雾霾天交通视频监控系统图像实时清晰化处理

针对交通视频监控系统在雾霾天气下图像像质退化的问题,从天气对系统成像影响入手分析造成系统图像降质和模糊的机理。基于全局暗原色先验和图像对比度增强方法,提出了一种有效、快速的雾霾天气下降质图像清晰化处理方法。结合实际应用,以现场可编程门阵列FPGA为核心处理单元,设计了一种高速图像采集、增强、去雾霾和传输系统平台,并基于该平台实际验证了文中提出的算法。最后,经过实际测试表明,该系统平台能实时、有效地提高交通视频监控系统图像的对比度和色彩清晰度,从而提高了其工作的可靠性、稳定性以及应对恶劣天气的能力。

Real-time Improving the Image Clarity for Traffic Video Monitoring System in Haze

According to the image of the traffic video monitoring system degradation in haze weather, the degradation reason and fuzzy mechanism were analyzed from the effect of weather to the imaging system. An efficient and fast improving the degraded image clarity method was proposed which based on the global dark-channel prior and image contract extending. The experimental results showed that the method can directly recover a high quality haze-free image and obtained satisfactory visual effect. For the practical engineering applications, a high speed and performance image acquisition, enhancement, haze removal and transmission platform was designed which used the field programmable gate array (FPGA) as the core processer, and the algorithm proposed in this paper was implemented on the platform. Finally, through the actual test results indicated that the system can real-time, effectively enhance the image contrast and color definition of traffic video monitoring system, thus improve its reliability, stability and the ability to cope with the bad weather.