大学毕业论文-关于图像三维降噪算法的模拟仿真论文正文.docVIP

摘 要 视频具有直观、可靠、方便、信息内容丰富等特点，因此广泛应用于金融、商业、交通、住宅、社区等场合，为这些领域的环境监控和安全防范起到了不可忽略的作用。然而，噪声的存在是实时视频监控中不可避免的问题，也是影响视频视觉质量的重要因素。因此，降噪技术是增强视频视觉质量的一个非常重要的手段。 本文从上述需求要求出发：1.研究比较传统二维降噪和三维降噪的区别，明确三维降噪优势所在；2.对二维去噪算法进行了详细地研究，通过对传统去噪算法、非局部均值算法和其改进后的自适应非局部均值算法进行比较，确立了最优的二维图像降噪算法。3.重点研究三维视频降噪算法，结合多帧图像采用基于 Non_local means 的 运动检测方法自适应地区分图像的运动区域和静止区域，对静止区域采用时域加权均值滤波，对运动区域采用空域 ANL 滤波。 实验结果表明，基于Non-local means的时空联合视频降噪算法由于准确地区分了图像的运动区域和静止区域，充分利用了视频的时域、空域信息，在不造成运动拖影的前提下，能够显著提高视频的峰值信噪比和图像的主观质量，同时基本满足实时性要求。 关键词 三维视频降噪，ANL算法，运动检测，时空联合 ABSTRACT Video, because of its advantage of intuition, conwenience, reliability, rich information, and so widely in areas such as finance, commerce, traffic, house and community, and plays an important role in monitoring and protection of the areas.However, the existence of noise makes it an inevitable problem in real-time video monitoring, and an important factor to evaluate of quality of video vision. Therefore, noise reduction technology is a very important means of enhancing the visual quality of the video. In this paper，three efforts are involved：1.compared the difference between traditional two-dimensional noise reduction and three-dimensional noise reduction, clear three-dimensional noise reduction advantages; 2.Study two-dimensional noise reduction algorithm in detail, through comparing the traditional two-dimensional noise reduction algorithm, non-local means algorithm and adaptive non-local means algorithm , establish the optimal two-dimensional image noise reduction algorithm; 3.Focus on the three-dimensional video noise reduction algorithm, by motion detection to multi-frame images based on non-local means, this algorithm can adaptively distinguish the still regions and motion regions of video image. Temporal weighted average filter to the still regions and spatial ANL filter to the motion regions are used separately. Experimental results show that because of the the precise distinction of the still regions and motion regions,ful