An Efficient Small Target Detection Method using Temperature Non-Linear Distribution and Double-Scale NWTH transformation in IR Image

摘要

In this paper, we proposed detection method of the small target using the temperature non-linear distribution elimination and double-scale New White Top-Hat (NWTH) transformation in Infrared (IR) image. At first, the IR image eliminates the temperature non-linearity noises using that atmosphere temperature is non-linear distribution with the rise of the ground height by a convective flow. And then, the double-scale NWTH transformation detects candidate targets and suppresses background clutters using double-scale structuring elements. Finally, the small target detected by comparing the Local Signal to Background Ratio (LSBR) among candidate targets. Experimental results show that the proposed method has better target detection performance than the conventional methods.