摘要
对于处于气泡等运动信息干扰情况下的水下视频图像,传统的增强手段不能达到令人满意的要求。本文提出了一种在水下视频图像中去除气泡场的方法。采用独立成分分析算法,将序列图像中的背景和运动目标进行分离处理,在灰度域上区分开气泡和其他运动目标并去除气泡,然后,在已经去除了气泡信号的图像中重新赋上其相邻序列的视频图像中相同位置但清晰的像素值,并使用光流法对实验结果进行了分析。结果证明,此方法能够成功的从水下图像中分离出气泡信号,达到了削弱、去除水下图像中气泡场的目的。
A method to remove air bubbles in underwater video image is put forward. In bubble movement information of underwater video image,the traditional means of enhancement cannot achieve a satisfactory requirement,thus the im-proved independent component analysis algorithm is proposed in this paper to deal with it. The background and moving targets in image sequences are separated by the algorithm,and then the bubbles are separated from other moving tar-gets and the air bubbles are removed in the gray domain. In the paper,the principle and the operation process of inde-pendent component analysis are introduced,and the simulation analysis was carried out on the experimental images.The results prove that this method can successfully from isolated bubbles in underwater image signal. the purpose of weaken-ing the bubble field in underwater image is achieved by this method. Finally,using the optical flow method in the ex-perimental results are analyzed, the results show that the method of removing image air bubble field water under the proposed is feasible.
出处
《长春理工大学学报（自然科学版）》
2014年第1期61-64,共4页
Journal of Changchun University of Science and Technology(Natural Science Edition)
关键词
独立成分分析
水下图像
气泡去除
光流法
independent component analysis
underwater image
removing bubbles
optical flow
