中文摘要：

以内蒙古呼伦贝尔草原为研究区,共选择放牧区和非放牧区的47个观测样地,利用数码相机获取草地冠层照片,利用HLS彩色变换与RGB结合法及其他方法分别提取植被覆盖度。结果表明：本文发展的HLS彩色变换与RGB结合法提取的植被覆盖度的平均精度达85.85%,仅次于最大似然监督分类法。基于本文方法提取的实测点草地植被覆盖度与Landsat-5 TM影像计算的6种植被指数均有较好的正相关关系,其中与减小的比值植被指数（RSR）的相关性最高,从而建立了植被覆盖度遥感估算模型。遥感估算的研究区植被覆盖度空间差异明显,大部分草地植被覆盖度在0.7～0.9之间,而放牧区的草地植被覆盖度大多在0.5～0.7之间。

英文摘要：

In this study,we chose forty-seven observation fields,including grazing regions and non-grazing regions on Hulunbeier prairie.A digital camera was used to get photos of the grass canopy in every experimental field.The method of combining HLS color transformation with RGB model proposed by the author,as well as other several methods were used respectively to extract the vegetation coverage from those digital photos.The results indicated that the method of combining HLS color transformation with RGB model had higher accuracy of extracting vegetation coverage,averaging 85.85%,which was only inferior to the maximum likelihood supervising classification method.The vegetation coverage in observation fields extracted by the method of combining HLS color transformation with RGB model had good positive correlations with six kinds of vegetation indexes calculated from Landsat-5 TM image,and it had the best correlation with reduced specific value vegetation index（RSR）,which was used to establish the remote-sensing estimation model of vegetation coverage.The vegetation coverage estimated by remote-sensing model had an obvious spatial distribution feature in the studied area,the vegetation coverage was in the range of 0.7~0.9 in most regions,while that mostly varied from 0.5 to 0.7 in the grazing regions.