基于FPGA的实时视频图像去雾系统

摘要

To meet the requirements of real-time video image processing,a kind of design based on FPGA for real-time video image dehazing system was proposed.Transmittance and global atmospheric light were estimated by making use of the mean fil-ter and recovered using the foggy image restoration model.The dehazing algorithm based on the atmospheric scattering model was researched.The algorithm is simple and effective and suitable for real-time video image dehazing.The algorithm was trans-planted in Terasic DE2-115 board,ADV7180 was taken as video decoder and VGA screen was taken for display,and Quartus II and Modelsim were used to carried out simulation verification.Verification results indicate that the algorithm has good effects on video image dehazing and real-time processing,the video image dehazing processing speed reaches 25 frame/s for 640∗480 images.%为满足现有视频有雾图像实时处理的需求，提出一种基于 FPGA平台实现视频图像去雾系统的方案。利用均值滤波估计透射率和全局大气光，通过雾天图像复原模型进行恢复；基于大气散射模型进行去雾算法的研究，该算法简单有效且适用于实时视频图像去雾；将该算法移植到友晶公司的 DE2-115视频图像开发板中，以 ADV7180作为视频解码器和VGA显示器作为显示终端，使用 Quartus II和 Modelsim进行仿真验证。验证结果表明，该算法在实时处理640∗480的视频图像时去雾效果较好，处理速度能够达到25帧/s的要求。