基于FPGA的实时视频增强  

Real-time Video Image Enhancement Based on FPGA

作　　者：杨光[1] 李晶[1] 吴钟建[1] 陈咸志[1] 代俊[1] 罗镇宝[1] 曾军[1] 罗冠泰[1] 

机构地区：[1]西南技术物理研究所,四川成都610041

出　　处：《电视技术》2013年第19期19-23,共5页Video Engineering

摘　　要：在图像制导应用中,针对雾霾等恶劣天气情况下形成的低对比度图像中目标跟踪稳定性不高的问题,需要对输入的低对比度视频图像进行实时增强预处理。限制对比度自适应直方图均衡算法(Contrast Limited Adaptive Histogram Equalization,CLAHE)是一种增强效果较好、计算简单且易于并行实现的方法,重点研究CLAHE算法架构以及FPGA硬件实现过程,并在Xilinx公司的Virtex-5系列FPGA硬件平台和MODELSIM仿真软件上对算法进行了时序分析、增强效果验证。通过软件仿真和实际平台测试验证,设计正确可行,在实时增强的基础上有效提高目标跟踪精度。In allusion to poor precision of target tracking within image guidance application,real-time enhancement pretreatment is essential for input low-contrast image.Contrast limited adaptive histogram equalization is a technique with simple calculation and better enhancement effect and it is convenient for parallel implementation.This paper mainly discusses the algorithm architecture and hardware realization in FPGA,function and timing analysis,enhancement effect verification are based on the hardware platform of FPGA and the MODELSIM simulation software.Via software simulation and platform verification,proposed design is feasible and effective in improving tracking precision on the basis of real-time image enhancement.

关 键 词：视频图像增强 CLAHE 实时性 FPGA 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]