J4

• 计算机 • 上一篇    下一篇

基于LVQ神经网络的植物种类识别

王路1, 张蕾2, 周彦军2, 曾晓云2, 孔俊2   

  1. 1. 吉林财税高等专科学校 信息系, 长春 130062; 2. 东北师范大学 计算机学院, 长春 130024
  • 收稿日期:2006-11-10 修回日期:1900-01-01 出版日期:2007-05-26 发布日期:2007-05-26
  • 通讯作者: 孔俊

Computeraided Plant Species IdentificationBased on LVQ Neural Network

WANG Lu1, ZHANG Lei2, ZHOU Yanjun2, ZENG Xiaoyun2, KONG Jun2   

  1. 1. Department of Information, Jilin College of Finance and Tax, Changchun 130062, China;2. School of Computer Science, Northeast Normal University, Changchun 130024, Cina
  • Received:2006-11-10 Revised:1900-01-01 Online:2007-05-26 Published:2007-05-26
  • Contact: KONG Jun

摘要: 提出一种基于学习矢量量化(LVQ)神经网络的计算机植物种类识别新方法. 使用2-D不变矩、 多尺度2-D Gabor滤波器等多种方法分别提取了叶片的几何特征和纹理特征, 应用LVQ神经网络识别植物种类. 实验结果表明, 该方法对植物种类的识别效率较高.

关键词: 计算机自动植物种类识别, 2-DGabor滤波, LVQ神经网络

Abstract: A new method of computeraided plant species identification based on learning vector quantization (LVQ) neural network is proposed. In this method 2-D moment invariants, multi-resolution Gabor filters and statistical moments are used to extract leaf information, such as geometry feature of leaf shape, texture feature of nervation, and LVQ neural network is used to distinguish the plant species. The experimental results illustrate the effectiveness of this method.

Key words: computeraided plant identification, 2-D Gabor filter, LVQ neural network

中图分类号: 

  • TP391