基于暗原色先验模型的水下图像增强算法-《计算机技术与发展》

文章信息/Info

Title:: Underwater Image Enhancement Algorithm Based on Dark Channel Prior Model

作者:: 李社蕾¹; 李海涛²; 崔聪颖¹; 1. 三亚学院信息与智能工程学院,海南三亚 572022; 2. 92961 部队,海南三亚 572021

Author(s):: LI She-lei¹ ; LI Hai-tao²; CUI Cong-ying¹; 1. School of Information &Intelligence Engineering,Sanya University,Sanya 572022,China; 2. The 92961 Unit of PLA,Sanya 572021,China

Keywords:: underwater image; color correction; image enhancement; dark channel prior; image processing

摘要:: 针对在水下环境中,由于光的散射和衰减导致的图像对比度低、颜色失真等问题,提出了一种暗原色先验模型和基于统计方法的颜色校正算法相结合的图像增强新方法。该方法根据水下彩色图像成像模型,首先利用暗原色先验算法对水下彩色图像进行去模糊处理,并针对大水深情况下获取水下图像时因使用人工照明而造成的水下彩色图像亮度不均匀问题,对暗原色先验模型中背景光强度 A 值的估计方法进行改进;然后利用统计方法对水下彩色图像的 R、G 和 B 三个颜色通道分别进行颜色校正,从而实现了水下图像色彩的整体校正。实验结果表明,该方法有效增强了水下彩色图像的对比度和亮度,消除了由于光的散射而造成的图像模糊,使水下彩色图像具有更好的可视性。

Abstract:: Aiming at the problem of low image contrast and color distortion in the underwater environment caused by scattering and attenuation of light,we propose a novel method of image enhancement based on the combination of a dark channel priori model and a color correction algorithm based on statistical method. According to the imaging model of underwater color image,firstly the dark channel prior algorithm is used to de-blur the image,and for the uneven brightness of underwater color image caused by artificial lighting in deep water,the estimation method of the background light intensity A in the dark channel priori model is improved. Then the color correction of the three color channels,namely R,G and B,are respectively carried out by statistical method,thus realizing the overall correction of underwater image color. The experiment shows that the proposed method can effectively enhance the contrast and brightness of underwater color images and eliminate the image blurring caused by light scattering,which makes the underwater color images have better visibility.