronics Technique
关键词
荧光磁粉探伤
图像处理
数学形态学
自动识别
fluorescent magnetic detection
image processing
mathematical morphology
automatic identification
作者简介
张强 男，1986年出生，河北人，硕士生。研究方向为检测技术与自动化装置。
