摘要
针对在雾霾、沙尘等不利的气候条件下,拍摄照片和视频会受到如对比度下降、色彩失真等严重影响。为了减少不利天气对视觉信息的干扰,该文将对暗通道先验算法进行改进,提出一种优化的加窗加权暗通道先验自适应阈值去雾算法,结合硬件FPGA(field programmable gate array)对图像视频进行去雾,并运用模块化设计的理念,对算法进行硬件级优化,利用FPGA的并行计算特性加速图像视频去雾。实验结果表明,该系统能够高效地还原清晰图片,硬件资源消耗及成本较低,视频帧率稳定在60帧/s,可满足实时性要求。
In adverse weather conditions such as haze and sandstorms,taking photos and videos can be severely affected by factors such as decreased contrast and color distortion.In order to reduce the interference of adverse weather on visual information,this paper will improve the dark channel prior algorithm.Propose an optimized windowed weighted dark channel prior adaptive threshold dehazing algorithm,combined with hardware FPGA(field programmable gate array) for dehazing images and videos,and apply modular design concept to optimize the algorithm at the hardware level,utilizing FPGA parallel computing characteristics to accelerate image and video dehazing.The experimental results show that the system can efficiently restore clear images,with low hardware resource consumption and cost.The video frame rate is stable at 60 frames per second,meeting real-time requirements.
作者
曹青正
李茜铭
汤小红
陈凯歌
CAO Qingzheng;LI Ximing;TANG Xiaohong;CHEN Kaige(School of Mechanical Engineering,Hunan Institute of Science and Technology,Yueyang 414006,China)
出处
《自动化与仪表》
2025年第2期144-147,161,共5页
Automation & Instrumentation
基金
湖南省研究生科研创新项目(CX20240967)
湖南省自然科学基金项目(2024JJ7204)。
