Aiming at the problem of image distortion caused by light refraction and the limitations of existing solutions, we propose a three-dimensional information extraction algorithm based on spatial mapping. Main procedures of the algorithm are as follows. First, we obtain the pixel coordinate information of the underwater image. Then, we convert the target underwater image into the equivalent air image through the pixel-pixel spatial mapping relationship. At last, the three-dimensional information is extracted from the equivalent air image. The algorithm can be applied to extract three-dimensional information of multiple underwater targets. Experiment results show that compared with the actual three-dimensional information of the target, the average error rate of the distance extracted by the proposed algorithm is 1.475%, and the average error of the attitude angle extracted by the proposed algorithm is 2.559°, which means the three-dimensional information of the target can be accurately extracted by the proposed algorithm.
The alert did not successfully save. Please try again later.
Ruiliang Zong, Xuyang Chen, Yuyao He, Baoqi Li, Yaohua Zhao, "3D information extraction algorithm of underwater target based on spatial mapping," Proc. SPIE 10850, Ocean Optics and Information Technology, 108500D (12 December 2018);