后向自适应的红外视频无损压缩算法

Backward-adaptive Lossless Compression of Infrared Video

红外视频广泛用于侦测、监控,在安全领域发挥重要作用。针对红外视频存储所需空间大,不便于传输等问题,本文提出针对红外视频的多预测模式无损压缩算法。首先分解红外视频为序列帧并对各帧图像分块,通过对时间和空间去相关预测;对预测获取的残差进行计算,对于依旧存在相关性的残差进行二次预测,对每个子块选择最优预测器;对处理好的残差进行熵编码,完成本算法。对多个红外视频分别使用本文算法、Gzip算法、LOCO-I算法进行比较,结果表明本文算法可以获得更高的压缩比,有效改善红外视频存储问题,实现高效信息的传输。

Infrared video is widely used in detection and monitoring and plays an important role in the security field. In view of the sizeable space required for infrared video storage and the inconvenience of transmission, this work proposes a multi-prediction mode lossless compression algorithm for infrared video. First, the infrared video is decomposed into sequence frames, and each frame is divided into blocks. The correlation prediction is performed as a function of time and space and the residual obtained by the prediction is calculated. Then, the residuals with correlations are predicted for each sub-block. The optimal predictor is selected and the processed residual is entropy encoded to complete the algorithm. The proposed algorithm, the Gzip algorithm, and the LOCO-I algorithm are compared in multiple infrared videos. The results show that the proposed algorithm can obtain a higher compression ratio, effectively improve the infrared video storage problem, and realize efficient information transmission.