A Novel Forest Dynamic Growth Visualization Method by Incorporating Spatial Structural Parameters Based on Convolutional Neural Network
Current visual methods of forest dynamic growth mostly focus on the plot or stand level, which cannot express the morphological and structural characteristics of individual trees, as well as their statistical linkages, and causes each tree in the stand to grow at the same rate. In addition, these vi...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10356626/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850041034015244288 |
|---|---|
| author | Linlong Wang Huaiqing Zhang Kexin Lei Tingdong Yang Jing Zhang Zeyu Cui Rurao Fu Hongyan Yu Baowei Zhao Xianyin Wang |
| author_facet | Linlong Wang Huaiqing Zhang Kexin Lei Tingdong Yang Jing Zhang Zeyu Cui Rurao Fu Hongyan Yu Baowei Zhao Xianyin Wang |
| author_sort | Linlong Wang |
| collection | DOAJ |
| description | Current visual methods of forest dynamic growth mostly focus on the plot or stand level, which cannot express the morphological and structural characteristics of individual trees, as well as their statistical linkages, and causes each tree in the stand to grow at the same rate. In addition, these visual growth models still have some space for improvement in terms of prediction accuracy and multirelational data mining. In this article, uneven-aged Chinese fir (<italic>Cunninghamia lanceolata</italic>) plantations were chosen as our study subject and proposed a novel method of forest dynamic growth visualization modeling by incorporating spatial structure parameters and using convolutional neural network technique (FDGVM-CNN-SSP) to explore the effect of spatial structure on the morphological growth and to develop a prediction growth model of Chinese fir plantations by introducing a convolutional neural network (CNN) model. The results show that: first, spatial structural parameters C and U have a certain contribution to the forest growth, and C and U can explain 21.5%, 15.2%, and 9.3% of the variance in DBH, H, and CW growth models, respectively; second, CNN model outperformed machine learning algorithms SVR, MARS, Cubist, RF, and XGBoost in terms of prediction performance; third, based on FDGVM-CNN-SSP, we simulated Chinese fir plantations at individual tree level and stand level from 2018 to 2022 and found that DBH and H's fitting performance in measured and predicted data was highly consistent with <italic>R</italic><sup>2</sup> and root-mean-square error (RMSE) of 86.8%, 2.06 cm in DBH and 79.2%, 1.11 m in H, but CW's <italic>R</italic><sup>2</sup> and RMSE of 72.2%, 0.65 m caused crowding (C) inconsistency. |
| format | Article |
| id | doaj-art-40c8e38aa1d5433389c094e23d87db3d |
| institution | DOAJ |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-40c8e38aa1d5433389c094e23d87db3d2025-08-20T02:55:53ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352024-01-01173471348810.1109/JSTARS.2023.334244510356626A Novel Forest Dynamic Growth Visualization Method by Incorporating Spatial Structural Parameters Based on Convolutional Neural NetworkLinlong Wang0https://orcid.org/0009-0004-3337-4399Huaiqing Zhang1https://orcid.org/0000-0003-3874-5326Kexin Lei2Tingdong Yang3Jing Zhang4Zeyu Cui5Rurao Fu6Hongyan Yu7Baowei Zhao8Xianyin Wang9Institute of Forest Resource Information Techniques and Research Institute of Forest Policy and Information, Chinese Academy of Forestry, Beijing, ChinaInstitute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, ChinaInstitute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, ChinaInstitute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, ChinaInstitute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, ChinaInstitute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, ChinaInstitute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, ChinaQinghai Service Support Center, Qilian Mountain National Park, Qinghai, ChinaQinghai Service Support Center, Qilian Mountain National Park, Qinghai, ChinaQinghai Service Support Center, Qilian Mountain National Park, Qinghai, ChinaCurrent visual methods of forest dynamic growth mostly focus on the plot or stand level, which cannot express the morphological and structural characteristics of individual trees, as well as their statistical linkages, and causes each tree in the stand to grow at the same rate. In addition, these visual growth models still have some space for improvement in terms of prediction accuracy and multirelational data mining. In this article, uneven-aged Chinese fir (<italic>Cunninghamia lanceolata</italic>) plantations were chosen as our study subject and proposed a novel method of forest dynamic growth visualization modeling by incorporating spatial structure parameters and using convolutional neural network technique (FDGVM-CNN-SSP) to explore the effect of spatial structure on the morphological growth and to develop a prediction growth model of Chinese fir plantations by introducing a convolutional neural network (CNN) model. The results show that: first, spatial structural parameters C and U have a certain contribution to the forest growth, and C and U can explain 21.5%, 15.2%, and 9.3% of the variance in DBH, H, and CW growth models, respectively; second, CNN model outperformed machine learning algorithms SVR, MARS, Cubist, RF, and XGBoost in terms of prediction performance; third, based on FDGVM-CNN-SSP, we simulated Chinese fir plantations at individual tree level and stand level from 2018 to 2022 and found that DBH and H's fitting performance in measured and predicted data was highly consistent with <italic>R</italic><sup>2</sup> and root-mean-square error (RMSE) of 86.8%, 2.06 cm in DBH and 79.2%, 1.11 m in H, but CW's <italic>R</italic><sup>2</sup> and RMSE of 72.2%, 0.65 m caused crowding (C) inconsistency.https://ieeexplore.ieee.org/document/10356626/Convolutional neural network (CNN)forest growth model (FGM)spatial structurethree-dimensional (3-D) visualization |
| spellingShingle | Linlong Wang Huaiqing Zhang Kexin Lei Tingdong Yang Jing Zhang Zeyu Cui Rurao Fu Hongyan Yu Baowei Zhao Xianyin Wang A Novel Forest Dynamic Growth Visualization Method by Incorporating Spatial Structural Parameters Based on Convolutional Neural Network IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Convolutional neural network (CNN) forest growth model (FGM) spatial structure three-dimensional (3-D) visualization |
| title | A Novel Forest Dynamic Growth Visualization Method by Incorporating Spatial Structural Parameters Based on Convolutional Neural Network |
| title_full | A Novel Forest Dynamic Growth Visualization Method by Incorporating Spatial Structural Parameters Based on Convolutional Neural Network |
| title_fullStr | A Novel Forest Dynamic Growth Visualization Method by Incorporating Spatial Structural Parameters Based on Convolutional Neural Network |
| title_full_unstemmed | A Novel Forest Dynamic Growth Visualization Method by Incorporating Spatial Structural Parameters Based on Convolutional Neural Network |
| title_short | A Novel Forest Dynamic Growth Visualization Method by Incorporating Spatial Structural Parameters Based on Convolutional Neural Network |
| title_sort | novel forest dynamic growth visualization method by incorporating spatial structural parameters based on convolutional neural network |
| topic | Convolutional neural network (CNN) forest growth model (FGM) spatial structure three-dimensional (3-D) visualization |
| url | https://ieeexplore.ieee.org/document/10356626/ |
| work_keys_str_mv | AT linlongwang anovelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT huaiqingzhang anovelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT kexinlei anovelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT tingdongyang anovelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT jingzhang anovelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT zeyucui anovelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT ruraofu anovelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT hongyanyu anovelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT baoweizhao anovelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT xianyinwang anovelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT linlongwang novelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT huaiqingzhang novelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT kexinlei novelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT tingdongyang novelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT jingzhang novelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT zeyucui novelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT ruraofu novelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT hongyanyu novelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT baoweizhao novelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork AT xianyinwang novelforestdynamicgrowthvisualizationmethodbyincorporatingspatialstructuralparametersbasedonconvolutionalneuralnetwork |