摘要
恶劣天气条件下图像退化的复原方法,对提高户外视觉系统的稳定性和实用性极为重要。雨作为其中一种场景,影响了户外图像或视频进行可靠目标检测、物体识别和跟踪、特征提取等基本操作,极易造成处理偏差,因此有必要对去雨方法进行研究,复原该类图像,消除降雨对图像质量的影响,以保证图像的清晰度。基于视频中雨滴的时域特性和像素直方图的全局特性,借助FCM(模糊C均值)聚类算法,提出了检测与去除视频中雨滴的方法。通过视频中4种不同场景的实验,验证了该方法的有效性。同时,该算法也可应用于其它类似雨的场景检测中。
Due to the random spatial distribution and fast motion,the visual effects of raindrops have a complex impact on various signals of videos and degrade the perceptual image quality of videos.Thus it is essential to develop algorithms for detecting and removing rain from videos.This paper gives a comprehensive analysis of FCM based on raindrops＇ spatio-temporal and histogram image.Experiments on videos of the four complex sceces prove the effectiveness of the proposed algorithms.Also,the algorithms described in this paper can be used in a wide range of applications,including the rain-like scence detection.
出处
《通信技术》
2012年第3期75-78,共4页
Communications Technology
基金
国家自然科学基金项目(批准号:11176018)
