特征融合在植物叶片识别中的应用研究  

Application Research on Multi-feature Fusion in Plant Leaf Recognition

作　　者：陆楷煜 夏春蕾[1] 戴曙光[1] 靖浩翔 马玉文 彭铄期 盛旭阳 LU Kai-yu;XIA Chun-lei;DAI Shu-guang;JING Hao-xiang;MA Yu-wen;PENG Shuo-qi;SHENG Xu-yang(School of Optical and Computing Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093

出　　处：《软件导刊》2020年第10期71-75,共5页Software Guide

摘　　要：利用模式识别领域的特征融合方法进行植物叶片识别,植物叶片对植物种类分辨与认知具有重大作用,其纹理、形状是分辨植物种类的一个极佳指标。以植物叶片为研究对象,提取叶片的LBP纹理特征、HOG纹理特征以及形状特征,设计多特征融合模型,基于ECOC-SVM多分类器对UCI数据集32种640张植物叶片图像进行训练、分类。实验结果表明,基于多特征融合模型训练的ECOC-SVM对数据集中的叶片有很好的描述能力,识别率达92%,识别效果较好。In recent years,as an important field of artificial intelligence,pattern recognition has developed rapidly.Plant leaves play an important role in identifying and recognizing plant species.At the same time,the texture and shape of plant leaves are very good indicators to distinguish plant species.This paper takes plant leaves as the research object,extracts LBP texture feature,HOG texture feature and shape feature of leaves,designs multi-feature fusion model,and uses ECOC-SVM multi-classifier to 640 plants of 32 species in UCI dataset Leaf images for training and classification.The experimental results show that the ECOC-SVM based on the multifeature fusion model has a good description ability for the leaves in the data set with a higher recognition rate of 92%.The model has a better recognition effect.

关 键 词：模式识别 植物叶片识别 SVM 人工智能 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]