Dynamic Detection of Forest Change in Hunan Province Based on Sentinel-2 Images and Deep Learning

• Published in: Remote sensing (Basel, Switzerland) Switzerland), 2023-02, Vol.15 (3), p.628
• Main Authors: Xiang, Jun, Xing, Yuanjun, Wei, Wei, Yan, Enping, Jiang, Jiawei, Mo, Dengkui
• Format: Article
• Language: English
• Subjects: Change detection; Classification; Datasets; Deep learning; Deforestation; dynamic detection; forest change; Forest management; Forest resources; Forests; Hunan Province; Land resources; Landsat satellites; Mapping; Methods; Monitoring; Monitoring methods; Mountain regions; Mountainous areas; Neural networks; Remote sensing; Sentinel-2
• ISSN: 2072-4292; 2072-4292
• DOI: 10.3390/rs15030628

Dynamic detection of forest change is the fundamental method of monitoring forest resources and an essential means of preserving the accuracy and timeliness of forest land resource data. This study focuses on a deep learning-based method for dynamic forest change detection using Sentinel-2 satellite data, especially within mountainous areas. First, the performance of various deep learning models (U-Net++, U-Net, LinkNet, DeepLabV3+, and STANet) and various loss functions (CrossEntropyLoss(CELoss), DiceLoss, FocalLoss, and their combinations) are compared on a self-made dataset. Next, the best model and loss function is used to predict the annual forest change in Hunan Province from 2017 to 2021, and the detection results are evaluated in 12 sample areas. Finally, forest changes are detected in Sentinel-2 images for each quarter of 2017–2021. In addition, a dynamic detection map of forest change in Hunan Province from 2017 to 2021 is drawn. The results reveal that the U-Net++ model and the CELoss performed the best on the self-made dataset, with a Precision of 0.795, a Recall of 0.748, and an F1-score of 0.771. The results of annual and quarterly forest change detection were consistent with the changes in the Sentinel-2 images with accurate boundaries. This result demonstrates the high practicality and generalizability of the method used in this paper. This paper achieves a rapid and accurate extraction of multi-temporal Sentinel-2 image forest change areas based on the U-Net++ model, which can be used as a benchmark for future large territorial areas monitoring and management of forest resources.