| 基于叶片外形特征的植物识别研究 |
| |
| 引用本文： | 侯铜,姚立红,阚江明.基于叶片外形特征的植物识别研究[J].湖南农业科学,2009(4). |
| |
| 作者姓名： | 侯铜 姚立红 阚江明 |
| |
| 作者单位： | 北京林业大学工学院,北京,100083 |
| |
| 摘 要： | 提出了一种基于植物叶片形状特征的方法对叶片进行识别,首先对叶片进行预处理,其中包括滤波和阈值分割,然后提取叶片的相对特征参数,得到了8个特征参数。使用BP神经网络对现有的几种叶片进行分类,选择隐含层数、调节学习效率、动量因子等参数。在VC 6.0环境下循环计算、调整权值、阈值,最终识别成功,并达到了87.5%的平均识别率。
|
| 关 键 词： | 预处理 特征提取 调节参数 神经网络分类 |
Plant Recognition Research Based on Shape Features of Leaf |
| |
| HOU Tong,YAO Li-hong,KAN Jiang-ming.Plant Recognition Research Based on Shape Features of Leaf[J].Hunan Agricultural Sciences,2009(4). |
| |
| Authors: | HOU Tong YAO Li-hong KAN Jiang-ming |
| |
| Institution: | College of Engineering;Beijing Forestry University;Beijing 100083;PRC |
| |
| Abstract: | The author introduced a method of recognizing plant leaves based on shape features.Firstly,image pre-processing was applied to get eight relative parameters including filtering and threshold partition.Secondly,BP neural network was applied to classify existing leaves,choose hidden layers,adjust learning efficiency,and momentum factor.The author calculated those parameters circularly,adjust weights,choose thresholds,and succeed in recognizing leaves in VC++.The average correct recognition rate was up to 87.5... |
| |
| Keywords: | pre-processing feature extraction adjust parameter neural network classification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
