摘要
综合运用图像处理中的边缘检测、形态学处理、频域滤波、最优阈值分割等基本方法,在VC++6.0环境下,实现了轮毂表面在线自动探伤系统。该系统能有效地抑制背景光照不均和刀纹、水渍、油渍等干扰的不利影响,较好地实现了对裂纹特征的检测。实验表明,系统所采用的方法具有运算速度快、误判漏判率低、满足在线检测需求的特点。
In this paper, based on the algorithms of edge detection, morphological operation, frequency domain filtering and optimal threshold segmentation,have constructed an on- line automatic flaw inspection system of the train nave surface by using VC＋ ＋ programming＇ language. The system works well for effectively restraining background asymmetric illumination, blade lines, water spots, oll stains, etc, while getting satisfying results of crack detection. The experiments indicate that the proposed method in the system is computationally efficient, has low flaw missing rate and false alarm rate,and meets the need of on- line flaw inspection.
出处
《计算机技术与发展》
2007年第4期214-217,共4页
Computer Technology and Development
基金
安徽省人才开发基金(2001Z021)
安徽大学211工程学术创新团队项目
关键词
磁粉探伤
边缘检测
形态学
频域滤波
阙值分割
裂纹
magnetic particle flaw inspection
edge detection
morphology
frequency domain filtering
threshold segmentation
crack
