基于改进的LVQ神经网络图像边缘提取研究  

The Research of Image Edge Extraction based on Improved LVQ Neural Network

作　　者：郎俊 白国振[1] 周媛[1] LANG Jun;BAI Guozheng;ZHOU Yuan(School of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai 200093, Chin)

机构地区：[1]上海理工大学机械工程学院,上海200093

出　　处：《电子科技》2018年第6期70-74,共5页Electronic Science and Technology

基　　金：上海市自然科学基金(13ZR1458500)

摘　　要：针对传统的学习向量量化神经网络(GLVQ)具有初始值敏感、数据浪费以及权值训练不稳定等缺点,提出了一种基于广义学习向量量化神经网络的改进算法(MGLVQ)。MGLVQ神经网络是利用人工鱼群算法对权值离线粗调之后,再使用引入损失函数的梯度寻优算法对其进行在线细调。将该改进算法用于对图像每个像素的分类,实现图像的边缘提取。仿真结果表明,在Lena图像中GLVQ神经网络边缘点为整个像素的45.08%,而MGLVQ提取边缘点的比例为28.58%,即在整个像素分类中GLVQ比MGLVQ错误分类率高16.51%。相比于GLVQ神经网络,MGLVQ拥有更强的去除噪声的能力和更好的边缘检测效果。Traditional learning vector quantization neural network（ GLVQ） sensitive to initial values,data waste and weight training instability and other shortcomings,is proposed based on generalized learning vector quantization（ GLVQ） improved neural network（ MGLVQ）. MGLVQ neural network is the use of artificial fish algorithm for offline coarse adjustment,and then using the introduction of loss function gradient optimization algorithm for online precise adjustment. The improved algorithm for each pixel of image classification,image edge detection. Simulation results show that in the Lena image edge points GLVQ neural network for 45. 08% of all pixels,compared with28. 58% for MGLVQ extract edge points,namely the GLVQ throughout the pixels classification is 16. 51% more than MGLVQ misclassification. So in compared to GLVQ neural network,The algorithm has stronger ability to remove noise and better edge detection.

关 键 词：LVQ神经网络 边缘提取 人工鱼群算法 特征向量 MGLVQ 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]