基于图像处理的叶斑病分级方法的研究

针对目前植物病害染病程度判别中以目测为主,存在着主观随意的缺陷,开发出一种基于计算机图像处理技术的病害分级新方法.同时,分析了图像分割中存在的各影响因素,运用Otsu法提取出叶片区域;提出在HSI颜色空间下选择H分量分割病斑以减少光照变化和叶脉的干扰,使用Sobel算子检测病斑边缘,分割出病斑区域,并通过计算病斑面积占叶片面积的百分比给出病害的染病级别.研究表明,使用该方法对植物叶部病害严重度进行分级具有快速精确的特点.

Grading Method of Leaf Spot Disease Based on Image Processing

Aimed at solving the problem that currently grading of plant diseases are mainly based on eyeballing,the paper developed a new disease grading method based on computer image processing.Having analyzed influencing factors exist in the process of image segmentation,leaf region was segmented using Otsu thresholding method;proposed to choose H component segmenting disease spot under HSI color space to reduce illumination changes and the leaf vein disturbance,used Sobel operator examine disease spot edges and then disease spot regions were segmented.Rank of disease infected is obtained by calculating the quotient of disease spot and leaf areas.Research indicate that using method put forward by the paper to grade plant leaf spot diseases is fast and accurate.