摘要
针对电气化铁路接触网支撑装置中旋转双耳耳片断裂问题,提出一种基于快速鲁棒性特征SURF(Speeded-up Robust Features)匹配的图像检测方法。利用标准旋转双耳图像与待检测图像局部不变特征点的匹配,实现旋转双耳的初识别;应用Hough变换实现耳片精确定位;通过耳片局部图像中的灰度方差判断耳片是否存在断裂状态。实验表明,该方法能在复杂图像中较准确地识别耳片断裂特征,为旋转双耳的状态检测提供参考。
A novel image detection method based on speeded-up robust features（SURF） was proposed, in order to detect the fracture of the ear pieces of the rotary double ears of the catenary support device of electrified railway. Firstly, the local invariant feature points were matched between standard image of rotary double ears and sampled image to be detected to realize the initial recognition of the rotary ears. Secondly, the ears were accurately located through Hough transform. Lastly, the fracture of the ear pieces was distinguished by the gray variance in the local image of the ear piece. The experiments showed that this method can accurately identify the fracture features of the ears in the intricate graphics, and can offer a reference to the state detection of the rotary double ears.
出处
《铁道学报》
EI
CAS
CSCD
北大核心
2016年第8期28-34,共7页
Journal of the China Railway Society
基金
西安铁路职业技术学院课题(XTZY15G05)
关键词
旋转双耳
SURF
局部不变性
灰度方差
rotary double ears
SURF
local invariant feature
gray variance
