集成电路与嵌入式系统2024，Vol.24Issue(12)：25-32,8.DOI:10.20193/j.ices2097-4191.2024.0043
基于FPGA的图像去雾算法优化设计
Optimized design of image defogging algorithm based on FPGA
摘要
Abstract
To address the shortcomings of traditional defogging algorithm and meeting the real-time requirements of the defogging sys-tem,an improved defogging algorithm is proposed and designed with FPGA.Firstly,a Gaussian low-pass filter is employed to retain the low-frequency information of the image.Secondly,the rough transmittance map of the foggy image is obtained by combining the a priori theory of the dark channel with the bilateral filtering to acquire the refined transmittance map.Lastly,the dehazed image is obtained by the foggy imaging model.Additionally,the improved algorithm is optimized with FPGA.The experimental results show that the im-proved defogging algorithm enhances the defogging effect on images with high-brightness areas compared to other defogging algorithms.When performing video image defogging on an FPGA platform,it can achieve a speed of 60 frames per second,meeting the real-time requirement.关键词
图像去雾/FPGA/暗通道先验/实时性Key words
image defogging/FPGA/dark channel priori/real-time分类
信息技术与安全科学引用本文复制引用
严飞,李楚,郑绪文,孟川,刘银萍..基于FPGA的图像去雾算法优化设计[J].集成电路与嵌入式系统,2024,24(12):25-32,8.基金项目
江苏省重点研发计划项目(BE2020006-2) （BE2020006-2）
国家自然科学基金项目(61605083). （61605083）
