融合彩色模型空间的非线性低照度图像增强

低照度图像存在图像整体亮度偏低、亮度不均匀、色彩饱和度过高、图像模糊等问题,针对此类问题,提出了一种融合彩色模型空间的低照度图像增强算法。在该算法中,将图像的亮度增强与图像色彩恢复转换至不同的彩色模型空间分别进行处理:在RGB彩色模型空间中,首先对图像的高灰度级进行预处理,随后进行滤波处理,最后再用三分量增强函数对图像进行亮度恢复;在HSV彩色模型空间中,利用非线性色彩饱和度校正函数与亮度增强函数进行图像的色彩恢复,最后将两个空间中的处理结果进行加权融合。最终的对比实验结果表明,该方法在避免图像出现过度增强、色彩恢复与图像照度增强方面有着良好的效果,所处理的图像符合人眼视觉特性。

Non-linear Low-Light Image Enhancement Based on Fusion Color Model Space

Low light images have problems such as low overall brightness of the image, uneven brightness, high color saturation, and image blur. To address these problems, a low light image enhancement algorithm fusing color model space is proposed, in which the images are processed by transforming image brightness enhancement and image color restoration to different color model spaces respectively. In the RGB color model space, the high gray levels of the image are first preprocessed and then filtered, and finally the image brightness is restored with the three component enhancement function; In the HSV color model space, the image brightness is restored with the non linear color saturation correction function and the brightness enhancement function, and finally the processing results in the RGB and HSV color model spaces are weighted and fused. The final comparative experimental results show that the proposed method has a good effect in avoiding excessive image enhancement, color restoration and image light enhancement, and the processed images conform to the human visual characteristics.