利用DCT变换和分段量化的成像声纳实时处理系统图像数据压缩技术研究  

Research on Imaging Data Compression in Real-time Sonar System Using DCT and Subsection Quantization

作　　者：江泽林[1] 张鹏飞[1] 刘维[1] 刘纪元[1] 

机构地区：[1]中国科学院声学研究所,北京100190

出　　处：《信号处理》2015年第1期1-7,共7页Journal of Signal Processing

基　　金：国家自然科学基金项目(11204343);中国科学院声学研究所青年人才领域前沿项目;哈尔滨工程大学水下机器人技术重点实验室基金项目(9140C27020112022601)

摘　　要：成像声纳实时处理系统在某些应用场合下传输图像数据时需要进行压缩,如搭载成像声纳的半潜式航行器与母船之间无线传输声纳数据。针对该应用需求,提出了利用离散余弦变换(DCT)和分段量化的声纳图像数据压缩技术。该方法处理的对象是成像声纳实时获取的图像行数据。方法的实施步骤为,首先对行数据进行DCT变换,其次利用DCT的能量集中特性,对变换后的DCT系数进行截断处理,最后对截断数据进行分段量化处理,进一步提高压缩效率。在显控端利用相逆的过程进行解压处理,实现成像结果的实时显示。对实际的声纳图像数据进行了处理,验证了方法的有效性。For the real-time imaging sonar,the imaging data require to be compressed to transmit to the littoral combat ship （LCS）in some cases such as in the wireless communication between a LCS and a remote multi-mission vehicle （RM-MV）.In this paper,a discrete cosine transform （DCT）and subsection quantization based imaging data compression meth-od was proposed,which could apply to the real-time sonar imaging row data.The steps of the method were as follows. Firstly the row data were transformed by DCT.Secondly the DCT sequence was truncated for its strong “energy compac-tion”property.Finally the truncated data were divided into several subsections,each of which was quantized separately.In the display end,which usually located in a LCS,after an adverse processing,the decompressed imaging raw data would display on the screen in real time.The method was applied to the lake trial sonar data,indicating this effective solution.

关 键 词：声纳 图像数据压缩 离散余弦变换 分段量化 

分 类 号：TB566[交通运输工程—水声工程]