基于残差集的CFA图像序列无损压缩新算法
Novel Residual Set Lossless Compression Method of CFA Image Sequences
-
摘要: 提出了一种基于残差集的颜色滤波阵列(CFA)图像序列的无损压缩新算法. 通过构造CFA图像序列的残差集,去除图像序列间的数据冗余;设计了一种最佳可逆滤波器对残差集图像进行通道分离滤波,去除图像内的冗余信息,提高CFA图像压缩性能. 将该算法与通道分离滤波JPEG和PNG算法进行了对比,结果证明,残差集无损压缩算法具有最高的数据总量压缩比、最低的平均压缩率和最好的重建CFA图像峰值信噪比.Abstract: A novel residual set lossless compression (RSLC) method of color filter array (CFA) image sequences is presented in the paper. The redundant data between image sequences are compressed by building the image residual set. The images of residual set are channel-separated filtered to compress the redundant data within the image by the designed best reversible filters which can improve the performance of CFA image compression. The experimental results prove that the RSLC method carries the best data compression ratio, the lowest bits per pixel and the highest PSNR performance among the channel-separated filter JPEG method, PNG method and the RSLC method.
下载: