Detection Of Aerial Moving Small Dim Infrared Targets Under Sea-sky Background

With the development of photo election technique, computer technique and detector technique, more and more applications are based on the automatic target recognition (ATR) technique。The task of infrared target detection is an important and difficult issue in automatic target recognition system. The module of target detection is in the front of ATR system whose performance mostly depends on the target detection. The weak target with size of several pixels or dozens of pixels embedded in complicated background has no obvious geometrical structure and gray-scale distribution. The disturbance of clutters makes the target detection even harder. This dissertation deeply studies fusion detection for small weak target in background of sea and sky, The main contributions of this thesis are given below.The infrared small target detection process generally is divided into pre-processing, segmentation and multi-frame confirmation. In this paper, a detection scheme about the infrared small target with low SNR is designed in allusion to the characteristics of the sea and the sky background. Firstly, compared with a variety algorithm of pre-treatment, the median filter is used for pre-processing which has higher SNR than other filters.Sea-sky-line detection in sea-sky background image or video sequences contributes to applications such as small moving object detection and tracking, flight control and port security. A new2-layered structure method based on horizontal projection and improved Hough transformation is proposed in this paper. In the first layer, sea-sky-line region is located using horizontal projection method, which is a preprocessing step to reduce the disturbance of complex background. In the second layer, sea-sky-line is detected with kernel-based Hough transformation algorithm in the sea-sky-line region. The experimental results indicate that the proposed method is fast and robust for both infrared images and visible light images.In the last chapter, the detection of moving small targets in image sequences is considered. Based on the moving characteristics of small target SNR image sequences. A judging method based on neighborhood is discussed and used. Emulation results show this method can distinguish small target from sequences effectively; Further more, it has lower computation requirements and uses fewer images for detection as well.