基于双域分解的夜间车辆红外图像研究

夜间车辆交通红外图像光照不均,导致车辆图像细节纹理较弱,识别难度较大。为此,提出基于双域分解的夜间车辆交通红外偏振图像增强方法。采用改进Retinex低照度图像光照补偿算法,分解图像为低频图像与高频图像,对低频图像去雾、优化其对比度,对高频图像去噪与增强,合成低频、高频图像实现夜间车辆交通红外偏振图像增强。实验测试结果证明,对比传统方法,所提方法增强后图像亮度与对比度得以优化,且细节信息更丰富,具有理想的视觉效果。

Research on infrared image of vehicle at night based on dual domain decomposition

The uneven illumination of infrared images of vehicle traffic at night leads to the weak detail texture of vehicle images,which is difficult to be identified. Therefore,an infrared polarization image enhancement method based on dual domain decomposition is proposed. The improved Retinex low illumination image illumination compensation algorithm is used to decompose the image into low frequency image and high frequency image. The low frequency image is defogged,and its contrast is optimized. Meanwhile,the high frequency image is denoised and enhanced. Thus,the low frequency and high frequency images are synthesized to realize the infrared polarization image enhancement of vehicle traffic at night. The experimental results show that this method optimizes the brightness and contrast of the enhanced image,providing more detailed information compared to the traditional method,and has an ideal visual effect.