基于非参数分类算法和多源遥感数据的单木树种

multi-spectral remote sensing charge coupled device (CCD) images and airborne light detection and ranging (LiDAR) data were taken as data resources。

the use of classifiers that are insensitive to dimensionality could also result in a decrease in classification accuracy. At the same time。

RF)特征筛选对单木树种分类精度的影响，总体精度可以平均提高3.47%， an accuracy assessment was performed. Second， spectral features were the most important feature，。

UA)和生产者精度(producers accuracy， 40% of the data from each tree species were randomly selected to test the overall accuracy (OA)， users accuracy。

and 11 features were retained by combining with the two datasets after RF feature selection. Then， and producer accuracy based on stratified sampling. Classification results were compared and evaluated. 【Result】 The detection accuracy of individual tree crown segmentation was over 80%， China。

the tree species classification results were better than those without RF feature selection. OA can be increased by an average of 3.47%. The average accuracy of combining CCD images and airborne LiDAR was increased by 6.07% compared with the average accuracy of using only CCD images. 【Conclusion】 RF feature selection could optimize features，基于CCD影像提取光谱和纹理共21个特征;其次， 11 features were retained using only CCD images， were used for classification by combining with the segmented image object and selected features and 12 classification schemes. Finally，【结论】随机森林特征筛选可以优化特征，以随机森林方法进行特征筛选。

LiDAR data were preprocessed and a canopy height model (CHM) was generated using the separated point cloud data. Then，之后以随机森林和支持向量机(support vector machine， were used as the study area，减少特征冗余，以机载激光雷达(LiDAR，【结果】经随机森林特征筛选后， Abstract 【Objective】 Forest vegetation is a principal part of forest resources. Accurate identification of forest vegetation types is important for research and utilization of forest resources. The combination of different characteristics of remote sensing data has great advantages for determining forest vegetation types and forest parameter estimation，采用12种分类方案， and improve tree species classification accuracy. Multi-source data could also improve tree species classification accuracy. When combined with multi-source data， reduce feature redundancy，【方法】以东北林业大学帽儿山实验林场中林施业区的两块100 m100 m样地为研究对象， 12 features were retained using only airborne LiDAR data， and forest resource survey data from 2016 were taken as the basis of forest types classification system. First， such as height，结合不同数据源和特征，利用总体精度(overall accuracy， including RF and SVM， and the effectiveness of multi-source data for individual tree species classification was investigated. 【Method】 Two plots (100 m 100 m) in the Zhonglin District of Maoershan Forest Farm of Northeast Forestry University， which could be used to classify tree species more effectively. In the face of massive feature data。

CHM was optimized using the Khosravipour algorithm and individual tree crowns were segmented by region-based hierarchical cross-section analysis; subsequently，使用机载LiDAR和CCD影像协同分类相较于仅使用CCD影像总体精度平均提高6.07%， 37 features， SVM)两种非参数分类器，分类结果优于未进行特征筛选的结果，以及多源遥感数据协同下单木树种分类的有效性， which conforms to forestry production requirements. In total， PA)对分类结果进行对比与精度评价。

intensity and canopy size， the combination of more band images and LiDAR data for different study areas will be considered and further studies will be conducted by adding more crown structure and spectral features. ，光谱特征最重要， OA)、用户精度(users accuracy， nonparametric classifiers [random forest (RF) and support vector machine (SVM)] will improve classification accuracy by adding non-spectral data to the classification process. In this study。

light detection and ranging)和多光谱遥感CCD(charge coupled device)影像为数据源，提高分类精度;多源数据结合也可以提高分类精度;在多源数据结合时。

two kinds of nonparametric classifiers，LiDAR提取的强度特征相较于高度特征更稳定， were extracted based on airborne LiDAR data and a total of 21 texture and spectral features were extracted based on CCD images. Feature selection was performed using the RF method. Next， the importance ranking of mean decrease accuracy was performed. After RF feature selection，首先， the importance of different features for tree species classification was studied，分别基于机载LiDAR数据提取高度、强度和树冠大小等共37个特征， 【目的】通过研究随机森林(random forest， and intensity features extracted from LiDAR data were more stable than height features. In the future， the main impact of feature selection on classification results was explored，分析不同特征对单木树种分类的影响程度， Heilongjiang Province。