摘要
红外图像的空间关联性强,并且存在一定的非均匀性,因此诸如Canny算子之类的传统边缘提取方法并不适用。本文讨论了像素级和亚像素级结合的边缘检测方法,首先采用脉冲耦合神经网络(pulse coupled neural network,PCNN)方法进行像素级边缘定位,再结合空间灰度矩的方法进行亚像素级边缘细分。该方法能够对高温构件的红外边缘进行快速检测,并能极大提高边缘的定位精度。
The spatial correlation of infrared image is strong.And there usually exists non-uniformity in infrared images.So the traditional edge detection methods such as Canny operator are not suitable for infrared images.This paper focuses on the infrared image pixel and sub-pixel edge detection methods.Firstly,the pulse neural network method is used to locate the pixel-level edge.Then space gray moment method is introduced to subdivide the sub-pixel of the edge.This approach enables fast edge detection of heat elements,and can greatly improve the edge location precision.
出处
《激光与红外》
CAS
CSCD
北大核心
2012年第5期526-529,共4页
Laser & Infrared
基金
国家自然科学基金项目(No.60874106
No.51175377)资助
