Automatic Segmentation of FLIR Image Based on Gradient Operator
In the FLIR automatic infrared object tracingsystem,the gradient histogram of the FLIR image is unimodalitybased on tremendous statistics of the gradient histogram of FLIRimage. Moreover the gradient histogram of both objects andbackground can be fitted by z2 distribution density functionwith different degree of freedom. According to the abovestatistical characteristics of the gradient histogram and the leasterror-classified probability, an automatic estimation method ofgradient histogram segmentation threshold is presented andvalidated in this paper. And the average gradient is theapproximation of the optimal segmentation threshold throughtheoretical analysis. Finally this method is proved to be effectiveand practical by the segmentation experiments of the FLIR imagein the real conditions.
Gradient operator histogram z2 distributionfunction Segmentation threshold
Xinsai Wang Jing He
Infrared & Imaging Navigation Technology Lab Air Defense Forces Command Academy Zhengzhou,450052,P.R.China
国际会议
宜昌
英文
79-82
2010-10-10（万方平台首次上网日期，不代表论文的发表时间）