高级检索
    李爽, 张潇巍, 谭旭, 徐凌飞, 吕生华, 文剑. 基于深度学习的树木根系探地雷达多目标参数反演识别[J]. 北京林业大学学报. DOI: 10.12171/j.1000-1522.20230259
    引用本文: 李爽, 张潇巍, 谭旭, 徐凌飞, 吕生华, 文剑. 基于深度学习的树木根系探地雷达多目标参数反演识别[J]. 北京林业大学学报. DOI: 10.12171/j.1000-1522.20230259
    shu
    Li Shuang, Zhang Xiaowei, Tan Xu, Xu Lingfei, Lü Shenghua, Wen Jian. Deep learning-based inverse identification of multi-target parameters for tree rooting ground-penetrating radar[J]. Journal of Beijing Forestry University. DOI: 10.12171/j.1000-1522.20230259
    Citation: Li Shuang, Zhang Xiaowei, Tan Xu, Xu Lingfei, Lü Shenghua, Wen Jian. Deep learning-based inverse identification of multi-target parameters for tree rooting ground-penetrating radar[J]. Journal of Beijing Forestry University. DOI: 10.12171/j.1000-1522.20230259
    shu

    基于深度学习的树木根系探地雷达多目标参数反演识别

    Deep learning-based inverse identification of multi-target parameters for tree rooting ground-penetrating radar

      摘要
    • 摘要:
      目的 使用深度学习方法实现对根系雷达图像的多目标检测和多参数估计。
      方法 构建了一种以YOLOv5s和CNN-LSTM为主要框架的网络模型实现对根系雷达扫描图像的多目标检测和多参数估计。首先，通过仿真模拟和预埋试验获取试验所需的根系雷达剖面图数据，同时为了增加数据的多样性，使用CycleGAN风格迁移网络获取了一批具有真实雷达图像特征的仿真数据；然后，使用YOLOv5s目标检测网络识别并提取根系响应区域；接着引入频域变换，获取频域特征，并将根系雷达图像的时域特征和频域特征融合；最后，利用卷积神经网络、卷积注意力机制以及长短期记忆网络强调和提取与根系参数相关的信息特征，并使用多任务学习的方法实现对根系半径、深度、相对介电常数以及水平倾角的预测。
      结果 （1）仿真试验中，根系半径估计的最大误差是4.3 mm，R2为0.980，均方根误差为1.32，深度估计的最大误差是35.1 mm，R2为0.962，均方根误差为17.68，相对介电常数估计的最大误差是3.1，R2为0.960，均方根误差为1.10，水平倾角估计的最大误差是10.2°，R2为0.821，均方根误差是4.96。（2）在实测数据上对根系半径估计的平均相对误差是9.112%，深度估计的平均相对误差是5.772%，水平倾角估计的平均相对误差是11.25%。
      结论 本文提出的基于深度学习与探地雷达的多目标检测方法可以为根系检测和根系参数估计提供便利。

       

      Abstract:
      Objective This paper takes multi-target detection and multi-parameter estimation of root radar images using deep learning methods.
      Method In this study, a network model with YOLOv5s and CNN-LSTM as the main framework was constructed to achieve multi-target detection and multi-parameter estimation of root radar scanning images. Firstly, the root system radar profile data required for the experiments were obtained through simulation and pre-embedding experiments, and at the same time, in order to increase the diversity of the data, a batch of simulation data with the characteristics of real radar images were obtained using CycleGAN style migration network; then, the root system response region was identified and extracted using the YOLOv5s target detection network; then, the frequency domain transform was introduced to obtain the frequency domain features and fuse the time domain features and frequency domain features of the root system radar image; finally, a convolutional neural network (CNN), a convolutional attention mechanism, and a long short-term memory network (LSTM) were used to emphasize and extract the information features related to the root system parameters, and a multi-task learning approach was used to achieve the prediction of the root system radius, depth, relative permittivity, and horizontal angle.
      Result (1) In the simulation experiments, the maximum error in the estimation of root radius was 4.3 mm with R2 of 0.980 and a root mean square error of 1.32. The maximum error in the estimation of depth was 35.1 mm with R2 of 0.962 and a root mean square error of 17.68, and the maximum error in the estimation of relative permittivity was 3.1 with R2 of 0.960 and a root mean square error of 1.10, and the maximum error in the estimation of the horizontal angle was 10.2°, with R2 of 0.821 and the root mean square error was 4.96. (2) The average relative error of root radius estimation on measured data was 9.112%, the average relative error of depth estimation was 5.772%, and the average relative error of horizontal angle estimation was 11.25%.
      Conclusion The experimental data show that the method proposed in this paper can facilitate root detection and root parameter estimation.

       

      • HTML全文
      • 参考文献(33)
      • 相关文章
      • 施引文献
    • 资源附件(0)

    /

    返回文章
    返回