中南林业科技大学学报2025，Vol.45Issue(3)：10-19,10.DOI:10.14067/j.cnki.1673-923x.2025.03.002
基于空间交叠主动学习的无人机光学影像桉树树冠检测
UAVs optical image-based Eucalyptus canopy detection using active learning with spatial overlap indicator
摘要
Abstract
[Objective]Deep learning-based canopy detection from remote sensing imagery is gradually becoming an important technique for forest inventory and monitoring.However,existing methodologies usually require a large number of labeled examples,resulting in high costs for both annotation and sample acquisition,thus limiting their applicability.To address this challenge,a novel active learning method was designed for comprehensive Eucalyptus canopy recognition utilizing UAV optical imagery.[Method]The proposed method employs a"Teacher-student"interactive learning mode.In each learning stage,the Teacher model generates candidate pseudo-samples,and high-value target pseudo-samples are obtained based on a pseudo-sample selecting strategy,then combined with existing labeled samples and input into the Student model for training.Subsequently,the parameters of the Student model are transferred to the Teacher model.After multiple rounds of interactive learning,the teacher model becomes the final model for tree canopy detection applications.Specifically,the method introduces a gradient Harmonized mechanism loss(GHM loss)in the Student model to reduce over-training on easy samples.It also designs a novel spatial overlap indicator to strengthen the model's learning emphasis on difficult pseudo-samples with severe canopy occlusions and coexistence of multiple tree species.Moreover,the method adopts multi-size grid mask and other data augmentation methods to enhance the model's adaptability to the spatial distribution patterns of trees,diverse lighting conditions and unconventional photographic angles.These enhancements collectively lead to a significant decrease in labeling workload and a improvement in model performance.[Result]Data collection was conducted on Guangxi Gaofeng forests using the DJI Phantom 4 Pro V2.0 UAVs,resulting in a dataset consisting of 8 000 annotated samples with 256 pixel×256 pixel images of both young and nearly mature forests.This dataset was used to compare our proposed method with other supervised learning and active learning approaches.The results demonstrate that our proposed active learning method outperforms both supervised learning methods and the latest active learning methods when using a limited number of samples.Specifically,with only 26%of the data used as samples,our method achieved an F1 score of 0.8,meeting the practical requirements for tree crown recognition.Furthermore,when the sample size increased to 34%,our method achieved an F1 score of 0.9,which is comparable to the performance of fully supervised learning methods.[Conclusion]The proposed active learning approach enables automatic acquisition of accurate tree crown boundaries under the constraint of limited samples,resulting in significant savings in data processing and sample preparation time.It offers advantages in terms of efficiency,convenience,and cost-effectiveness,thereby playing a crucial role in enhancing forest monitoring efficiency and automation level.关键词
树冠识别/主动学习/伪样本筛选/目标检测Key words
tree canopy recognition/active learning/pseudo sample filtering/object detection分类
农业科技引用本文复制引用
段炼,梁波,李震,罗天啸,黄超群,黄国斌..基于空间交叠主动学习的无人机光学影像桉树树冠检测[J].中南林业科技大学学报,2025,45(3):10-19,10.基金项目
广西林业科技推广示范项目(2023GXLK05) （2023GXLK05）
南宁市武鸣区科学研究与技术开发计划项目(20220107) （20220107）
广西大学生创新创业训练计划项目(S202410603111). （S202410603111）
