低信噪比环境下光斑图像信号的定位优化研究  

Research on Location Optimization of Spot Image Signal in Low SNR Environment

作　　者：李伟[1] 朱敏[1] 左常玲[1] LI Wei;ZHU Min;ZUO Changling(School of Electronic and Electrical Engineering,Anhui Sanlian University,Hefei 230601,China)

机构地区：[1]安徽三联学院电子电气工程学院,合肥230601

出　　处：《重庆科技学院学报（自然科学版）》2022年第5期70-74,共5页Journal of Chongqing University of Science and Technology：Natural Sciences Edition

基　　金：安徽省教育厅科学研究项目“基于声纹识别的智能家居声控信号语音识别算法的研究”(KJ2021A1190);“基于深度数据分析的老人异常行为检测算法研究”(KJ2020A0810)。

摘　　要：针对低信噪比环境下激光通信光斑图像中心定位不准的问题,提出了一种基于BP神经网络模型的定位算法。首先,对激光光斑图像进行预处理,并基于维纳滤波器去除环境噪声干扰;然后,利用最大类间方差法进行灰度阈值分割与光斑图像特征提取;最后,将特征元素输入神经网络模型中,通过训练降低输入量和输出量之间的误差,调整中心定位偏导数与隐含层神经元之间的权值比例,以降低算法的计算成本。仿真结果显示,本算法在低信噪比环境下仍可获取到完整的光斑图像,且光斑中心的两个轴向偏差较小,类间方差值低。Aiming at the problem of inaccurate positioning of the image center of laser communication spot under the condition of low signal-to-noise ratio,a positioning algorithm based on BP neural network model is proposed.Firstly,the laser spot image is preprocessed,and the ambient noise interference is removed based on Wiener filter to improve the signal-to-noise ratio.Then,the Otsu algorithm is used for the gray threshold of the image to extract the features of the target image.Finally,these feature elements are input into the neural network model as input items.Through training,the error between input and output is reduced,and the weight proportion relationship between the partial derivative of central positioning and the neurons in the hidden layer is adjusted,to reduce the computational cost of the algorithm.The simulation results show that the proposed algorithm can still obtain the complete spot image under the condition of low signal-to-noise ratio,and the two axial deviations of the spot center are small,and the mean value of inter class variance is lower.

关 键 词：低信噪比 光斑图像 定位 BP神经网络 类间方差 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]