基于去雾处理的无线视频传感系统设计

发布时间：2018-06-15 06:32

本文选题：去雾处理 + 无线视频传感系统　；参考：《南京邮电大学》2017年硕士论文

【摘要】：在缺乏电力和通信网络基础设施的野外环境下,传统的视频感知系统无法满足应用需求,无线视频传感器网络自组织、低功耗等特点能很好的解决上述问题。然而,野外场景通常笼罩在雾霾天气条件下,导致采集的视频图像模糊不清,难以从中获取场景的细节信息。为此,本文设计实现了一个基于去雾处理的无线视频传感系统。首先,针对系统野外环境视频采集的应用场景要求,本文结合视频图像去雾和无线传感网技术对系统进行了需求分析,设计了系统整体架构和节点体系结构,并详细介绍了系统各功能模块的软硬件实现。其次,针对Tarel算法存在的缺陷,提出一种改进的单幅图像去雾算法。用全变分模型滤波代替原算法的中值滤波,在大幅度提高计算效率的基础上保持了图像边缘特性;针对估计的大气耗散函数在明亮区域失效的情况,提出了一种基于雾气浓度特征的容错机制,提高了算法的普适性。实验对比结果表明,该算法能有效保持复原图像的边缘特性,恢复图像清晰自然,较目前的主流去雾算法,去雾效果和效率都具有一定的优势。然后,在改进的单幅图像去雾算法基础上,本文提出一种视频快速去雾算法。根据视频帧之间的关联性,将有雾视频序列按图像组为单位分别进行去雾,图像组的关键帧复原采用上文所提的单幅图像去雾算法,其余非关键帧基于运动矢量进行复原,这样可以大大减少去雾冗余,保证了视频去雾的实时性。实验结果表明,恢复的视频图像对比度和清晰度得到明显改善,且视频流能实时流畅输出。最后,结合本文所提的视频去雾算法,搭建和实现了原型系统。功能测试和性能测试的结果表明,本文设计的基于去雾处理的无线视频传感系统能满足野外环境的应用需求,并具有可移植性。
[Abstract]:In the absence of power and communication network infrastructure in the field environment, the traditional video sensing system can not meet the needs of the application, wireless video sensor network self-organization, low power consumption and other characteristics can solve the above problems. However, field scenes are usually enveloped in smog weather conditions, resulting in the blurred video images collected from which it is difficult to obtain the details of the scene. Therefore, a wireless video sensing system based on de-fogging is designed and implemented in this paper. First of all, according to the requirements of the application scene of video capture in the field environment of the system, this paper analyzes the requirements of the system by combining the video image de-fogging and wireless sensor network technology, and designs the overall architecture and node architecture of the system. The software and hardware implementation of each function module of the system is introduced in detail. Secondly, aiming at the defects of Tarel algorithm, an improved single image de-fogging algorithm is proposed. The median filter of the original algorithm is replaced by the total variational model filter, and the edge characteristic of the image is maintained on the basis of greatly improving the computational efficiency, and the estimated atmospheric dissipation function fails in the bright region. A fault tolerant mechanism based on fog concentration features is proposed to improve the universality of the algorithm. The experimental results show that the algorithm can effectively maintain the edge characteristics of the restored image and restore the image clear and natural. It has some advantages over the current mainstream de-fogging algorithm, the effect and efficiency of fog removal. Then, based on the improved single image de-fogging algorithm, a fast video de-fogging algorithm is proposed. According to the correlation between the video frames, the fogged video sequence is de-fogged according to the image group, the key frame of the image group is restored using the single image de-fogging algorithm mentioned above, and the rest of the non-key frames are restored based on the motion vector. In this way, the redundancy of de-fogging can be greatly reduced, and the real time of video de-fogging can be guaranteed. The experimental results show that the contrast and definition of the restored video images are improved obviously, and the video stream can be output smoothly in real time. Finally, the prototype system is built and implemented with the video de-fogging algorithm proposed in this paper. The results of function test and performance test show that the wireless video sensing system based on de-fogging in this paper can meet the requirements of field application and has portability.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41;TP212.9;TN919.8

【参考文献】