对行车视程估值的车前图像去雾算法

针对雾天条件下车前图像环境复杂,去雾困难的问题,提出了一种车前图像去雾算法。分析并简化了大气物理模型;并在分析了影响透射率衰减因子的因素的基础上,借助大气消光系数提出了透射率的一种表现形式。接着借助车行可视距离与能见度的关系,最终建立了能见度和透射率的数学关系。最后利用引导滤波求取大气光幕的值,并通过修复函数完成了单幅车前图像的去雾处理。去雾实验结果显示,经过算法处理后的图像亮度得到了提高,图像变得更加清晰,图像中的细节变得明显,图像的对比度也变得更强。

Image defogging algorithm based on the valuation of the driving range

In order to solve the problem of Image defogging in front of the car, a new image processing algorithm was proposed. Firstly, the atmospheric physical model was analyzed and simplified; then, With the help of the atmospheric extinction coefficient a form of transmission is proposed; next, with the help of the relationship between driving visibility and visibility, the mathematical relationship between visibility and transmittance was established; finally, guided filtering was used to estimate the value of atmospheric light and with the help of the repair function, the fog image is restored. After the algorithm the image has higher brightness and clarity as well as better image contrast.