一种多尺度分解与暗原色先验结合的去雾算法  

Defogging algorithm combining multi-scale decomposition and dark channel prior

作　　者：姜祖运 王小春[1] 李红军[1] JIANG Zuyun;WANG Xiaochun;LI Hongjun(College of Science,Beijing Forestry University,Beijing 100083,China)

机构地区：[1]北京林业大学理学院,北京100083

出　　处：《现代电子技术》2021年第7期44-49,共6页Modern Electronics Technique

基　　金：国家自然科学基金(61571046)。

摘　　要：为了提高图像去雾效果,提出一种在图像多尺度分解的基础上,通过改进的暗原色方法实现图像去雾的算法。该算法在HSV颜色空间中,首先对亮度分量V进行多尺度分解,然后对高频部分进行软阈值消噪,并通过梯度运算增强边缘,对低频部分采用改进的暗原色先验模型进行处理;其次,对饱和度分量S先进行非局部均值滤波去噪,再通过线性拉伸使其复原;最后,将图像由HSV空间转换到RGB空间得到最终的去雾图像。实验结果表明,在去雾效果和时间复杂度方面,该算法比一些经典算法都有明显的改善。An image defogging algorithm based on the image multi-scale decomposition and the improved dark channel prior is proposed to enhance the defogging effect.In the algorithm,a multi-scale decomposition is performed on the brightness component V in the HSV color space.And then,its high frequency part is processed with soft threshold denoising and edge enhancement;its low frequency part is processed with the improved dark channel prior model.The saturation component S is restored by non-local mean filtering denoising and contrast stretching.Finally,the image is transformed from HSV space to RGB space to get the final defogged image.The experimental results show that the algorithm is better than some classical algorithms in terms of both the defogging effect and time complexity.

关 键 词：图像去雾 多尺度分解 暗原色先验 W-BEMD分解 软阈值消噪 形态学滤波 边缘增强 

分 类 号：TN911.73-34[电子电信—通信与信息系统] TP391.41[电子电信—信息与通信工程]