摘要
为了消除光照不均对刀具磨损检测的影响,在研究刀具磨损特点的基础上,提出了一种刀具磨损检测的方法。首先,对刀具图像灰度值取对数并进行小波的单层变换;其次,对小波单层分解的近似子图进行同态滤波以消除光照不均的影响;然后,对滤波后的近似子图、水平细节及垂直细节子图进行标准化处理;在此基础上,对分解的各子图进行选择性的图像融合处理;最后,通过最大类间方差法进行图像分割。实验表明:所采用的方法能够有效抑制光照不均的干扰,能够有效的实现刀具磨损检测。
In order to eliminate the influence of illumination, a kind of tool wear detection method is put forward through analysis of tool wear characteristics. First of all, the logarithm of tool image gray value is decomposed into a se- ries of sub-graph through the wavelet transform. Secondly, approximate sub-graph of wavelet decomposition is treated by homomorphism to eliminate the influence of illumination. Then, the approximate sub-graph, level detail and vertical de- tail sub-graphs are standardized. On this basis, sub-graphs are selected for image fusion. Finally tool wear is detected through the segmentation method of maximum between-cluster variance. Experiments show that this method can effective- ly suppress the image background interference, and effectively realize the rapid detection of tool wear.
出处
《机械科学与技术》
CSCD
北大核心
2013年第11期1703-1707,共5页
Mechanical Science and Technology for Aerospace Engineering
基金
陕西省教育厅科研计划项目(2013JK1083)
西安工程大学博士科研启动基金项目(BS1005)资助
关键词
刀具磨损
小波变换
同态滤波
图像分割
磨损检测
algorithms
efficiency
experiments
imagewavelet decomposition
wavelet transforms
homomorphism filtering
wear detectionfusion
image segmentation
standardization
tools
wear of materials
tool wear
wavelet transforms
