自适应水下彩色图像增强算法

An adaptive underwater image enhancement algorithm

在水下环境中,水体介质对光照的吸收和散射是导致水下成像失真的两大主要成因。水体对光线的吸收导致能量衰退大大降低了清晰度,另水体介质对各波段的散射和吸收的无规律性导致水下成像颜色失真。为了提高对水下成像的认知,本文提出了一种自适应水下图像增强算法。该方法首先对原始图像进行降噪和对比度提升得到两幅输入图;其次,以原始图像的显著图为引导,结合照度和色度图,作为图像融合的权重图;最后,根据权重图对两幅输入图像进行自适应加权求和,从而实现图像增强的目的。实验结果证明本文算法具有良好的实时性和鲁棒性,有效地增强了水下降质图像。

In underwater situation,light scattering and absorption are two major sources of distortion for underwater photography. The clarity is greatly reduced due to the absorption. And the irregular light scattering and absorption leads to the color distortion. In order to improve the perception of underwater image,this paper introduces an adaptive underwater image enhancement algorithm. The method first applies the denoising and global contrast enhancement technologies to the original image respectively,then taking these two adapted versions of the original image as inputs that are weighted by specific maps. Then,guiding by the saliency map,the method also employs the luminance and the chrominance information to be the weight maps. At last,the method adaptively computes the weight sum of the two inputs in a per-pixel fashion. The experimental results demonstrate that the proposed method can obtain good visual quality and has good real-time performance and robustness.