基于Lab颜色空间数字照片的苔藓覆盖度提取研究

摘要

苔藓覆盖度既是描述其生长变化的重要参数,也是区域环境变化的重要指标.传统方法是通过统计苔藓所占网格数占总网格数的百分比来计算苔藓覆盖度,该方法费时费力,人为因素多,误差较大.本研究利用数码相机采集苔藓照片,通过分析数字图像Lab颜色空间的颜色分布特征,提出了提取苔藓覆盖度的算法,并将该方法结果与最大似然法、ISODATA法、自动分类法和传统网格目估法进行了比较.结果表明,基于 Lab 颜色空间变换方法的提取精度可达90%以上,优于其它4种,且该方法分类迅速、自动化程度高,是一种精确实用的方法.%The moss vegetation cover is not only a major parameter to describe its growth change,but also an important indicator of regional environmental changes.Traditional methods for extracting moss vegetation cover by counting the percentage of moss-occupied grids in the total number of grids,which is time-consuming,more human factors and more errors.This study uses the digital camera to collect the photos of moss at the first,then analyzes the photos'color distribution in Lab color space and proposes an algorithm to extract the mosses vegetation cover,finally,compares the results with the maximum likelihood method,ISODATA method,auto classification method and traditional grid statistical method.The consequences indicate that the method based on Lab color space conversion can improve the classification accuracy to 90%,moreover,this method is rapidly and higher automation,obviously better than the other 4 methods,which is a precise and practical method.