摘要
在雾天环境下,大气介质中悬浮颗粒的散射作用导致图像质量严重下降,限制了其使用范围,因此对雾天图像进行去雾处理是必要的。根据暗原色先验去雾的原理,在局部区域内基于图像分割的思想来较准确快速估计雾天图像的传播图,然后应用大气散射模型对雾天图像进行去雾处理,并通过直方图拉伸来增大处理后的图像的对比度。实验结果表明,该算法能有效去除雾气对图像的影响,与传统去雾算法相比较,具有较快的处理速度和较强的实用性。
It is very necessary to defog the images captured in the foggy days, because the scattering effect of atmospheric suspending particles will lead to serious image degradation, which directly influences the visual effect and its application range. Based on the principle of dark channel priority, a segmentation model is introduced to estimate a image transmission map precisely and fast in local area, and then the atmosphere scattering model is applied to defog the images, and the image contrast after defogging is enhanced by histogram stretching. Experiment results show that the proposed algorithm efficiently removes the influence of the fog to the image with high processing speed and wide application range.
出处
《太赫兹科学与电子信息学报》
2013年第2期254-259,共6页
Journal of Terahertz Science and Electronic Information Technology
基金
国家自然科学基金资助项目(61071161)
关键词
图像去雾
图像分割
大气散射模型
暗原色先验
直方图均衡
image defogging
image segmentation
atmosphere scattering model
dark channelpriority
histogram equalization
