Volume 37, Number 3, June 2019
|Page(s)||471 - 478|
|Published online||20 September 2019|
Study on Underwater Image Conversion Algorithm Based on Light Refraction Model
School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an, 710072, China
2 School of Electronic and Information, Northwestern Polytechnical University, Xi’an, 710072, China
Aiming at the problem of image distortion caused by light refraction during underwater imaging, and the conversion error caused by existing image conversion algorithms due to neglecting the secondary refraction of light, an underwater image conversion algorithm based on the light refraction model is proposed in this paper. The algorithm firstly obtains the pixel information of the underwater image, then calculates the corresponding coordinate information of the pixel points in the equivalent air image through the mapping relationship, and finally obtains the equivalent air image through image interpolation. Experimental results shows, compared with the existing image conversion algorithms, the proposed algorithm reduces the average error of u direction from 2.289 5 to 1.213 3, which is a decrease of 47.01%. The average error of v direction is reduced from 3.252 5 to 1.526 3, which is a decrease of 53.07%. At the same time, the mean value of ranging error was reduced from 58.83 mm to 28.88 mm, a decrease of 50.91%。
针对水下目标成像时光线折射所造成的图像失真问题，以及现有图像转换算法因忽略光线二次折射所造成的转换误差，提出一种基于光线折射模型的水下图像转换转算法。该算法首先获取水下图像的像素点信息，通过映射关系计算得到像素点在等效空气图像中的对应坐标信息，从而获得水下目标的等效空气图像。实验结果显示，文中所提算法较之现有图像转换算法，u方向图像转换误差均值由2.289 5降为1.213 3，降低47.01%，v方向图像转换误差均值由3.252 5降为1.526 3，降低53.07%。同时，测距误差均值由58.83 mm降为28.88 mm，降低50.91%。
Key words: binocular stereo vision / underwater image processing / light refraction model / ocean optics / average error / ranging error / image conversion algorithms
关键字 : 双目立体视觉 / 图像处理 / 折射模型 / 海洋光学
© 2019 Journal of Northwestern Polytechnical University. All rights reserved.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.