Extraction of Pine Wilt Disease Regions Using UAV RGB Imagery and Improved Mask R-CNN Models Fused with ConvNeXt
Pine wilt disease (PWD) is one of the most concerning diseases in forestry and poses a considerable threat to forests. Since the deep learning approach can interpret the raw images acquired by UAVs, it provides an effective means for forest health detection. However, the fact that only PWD can be de...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Forests |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4907/14/8/1672 |
| _version_ | 1797584728074747904 |
|---|---|
| author | Zhenyu Wu Xiangtao Jiang |
| author_facet | Zhenyu Wu Xiangtao Jiang |
| author_sort | Zhenyu Wu |
| collection | DOAJ |
| description | Pine wilt disease (PWD) is one of the most concerning diseases in forestry and poses a considerable threat to forests. Since the deep learning approach can interpret the raw images acquired by UAVs, it provides an effective means for forest health detection. However, the fact that only PWD can be detected but not the degree of infection can be evaluated hinders forest management, so it is necessary to establish an effective method to accurately detect PWD and extract regions infected by PWD. Therefore, a Mask R-CNN-based PWD detection and extraction algorithm is proposed in this paper. Firstly, the extraction of image features is improved by using the advanced ConvNeXt network. Then, it is proposed to change the original multi-scale structure to PA-FPN and normalize it by using GN and WS methods, which effectively enhances the data exchange between the bottom and top layers under low Batch-size training. Finally, a branch is added to the Mask module to improve the ability to extract objects using fusion. In addition, a PWD region extraction module is proposed in this paper for evaluating the damage caused by PWD. The experimental results show that the improved method proposed in this paper can achieve 91.9% recognition precision, 90.2% mapping precision, and 89.3% recognition rate of the affected regions on the PWD dataset. It can effectively identify the distribution of diseased pine trees and calculate the damage proportion in a relatively accurate way to facilitate the management of forests. |
| first_indexed | 2024-03-10T23:56:43Z |
| format | Article |
| id | doaj.art-deec61afb9224d8a850d0f44f80956fd |
| institution | Directory Open Access Journal |
| issn | 1999-4907 |
| language | English |
| last_indexed | 2024-03-10T23:56:43Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Forests |
| spelling | doaj.art-deec61afb9224d8a850d0f44f80956fd2023-11-19T01:10:22ZengMDPI AGForests1999-49072023-08-01148167210.3390/f14081672Extraction of Pine Wilt Disease Regions Using UAV RGB Imagery and Improved Mask R-CNN Models Fused with ConvNeXtZhenyu Wu0Xiangtao Jiang1College of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410018, ChinaCollege of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha 410018, ChinaPine wilt disease (PWD) is one of the most concerning diseases in forestry and poses a considerable threat to forests. Since the deep learning approach can interpret the raw images acquired by UAVs, it provides an effective means for forest health detection. However, the fact that only PWD can be detected but not the degree of infection can be evaluated hinders forest management, so it is necessary to establish an effective method to accurately detect PWD and extract regions infected by PWD. Therefore, a Mask R-CNN-based PWD detection and extraction algorithm is proposed in this paper. Firstly, the extraction of image features is improved by using the advanced ConvNeXt network. Then, it is proposed to change the original multi-scale structure to PA-FPN and normalize it by using GN and WS methods, which effectively enhances the data exchange between the bottom and top layers under low Batch-size training. Finally, a branch is added to the Mask module to improve the ability to extract objects using fusion. In addition, a PWD region extraction module is proposed in this paper for evaluating the damage caused by PWD. The experimental results show that the improved method proposed in this paper can achieve 91.9% recognition precision, 90.2% mapping precision, and 89.3% recognition rate of the affected regions on the PWD dataset. It can effectively identify the distribution of diseased pine trees and calculate the damage proportion in a relatively accurate way to facilitate the management of forests.https://www.mdpi.com/1999-4907/14/8/1672pine wilt diseasedisaster assessmentUAV-based RGB imageryinstance segmentation |
| spellingShingle | Zhenyu Wu Xiangtao Jiang Extraction of Pine Wilt Disease Regions Using UAV RGB Imagery and Improved Mask R-CNN Models Fused with ConvNeXt Forests pine wilt disease disaster assessment UAV-based RGB imagery instance segmentation |
| title | Extraction of Pine Wilt Disease Regions Using UAV RGB Imagery and Improved Mask R-CNN Models Fused with ConvNeXt |
| title_full | Extraction of Pine Wilt Disease Regions Using UAV RGB Imagery and Improved Mask R-CNN Models Fused with ConvNeXt |
| title_fullStr | Extraction of Pine Wilt Disease Regions Using UAV RGB Imagery and Improved Mask R-CNN Models Fused with ConvNeXt |
| title_full_unstemmed | Extraction of Pine Wilt Disease Regions Using UAV RGB Imagery and Improved Mask R-CNN Models Fused with ConvNeXt |
| title_short | Extraction of Pine Wilt Disease Regions Using UAV RGB Imagery and Improved Mask R-CNN Models Fused with ConvNeXt |
| title_sort | extraction of pine wilt disease regions using uav rgb imagery and improved mask r cnn models fused with convnext |
| topic | pine wilt disease disaster assessment UAV-based RGB imagery instance segmentation |
| url | https://www.mdpi.com/1999-4907/14/8/1672 |
| work_keys_str_mv | AT zhenyuwu extractionofpinewiltdiseaseregionsusinguavrgbimageryandimprovedmaskrcnnmodelsfusedwithconvnext AT xiangtaojiang extractionofpinewiltdiseaseregionsusinguavrgbimageryandimprovedmaskrcnnmodelsfusedwithconvnext |