An algorithm of background extracting from low updated-rate CCTV image sequences

摘要

Closed-Circuit Television (CCTV) images are potential data source of Intelligent Transportation Systems (ITS). However, these images sequence's interval is from 2s to 300s, so it is difficult to extract background from these images sequences by conventional methods, let alone to directly estimated traffic information. In order to make use of such online low updated-rate CCTV image sequences, a background extract algorithm is proposed to subtract background from CCTV image sequence. The proposed method has been examined with data collected at one selected locations in Hong Kong under different traffic conditions. It is shown that the proposed algorithm can obtain satisfactory result.