JPEG2000感兴趣区域编码的新方法

A NEW REGION-OF-INTEREST COMPRESSION METHOD FOR JPEG2000 IMAGES

对于感兴趣区域编码一直是图像压缩的热点之一。结合新一代静止图像压缩标准JPEG2000,介绍了JPEG2000中两种经典的ROI压缩算法:最大化位移法和一般位移法,分析了这两种方法的实现原理和优缺点,提出了一种新的JPEG2000感兴趣区域编码方法。该方法利用JPEG2000中"失真可伸缩性"的特点,对于感兴趣区域和背景区域给定的目标质量,在质量层上根据给定的目标质量对包进行丢弃处理以达到压缩的目的,没有经典ROI(Region Of Interest)压缩方法的移位处理,所以压缩效率略好于经典压缩方法,压缩时间小于经典ROI方法所需的时间。实验结果表明,在相同的压缩比特率下。

The compression of region-of-interest is a hot topic. According to JPEG2000, a new still image compression standard, two common ROI compression methods, max-shift method and general scaling-based method are introduced, and the working principle, the advantage and disadvantage of the two methods are analyzed. A new region-of-interest compression method based on distortion scalable property of JPEG2000 is proposed. In this method, ROI（ Region of Interest） and background packets are deleted according to the desired quality parameters for ROI and background. Bits shifting used in common ROI compression methods is avoided. Satisfied compression effect is achieved, and the compression time is reduced. The experimental results show that the new method has better compression effect and takes less compression time than the common ROI compression methods in the same compression bit rate.