论文部分内容阅读
【目的】为了加强热带林资源的保护,采用遥感技术对热带林植被进行分类研究。【方法】基于SPOT6高分辨率遥感影像,采用ESP多尺度分割评价模型与专家知识结合的方法确定最优分割尺度参数,在分割的基础上充分挖掘目标地物的光谱、形状及纹理信息,合理选择分类特征组合,建立分类规则,构建了一套基于面向对象的热带林多尺度分类方法。【结果】与单一尺度的分类方法相比,该方法分类精度有明显提高,分类总体精度达到84.46%,并且缩短了传统目视确定最优分割参数的时间,提高了分割效率和精度。【结论】基于面向对象的多尺度分类方法能够实现高精度的热带林植被信息提取,可为遥感分类技术在热带林的应用提供参考。
【Objective】 In order to strengthen the protection of tropical forest resources, remote sensing technology was applied to study the classification of tropical forest vegetation. 【Method】 Based on SPOT6 high-resolution remote sensing images, the optimal segmentation scale parameters were determined by combining ESP multiscale segmentation model with expert knowledge. The spectral, shape and texture information of the target object were fully tapped on the basis of segmentation, which was reasonable The classification features are selected and the classification rules are established. A set of object-oriented tropical forest multi-scale classification method is constructed. 【Result】 Compared with the single-scale classification method, the classification accuracy of this method was significantly improved, the overall accuracy of the classification reached 84.46%, and the time for the traditional visual determination of the optimal segmentation parameters was shortened, which improved the segmentation efficiency and accuracy. 【Conclusion】 The multi-scale object-oriented classification method can extract high-precision tropical forest vegetation information and provide a reference for the application of remote sensing classification in tropical forests.