作者: Bahadir Karasulu , Serdar Korukoglu
DOI: 10.1007/S11042-010-0591-2
关键词:
摘要: Digital video content analysis is an important item for multimedia content-based indexing (MCBI), retrieval (CBVR) and visual surveillance systems. There are some frequently-used generic object detection and/or tracking (D&T) algorithms in the literature, such as Background Subtraction (BS), Continuously Adaptive Mean Shift (CMS), Optical Flow (OF) etc. An problem performance evaluation absence of stable flexible software comparison different algorithms. This able to compare them with same metrics real-time at platform. In this paper, we have designed implemented well-known D&T (for people D&T) The works automatic semi-automatic test environment real-time, which uses image processing essentials, e.g. morphological operations filters, ground-truth (GT) XML data files, charting/plotting capabilities