结合边界约束和引导滤波的航空图像去雾算法

Aerial image haze removal algorithm based on boundary constraint and guided filter

针对尘雾天气条件下,航空图像受大气粒子的散射作用而导致对比度和色彩保真度差等问题,提出一种结合边界约束和引导滤波的航空图像去雾方法。该方法基于暗通道优先原理估算出大气光强度,通过添加边界约束条件来解目标函数,进而求出透射率,再用引导滤波对透射率进行精细化调整,达到去雾目的。与Fattal、Tarel、He和Meng算法对比分析表明,本文方法运算速度快,能够提高图像的细节清晰度,增大去雾深度,有效克服颜色过饱和现象,去雾图像自然逼真。

In fog weather conditions,contrast and color fidelity of aerial image are poor due to the scattering of the atmosphere particles. A fast defogging method combined with boundary constraint and guided filter in the aerial image is proposed. The method based on the principle of dark channel is used to estimate the atmospheric light intensity. By adding the boundary constraint condition and solving the objective function to calculate the transmission,and then refining the transmission by the guided filter to get the purpose of defogging,the results show that compared with the methods of Fattal,Tarel,He and Meng algorithm,the speed of the proposed method is fast. It can improve the detail clarity and increase the depth of defogging. What is more,it can overcome the color saturation phenomenon effectively and the defogging images are natural and realistic.