摘要
为高效估算草地生物量,需要一种方法来提高草地分类精度和降低数据处理时间。该研究基于原始RGB图像采用IHS变换,进行绿度波段图像的融合,并对融合后的图像进行Mean Shift算法分类。结果表明,1)基于IHS图像的草地分类,在视觉上与实际地物更为吻合;2)与其他文献方法对比,本研究方法性能优越,精度达到95%以上;3)可以批量处理多张图像,提高了数据处理效率。
In order to estimate grassland biomass efficiently,a method is needed to improve grassland classification accuracy and reduce data processing time.Firstly,IHS transform was adopted based on the original RGB image,then the green band image was fused,and finally the fused image was classified by mean shift algorithm.Compared with many classification methods,the results showed that 1)the grassland classification based on IHS image was more consistent with the actual features visually.2)Compared with other methods in literature,this method had superior performance with the accuracy of more than 95%.3)It could process multiple images in batch,which greatly improved the data processing efficiency.
作者
康乐
陈伟
赵安琳
杨延征
KANG Le;CHEN Wei;ZHAO An-lin;YANG Yan-zheng(East China Inventory and Planning Institute,National Administration of Forestry and Grassland,Hangzhou 310019,Zhejiang,China;State Key Laboratory of Urban and Regional Ecology,Research Center for Eco-environmental Sciences,Chinese Academy of Sciences,100085,Beijing,China)
出处
《西北林学院学报》
CSCD
北大核心
2022年第5期188-193,共6页
Journal of Northwest Forestry University
基金
国家自然科学基金(41801181)。