CAJ | 학술논문

针对快速场景变换环境中基于压缩感知的分布式视频编码所存在的问题,提出了一种基于自适应动态分组与压缩感知的分布式视频编码方法。对采集的视频帧在基本分组的基础上插入一个自适应分组分离器，通过设定门限阈值实现能够根据视频场景变换的动态调整分组。实验表明，与基本分组方法重构效果相比，采用自适应动态分组方法的重构效果有了较大改进，其峰值信噪比也有大大改善和提高。有效地解决了视频帧中场景变化较快造成的关键信息边缘纹理损失的问题。
침대쾌속장경변환배경중기우압축감지적분포식시빈편마소존재적문제,제출료일충기우자괄응동태분조여압축감지적분포식시빈편마방법。대채집적시빈정재기본분조적기출상삽입일개자괄응분조분리기，통과설정문한역치실현능구근거시빈장경변환적동태조정분조。실험표명，여기본분조방법중구효과상비，채용자괄응동태분조방법적중구효과유료교대개진，기봉치신조비야유대대개선화제고。유효지해결료시빈정중장경변화교쾌조성적관건신식변연문리손실적문제。
In this paper, we proposed an adaptive grouping and compressive sensing distributed videocodingmethod to the problem of the current distributed video encoding under fast changing scene. On the basis of the basic grouping method, adaptive grouping separator sets the threshold to achieve the purpose of changing groups dynamically. Experiments show that the adaptive method has been greatly improved compared to the basic grouping method, and the PSNR has significantly improved and enhanced, it can effectively solve the problem of the video frame edge of key information in the rapid scene.