摘要
随着遥感技术的发展,遥感技术也在土地的分类、利用及覆盖中广泛应用。然而,遥感分类方法的优劣直接关系到分类能否成功或符合要求。本文详细分析了监督分类方法、非监督分类方法和卷积神经网络分类方法,并归纳了每种方法的优缺点。最终以WorldView-3卫星影像为例,运用不同的分类方法对遥感影像进行分类,并对处理结果进行分析评价。整体来说,神经网络分类方法在分类效果上要比传统的分类方法处理速度快、精度高。
With the development of remote sensing technology,remote sensing technology is also widely used in the classification,utilization and coverage of land.However,the quality of remote sensing classification methods is directly related to the success of classification and compliance with the requirements.In this paper,the supervised classification method,the unsupervised classification method and the artificial neural network classification method were analyzed in detail,and the advantages and disadvantages of each method were summarized.Finally,taking worldView-3 satellite imagery as an example,different classification methods were applied to classify remote sensing images,and the processing results were analyzed and evaluated.Overall,the classification method of neural network was faster and more accurate than the traditional classification method in terms of classification effect.
作者
王凌云
李可新
陈曦
蔡鑫垚
WANG Lingyun;LI Kexin;CHENG Xi;CAI Xinyao(The Third Topographic Surveying Brigade,Ministry of Nature Resources,Harbin Heilongjiang 150006,China)
出处
《北京测绘》
2023年第5期643-648,共6页
Beijing Surveying and Mapping
关键词
遥感
分类方法
神经网络
精度分析
remote sensing
classification methods
neural networks
precision analysis
