噪声信道的鲁棒图像编码

Robust SPIHT Image Coding for Noisy Channels

在一鲁棒图像编码方案的基础上提出了一种改进方案。通过改变部分比特输出的先后次序 ,保留它将 SPIHT输出比特分类以体现各类信息的重要性及出错敏感性 ,使用不等保护能力纠错编码 ,并且利用图像小波系数数据幅度的非均匀分布规律 ,解码时添加后处理步骤 ,将小波系数适当减小 ,得到更高质量的图像。该方案算法简单 ,在实现鲁棒编码结构的同时提高了压缩效率。失真率分析和实验结果都表明该方案提高了峰值信噪比 。

In this paper, the authors put forward an improved scheme based on robust image coding algorithm. The fundamental approach is to change partly the encoder's output bit sequence into subsequences whereas to remain the classification of SPIHT encoder's output bit according to their importance and error sensitivity. And this method is to protect differently the error-correcting encoding and with the non-uniform distribution of wavelet coefficients, decoder gets its higher quality image through post-processing and appropriately diminishes the wavelet coefficients. Both rate-distortion optimization analyses and experimental results indicate that this method achieves better compression performance with robustness.