基于彩色通道相似性图像分割方法的植物叶面积计算  

Measurement of plant leaf area based on image segmentation of color channel similarity

作　　者：韩殿元[1,2] 黄心渊[1] 付慧[1] 

机构地区：[1]北京林业大学信息学院,北京100083 [2]潍坊学院计算机工程学院,潍坊261061

出　　处：《农业工程学报》2012年第6期179-183,共5页Transactions of the Chinese Society of Agricultural Engineering

基　　金：国家948计划项目(2009-4-41)

摘　　要：为了快速、准确地测量植物叶面积,该文提出了基于彩色通道相似性图像分割的植物叶面积测量方法。该文基于彩色图像,利用像素彩色通道的相似性和自适应方法得到的阈值分割叶片区域,并分别统计叶片和参考矩形的像素数,进而计算植物叶面积。试验表明,该方法得到的植物叶片区域更准确,并且对叶片阴影、花斑等具有很强的鲁棒性,错分率为仅为1.23%,具有较高的精度。该方法是一种实用的通过拍照计算植物叶面积的方法,可嵌入到手机等移动设备中。To measure plant leaf area quickly and accurately, a method of plant leaf area measurement based on color channel similarity image segmentation was put forward. In this paper, plant leaf area was separated by using the color channel similarity and an auto adaptive threshold. After the pixels of the leaf area and the referenced rectangle were counted, the leaf area was calculated. Experiments showed that this method was more accurate and had a strong robustness to the shadow and piebald, and the misclassification error was only 1.23%. So this method greatly improves the accuracy of measurement and is a practical method of plant leaf area measurement by taking pictures and can be embedded in mobile devices such as smart mobile phones.

关 键 词：图像处理 图像分割 测量 叶面积 彩色通道相似性 

分 类 号：S126[农业科学—农业基础科学]