作者: M.A. Zaveri , S.N. Merchant , U.B. Desai
DOI: 10.1109/ISCAS.2003.1205987
关键词:
摘要: It is difficult to detect and classify multiple point targets in the presence of clutter noise, an air borne IR image sequence. Most existing approaches assume a target several pixels size or Gaussian shape with variance 1.5 pixels. We propose wavelet based detection technique for single pixel dim sequence noise. Temporal multiscale decomposition used temporal change. This change map postprocessed reduce probability false alarm misdetection.