이웃 픽셀 값 기반의 가중치지도를 이용한 표적 검출 방법

Abstract

In this paper, we propose a small target detection method using weighted map based on neighboring pixel values to detect small targets of various sizes in infrared images. Small target has characteristics that it is brighter than background and the brightness change is large in images. Using these properties, the proposed method generates brightness change map using input image. Through the weighted maps of the input image and brightness map, the target is improved and the background is suppressed. A target enhanced map is generated using the two weighted maps. Finally, the target is detected through thresholding. Experimental results show that the proposed method has better detection performance than the existing methods.