摘要
感兴趣区域(ROI)编码与图像压缩算法的相结合,可使ROI获得高质量。本文通过在多级树集合分裂(SPIHT)算法中嵌入ROI,形成基于SPIHT算法的ROI图像编码方法。该方法通过下移背景(BG)位平面使ROI位平面置于BG位平面之前,解码时优先传输ROI系数。实验选取一般位移法,将此算法与传统的SPIHT算法作比较,并研究了不同条件下各区域的质量。实验证明,本文算法明显优于传统的SPIHT算法,选择合理的参数能得到好的视觉效果。
The combination of Region of Interest（ROI） coding and image compression algorithm can make ROI get high quality. ROI is embedded into Set Partitioning in Hierarchical Trees（SPIHT） algorithm in this paper, it forms an ROI coding method based on SPIHT algorithm. This method shifts background（BG） bit planes to make ROI bit planes place before BG bit planes, then ROI coefficients will be transmitted firstly when decoding. Scaling-based method is used for experiments, this paper compares this algorithm with traditional SPIHT algorithm and studies the quality of every region under different conditions. Experiments show that this algorithm is superior to the traditional SPIHT algo-rithm, choosing reasonable parameters can get very good visual effect.
出处
《软件》
2015年第2期112-117,共6页
Software
