高级检索
    汪献义, 邢艳秋, 尤号田, 邢涛, 舒苏. 基于近邻几何特征的TLS林分点云分类研究[J]. 北京林业大学学报, 2019, 41(6): 138-146. DOI: 10.13332/j.1000-1522.20180308
    引用本文: 汪献义, 邢艳秋, 尤号田, 邢涛, 舒苏. 基于近邻几何特征的TLS林分点云分类研究[J]. 北京林业大学学报, 2019, 41(6): 138-146. DOI: 10.13332/j.1000-1522.20180308
    Wang Xianyi, Xing Yanqiu, You Haotian, Xing Tao, Shu Su. TLS point cloud classification of forest based on nearby geometric features[J]. Journal of Beijing Forestry University, 2019, 41(6): 138-146. DOI: 10.13332/j.1000-1522.20180308
    Citation: Wang Xianyi, Xing Yanqiu, You Haotian, Xing Tao, Shu Su. TLS point cloud classification of forest based on nearby geometric features[J]. Journal of Beijing Forestry University, 2019, 41(6): 138-146. DOI: 10.13332/j.1000-1522.20180308

    基于近邻几何特征的TLS林分点云分类研究

    TLS point cloud classification of forest based on nearby geometric features

    • 摘要:
      目的在地面激光雷达点云分类任务中多存在特征维度较高的问题,然而当点云数量较多,分类任务中构造较高维度的特征往往需要较多的计算成本和运行内存。为了解决这一问题,本研究提出用近邻点构造5个几何特征训练成熟分类器,以期在将林分点云分为地面、树干与枝叶3个类别的同时达到降低特征维度的目的。
      方法在构造特征的过程中采用近邻值为140的快速KDtree搜索近邻点,获得近邻点后利用其计算协方差矩阵特征值、法向量、曲率、方差和最大高程差构造5个几何特征训练分类器。为了检验本研究构造的特征在林分点云分类中的稳定性,分类器分别采用随机森林和xgboost做比较研究。本研究的实验数据均来自地面激光雷达扫描获得的单站蒙古栎人工林点云数据。
      结果使用随机森林和xgboost分类器训练的模型在测试集中正确估计样本数量和样本总量的比值分别为0.932 1和0.936 3。这两个分类器在地面、树干和枝叶这3个类别中的查准率达到0.97、0.93、和0.91以上,且在这3个类别中的分类结果中xgboost较随机森林均有千分级的优势。
      结论结果表明本研究构造的特征能够完成林分点云分类任务,在保证点云分类准确率的基础上,既减少了特征维度,又有助于提高特征计算效率,具有较高的稳定性。本研究的分类结果可为林分参数反演和生物量估计等研究奠定基础。

       

      Abstract:
      ObjectiveMost researches on TLS point cloud classification always calculate high dimensional features. But the higher the dimensional features calculated, the more the calculating consumption and running memory needed. So to solve the problem, we designed five geometric features of nearby points to train existing classifier. And then it was used to label the forest point clouds into ground, stem and leaf.
      MethodThe 140 neighbors gotten by fast KDtree were used to compute the five features, including eigen values of covariance matrix, normal vector, curvature, variance and maximum distance of elevation. And the classifier could be trained with all of them. In order to check the stability of the five features in forest point cloud classification, both random forest and xgboost were introduced. The data in this research were all obtained from Mongolian oak plantation by TLS.
      ResultIn test sets, the ratios between correct prediction samples and total samples were 0.932 1 and 0.936 3 with random forest and xgboost. Both classifier precision in ground, stem and leaf reached 0.97, 0.93 and 0.91 or more. And compared with random forest, the xgboost’s performance in the three categories had millesimal advantage.
      ConclusionOn the basis of ensuring the accuracy of point cloud classification, the five features are not only with less dimensions but also helpful to enhance the efficiency of feature computation. It shows they can deal with point cloud classification problem in forest well and have high stability. And the classification result in this research will be helpful for forest parameter extraction and biomass estimation.

       

    /

    返回文章
    返回