未找到相关数据
Polarization calculation and underwater target detection inspired by biological visual imaging
被引：6
作者：
Shen, Jie
[1
]
Wang, Huibin
[1
]
Chen, Zhe
[1
]
Wei, Yi
[2
]
Wu, Yurong
[1
]
机构：
[1] College of Computer and Information Engineering, Hohai University, Nanjing, China
[2] Institute of Communication Engineering, PLA University of Science and Technology, Nanjing, China
关键词：
In challenging underwater environments;
the polarization parameter maps calculated by the Stokes model are characterized by the high noise and error;
harassing the underwater target detection tasks. In order to solve this problem;
this paper proposes a novel bionic polarization calculation and underwater target detection method by modeling the visual system of mantis shrimps. This system includes many operators including a polarization-opposition calculation;
a factor optimization and a visual neural network model. A calibration learning method is proposed to search the optimal value of the factors in the linear subtraction model. Finally;
a six-channel visual neural network model is proposed to detect the underwater targets. Experimental results proved that the maps produced by the polarization-opposition parameter is more accurate and have lower noise than that produced by the Stokes parameter;
achieving better performance in underwater target detection tasks. © 2014 IFSA Publishing;
S;
L;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
引用
收藏
页码：33 / 41
相关论文