An efficient two-stage error detector based on syntax and continuity

doi:10.1109/TCE.2007.4429212

An efficient two-stage error detector based on syntax and continuity

2007-11-01

5

PRE

AI

DOI：10.1109/TCE.2007.4429212

原文链接

原文求助

摘要

En 中文

In video transmission, the compressed signal is extremely vulnerable to bit errors. Error concealment is widely used in video decoder to recover corrupted frames. Due to the use of variable length codes (VLCs), a single bit error in block-based bitstream may propagate to following of this slice. Especially, the effectiveness of error concealment technique relies heavily on correctly locating errors. In this paper an error detector based on syntax and continuity which consists of two stages is proposed. The first stage is to find erroneous slice according to checking constraints imposed on video bitstream syntax or detecting special error patterns. In the second stage, the precise position of the first erroneous macroblock (MB) is found based on the continuity of spatial or temporal data of neighbor MBs. Simulation results demonstrate that the proposed method improves performance of error detection significantly compared with the traditional syntax-based method(1).

Keyword:

error detection (ED)

error concealment

special error pattern

spatial continuity

motion compensation

An efficient two-stage error detector based on syntax and continuity

摘要

期刊

机构

引用论文