摘要
针对传统基于视觉显著性和卷积神经网络定位技术、基于深度学习的输送带标记定位技术受到噪声信号影响,而导致定位精准度低的问题,提出机器视觉视角下船舶无线通信系统定位技术。根据机器视觉视角定位原理,计算实际延时,获取总节点数量。分析在真实环境下存在的干扰情况,构建信号损耗模型,计算从节点处接收端采集信号强度。分析图像噪声特性,处理带标记输送带图像,使用卡尔曼滤波去噪方式剔除噪声数据,完成无线通信系统定位。由实验结果可知,该系统定位精准度最高为98%,能够对船舶无线通信系统精准定位。
Aiming at the problem that the traditional positioning technology based on visual saliency and convolution neural network, and the conveyor belt marking positioning technology based on deep learning are affected by noise signal,resulting in low positioning accuracy, this paper proposes the positioning technology of ship wireless communication system from the perspective of machine vision. According to the principle of machine vision, the actual delay is calculated and the total number of nodes is obtained. This paper analyzes the interference in the real environment, constructs the signal loss model, and calculates the signal strength collected from the receiving end of the node. The noise characteristics of the image are analyzed, and the image of the marked conveyor belt is processed, and the noise data is eliminated by Kalman filter to complete the positioning of the wireless communication system. The experimental results show that the positioning accuracy of the system is up to 98%, which can accurately locate the ship wireless communication system.
作者
赵菲
ZHAO Fei(Zhengzhou University of Industrial Technology,Zhengzhou 451150,China)
出处
《舰船科学技术》
北大核心
2021年第2期136-138,共3页
Ship Science and Technology
关键词
机器视觉视角
船舶无线通信系统
定位技术
去噪方式
machine vision perspective
ship wireless communication system
positioning technology
denoising method
