A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Monitoring of Discolored Trees Caused by Pine Wilt Disease Based on Unsupervised Learning with Decision Fusion Using UAV Images
Pine wilt disease (PWD) has caused severe damage to ecosystems worldwide. Monitoring PWD is urgent due to its rapid spread. Unsupervised methods are more suitable for the monitoring needs of PWD, as they have the advantages of being fast and not limited by samples. We propose an unsupervised method with decision fusion that combines adaptive threshold and Lab spatial clustering. The method avoids the sample problem, and fuses the strengths of different algorithms. First, the modified ExG-ExR index is proposed for adaptive threshold segmentation to obtain an initial result. Then, k-means and Fuzzy C-means in Lab color space are established for an iterative calculation to achieve two initial results. The final result is obtained from the three initial extraction results by the majority voting rule. Experimental results on unmanned aerial vehicle images in the Laoshan area of Qingdao show that this method has high accuracy and strong robustness, with the average accuracy and F1-score reaching 91.35% and 0.8373, respectively. The method can help provide helpful information for effective control and tactical management of PWD.
Monitoring of Discolored Trees Caused by Pine Wilt Disease Based on Unsupervised Learning with Decision Fusion Using UAV Images
Pine wilt disease (PWD) has caused severe damage to ecosystems worldwide. Monitoring PWD is urgent due to its rapid spread. Unsupervised methods are more suitable for the monitoring needs of PWD, as they have the advantages of being fast and not limited by samples. We propose an unsupervised method with decision fusion that combines adaptive threshold and Lab spatial clustering. The method avoids the sample problem, and fuses the strengths of different algorithms. First, the modified ExG-ExR index is proposed for adaptive threshold segmentation to obtain an initial result. Then, k-means and Fuzzy C-means in Lab color space are established for an iterative calculation to achieve two initial results. The final result is obtained from the three initial extraction results by the majority voting rule. Experimental results on unmanned aerial vehicle images in the Laoshan area of Qingdao show that this method has high accuracy and strong robustness, with the average accuracy and F1-score reaching 91.35% and 0.8373, respectively. The method can help provide helpful information for effective control and tactical management of PWD.
Monitoring of Discolored Trees Caused by Pine Wilt Disease Based on Unsupervised Learning with Decision Fusion Using UAV Images
Jianhua Wan (author) / Lujuan Wu (author) / Shuhua Zhang (author) / Shanwei Liu (author) / Mingming Xu (author) / Hui Sheng (author) / Jianyong Cui (author)
2022
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
A Detection Method for Individual Infected Pine Trees with Pine Wilt Disease Based on Deep Learning
| DOAJ | 2022
Pine-YOLO: A Method for Detecting Pine Wilt Disease in Unmanned Aerial Vehicle Remote Sensing Images
| DOAJ | 2024
Hydraulic Modeling and Discolored Water
| British Library Conference Proceedings | 1994
Early-Stage Pine Wilt Disease Detection via Multi-Feature Fusion in UAV Imagery
| DOAJ | 2024