Atmospheric scattering-based multiple images fog removal

摘要：

Current video and image systems are typically of limited use for their poor visibility and contrast caused by the presence of fog, haze and mist. However, the local hazy areas of these images are often interested in and required their sharpness. In this paper, a novel fast haze removal algorithm from multiple images in uniform bad weather conditions which bases on the atmospheric scattering model is proposed. The key idea is to establish an overdetermined system by modeling the hazy images and corresponding images taken in clear days so that the transmission and global airtight can be obtained. Then the two parameters solved from the equations are applied to the local hazy area. Experimental results demonstrate that the proposed algorithms remove haze effectively and achieve accurate restoration.

关键词： atmospheric scattering model transmission scene depth airlight

作者: Zhang Tao Shao Changyan Wang Xinnian

作者单位: School of Information Science and Technology Dalian Maritime University Dalian, China, PRC School of School of Information Science and Technology Dalian Maritime University Dalian, China, PRC

会议类型: 国际会议

会议名称: 2011 4th International Congress on Image and Signal Processing(第四届图像与信号处理国际学术会议 CISP 2011)

会议地点: 上海

会议语种:英文

页码: 108-112

在线出版日期: 2011-10-15（万方平台首次上网日期，不代表论文的发表时间）