Ir Image Weak Target Detection And Analysis

A complete infrared image processing system based on video collection board is designed for an engineering project in this dissertation, which could be used to detect and analyse dim target from infrared image with low SNR. New algorithms for dim target detection in infrared image are studied and proposed, it has been proved that the whole algorithm is quite effective in software actualize.With the view of image enhancement and segmenting a target accurately, OSTU and clustering thought are combined to give a practical method to select the threshold. The properties are the same or similar within regions while different in different regions. The maximum distance is adopted to select threshold automatically. Wavelet transformation method is used to enhance modulus of high frequency subbands at each decomposition. The experiment result prove that the method is practical.