基于ROI编码的任意尺寸测量图像的压缩方法  

Efficient encoding method for arbitrary resolution photogrammetric image based on ROI coding

作　　者：杨晓[1] 刘俊杰 杨学友[1] 

机构地区：[1]天津大学精密测试技术及仪器国家重点实验室,天津300072

出　　处：《计算机工程与应用》2013年第4期14-17,共4页Computer Engineering and Applications

基　　金：国家自然科学基金资助

摘　　要：针对基于感兴趣区域(ROI,Region of Interest)的测量图像算法中存在的编码冗余、适用范围受限等问题,提出了一种适用于任意分辨率的测量图像压缩算法。该方法通过加入ROI中A类集合的分类,优化了码流的输出,进一步提高了算法的编码效率;根据小波变换的特点和测量图像对编码算法的要求,通过分区域编码解决了测量图像在编码过程中分辨率受限的问题。实验表明,方法实现了对任意分辨率测量图像的编码,并在不影响测量精度的前提下,获得了优于基于感兴趣区域的测量图像压缩算法的压缩性能。An efficient coding method for arbitrary resolution photogrammetric image is proposed to solve the problems of the compression method for photogrammetric image based on ROI（Region of Interest） in low coding efficiency and the limited applications. The proposed method optimizes the coding bit stream and further improves the coding efficiency by judging the type of A set in ROI. According to the characters of wavelet transform and the coding requirements for photogrammetric image, the problem of the limited resolution existing in the coding of photogrammetric image is resoNed by using the region coding. Experiment results show that the proposed method achieves the coding for arbitrary resolution photogrammetric image. Furthermore, the proposed method has achieved higher compression performance as compared with the compression method for photogrammetric im- age based on ROI without affecting the measurement precision.

关 键 词：任意分辨率 测量图像 感兴趣区域 小波变换 分区域编码 

分 类 号：TN919.8[电子电信—通信与信息系统]