摘要
针对现有去雾算法实时性和去雾效果差的问题,本文改进了基于暗通道的去雾算法,使其更加适用于航拍图像。首先对图像进行四分法,求取每一个图像块内像素最大的值,求平均值,即可得到大气光值。其次,用均值滤波求取透射率,并用导向滤波进行优化。最后复原出去雾后的图像。通过实验证明,本文算法简单实用,实时性好,具有一定的应用价值。
Aiming at the problem of real-time and poor de-fog effect of the existing de-fog algorithm,this paper improves the de-fog algorithm based on dark channel to make it more suitable for aerial image.First,the image is quadrilateral method,the maximum value of the pixels in each image block is obtained,and the average value is obtained,and the atmospheric light value can be obtained.Secondly,the transmittance is obtained by mean filtering and optimized by guided filtering.Finally restored the image after the fog.Experiments show that the algorithm is simple and practical,real-time is better,has a certain application value.
出处
《信息技术与信息化》
2017年第10期61-63,共3页
Information Technology and Informatization
关键词
图像去雾
大气物理模型
暗通道
均值滤波
Haze removal
The physical model of atmosphere
Dark channel
Mean filtering
