Real-time enhancement of the image clarity for traffic video monitoring systems in haze
Real-time enhancement of the image clarity for traffic video monitoring systems in haze
复制标题
雾霾天气下交通视频监控系统图像清晰度的实时增强
DOI：
10.1109/cisp.2014.7003741
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
Meijiao Wang
中科院分区：
文献类型：
--
作者：
X. Ji;Jiezhang Cheng;J. Bai;Tingting Zhang;Meijiao Wang
在雾霾天、雾天等天气条件下获取的室外场景图像对比度差、色彩逼真度低。为了有效改善雾霾天气下图像质量下降的问题，降低雾霾天气对室外交通视频监控系统的影响，本文分析了雾霾天气下图像质量下降的原因和模糊机理。从图像复原和图像增强的角度出发，提出了一种基于全局暗通道先验理论和图像对比度扩展的实时图像去雾方法。该算法首先采用全局暗通道先验方法去除图像中的雾霾，然后采用直方图均衡化方法增强图像的对比度和亮度。实验结果表明，该方法直接恢复出清晰、质量好的无雾图像，取得了满意的视觉效果。针对实际工程应用，设计了一种高性能的图像采集、增强、去雾及传输平台。该平台以FPGA（现场可编程门阵列）为核心处理器，实现了本文提出的算法。时序仿真结果和实际测试结果验证了该方法的可靠性和有效性。最后，通过实际测试结果表明，该系统能够实时、有效地增强交通视频监控系统的图像对比度和色彩清晰度，从而提高其可靠性、稳定性和科普雾、霾等恶劣天气的能力。
The images of outdoor scenes obtained in haze, fog and other weather days are usually have poor contrast and color fidelity. In this paper, in order to effectively improve the degraded image in haze quality, reduce the effect of the haze to outdoor traffic video monitoring systems, we analyzed the image degradation reason and fuzzy mechanism of image in haze. From the viewpoints of image restoration and image enhancement, an efficient and real-time image haze removal approach in view of the global dark-channel prior theory and image contract extending was proposed. Firstly, we used the global dark-channel prior method to remove the haze and fog, and then adopted the histogram equalization to enhance the contract and the brightness of images. The experimental results showed that the approach directly recovered a clear and quality haze-free image, obtained satisfactory visual effect. For the practical engineering applications, a high performance image acquisition, enhancement, haze removal and transmission platform was designed. It used the FPGA (field programmable gate array) as the core processer, and the algorithm proposed in this paper was implemented on the platform. The simulation result of timing sequence and the actual testing result were verified the reliability and validity of the method. 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 systems, thus it can improve its reliability, stability and the ability to cope with the bad weather such as the fog, haze and so on.