基于OSTU的光照不均匀图像自适应增强算法

Adaptive Enhancement Algorithm for Non-Uniform Illumination Images Based on OSTU

针对现有算法在增强光照不均匀图像暗区域时存在亮区域亮度失真的问题,提出了基于OSTU的光照不均匀图像自适应增强算法。首先,将图像的色彩空间由RGB转换至HSV,使用多尺度引导滤波算法计算出V分量的照度分量与反射分量,然后,通过OSTU法计算V分量的最佳分割阈值,根据该阈值设计出自适应调整照度分量不同区域的两条伽马曲线,同时,将调整后的照度分量与原照度分量融合并对反射分量进行增强,最后,合并照度分量与反射分量,得到处理后的V分量。为验证算法的有效性,将该算法与MSR算法、CLAHE算法及相关文献所提算法进行对比,实验结果表明,所提算法在提高图像暗区域亮度的同时很好的避免了图像亮区域亮度失真现象,图像增强后具有较高的SSIM值和PSNR值。

In order to solve the problem of brightness distortion existing in the existing algorithms when enhancing the dark area of non-uniform illumination image, an adaptive enhancement algorithm of non-uniform illumination images based on OSTU is proposed. Firstly, the color space of the image was converted from RGB to HSV,the illuminance and reflection components of the V component were estimated by multi-scale guided filter algorithm. Then, two gamma curves were designed to adjust the different area of illuminance component adaptively based on the V component threshold, which was calculated by OSTU. Additionally, by integrating the adjusted illuminance components with the original illuminance component, and the reflection component was enhanced. Finally, combining the illuminance component and reflection component, the processed V component was obtained. In order to verify the effectiveness of the algorithm, the algorithm was compared with MSR algorithm, CLAHE algorithm and related literature. The experimental results show that the algorithm proposed can improve the brightness of the dark area of the image and preform better in avoiding the brightness distortion of the bright area, and the enhanced images have higher SSIM and PSNR values.