一种易于硬件实现的图像有损压缩算法  

A Novel Method for Image Lossy Compression Convenient for Hardware Implementation

作　　者：沙莎[1] 邓家先[1] 易巧玲[1] 李莎[1] 

机构地区：[1]海南大学信息科学与技术学院,海南海口570228

出　　处：《光学与光电技术》2010年第2期19-23,共5页Optics & Optoelectronic Technology

摘　　要：原始图像数据经过小波变换后,根据子带系数之间在空间上和方向上所呈现出的带间相似性,将像素集合重新定义为集合树。通过对直流系数量化编码和交流系数比特深度编码中的参数K和熵编码中长度分别为3 bit码字(choicel)和4 bit码字(choice2)的编码方式的选择实现数据的渐进性编码,使产生的码流具有内嵌特性,可以对任意码流进行截取和译码,实现对码率的有效控制。提出了一种具有复杂度低、易于硬件实现的图像有损压缩算法。经过系统仿真证明此算法有利于提高重建图像的质量。After the DWT of the original image data. and according the similarity between the band shown by subband coefficients in the space and the direction, the collection of pixels is redefined as set tree. Through selecting the parameters K of DC coefficients quantization coding and AC bit depth encoding and choosing the code method of 3-bit words （choicel） or 4- bit words （choice2） in entropy coding implementation the data progressiveness coding, these code streams have embedded feature, then any code streams can be truncated and decoded, and the code rate can be controlled efficiently. A low complexity. easy for hardware implementation of image lossy compression algorithm is presented. The results prove that the quality of the reconstructed image is improved.

关 键 词：图像有损压缩算法 离散小波变换 比特平面编码 参数优化选择 

分 类 号：TN919.81[电子电信—通信与信息系统]