快速高效无损图像压缩系统的低功耗硬件实现

发布时间：2018-04-27 06:02

本文选题：快速高效无损图像压缩系统 + Golomb-Rice编码　；参考：《哈尔滨工程大学学报》2014年03期

【摘要】：针对快速高效无损图像压缩系统(FELICS)中,Golomb-Rice编码k参数选取增加系统复杂度和限制系统吞吐量的问题,依据基于上下文的图像压缩理论,采用JPEG-LS中的序列参数估计方法,研究了限长Golomb-Rice编码。给出了面向超大规模集成电路(VLSI)实现的低功耗快速高效无损图像压缩系统。VLSI-oriented FELICS简化了k参数的选取,提高了系统的吞吐率;针对贝尔图像格式的算法扩展提高了FELICS算法对Bayer图像的压缩效果。在SMIC 0.13μm工艺条件下,基于该方法完成了面向低功耗胃窥镜的VLSI设计。当系统工作在25 MHz,图像数据时钟为24 MHz时,编码每帧图像的功耗仅为11.15μW,VGA图像的吞吐率可以达到60 f/s。
[Abstract]:Aiming at the problem of how to select parameters of Golomb-Rice coding k to increase system complexity and limit system throughput in fast and efficient lossless image compression system, according to context-based image compression theory, this paper adopts the method of sequence parameter estimation in JPEG-LS. The limited length Golomb-Rice coding is studied. A low power fast and efficient lossless image compression system. VLSI-oriented FELICS for VLSI (VLSI) is presented, which simplifies the selection of k parameters and improves the throughput of the system. The algorithm extension for Bell image format improves the compression effect of FELICS algorithm to Bayer image. Under the condition of SMIC 0.13 渭 m, the VLSI design for low-power gastroscope is completed based on this method. When the system operates at 25 MHz and the image data clock is 24 MHz, the power consumption of each frame image is only 11.15 渭 W / VGA image throughput can reach 60 f / s.
【作者单位】：中国科学院微电子研究所;
【基金】：国家863计划资助项目(2008AA010704)
【分类号】：TN919.81

【相似文献】