Data archive for 'Convolutional neural networks facilitate river barrier detection and evidence severe habitat fragmentation in the Mekong River biodiversity hotspot'

This repository contains the code and database used in the paper 'Convolutional neural networks facilitate river barrier detection and evidence severe habitat fragmentation in the Mekong River biodiversity hotspot'. The 'FCOS' folder contains Python files of the barrier detection model (FCOS ResNext-101-FPN) which was trained to detect river barriers from remotely sensed photographic images in the MMDetection framework. The 'R_script' folder contains R files used in the paper. Coordinate.R was used to extract coordinates from bounding boxes in each TIF image. CAFI.R was used to calculate the CAFI index in each sub-catchment. The 'Barrier Database' folder contains the 'Training Validation Database' used during the training process, and the 'Mekong River Barrier Database' generated in the study. For more information on the MMDetection framework, refer to the following GitHub repository: https://github.com/open-mmlab/mmdetection本仓库收录了论文《卷积神经网络助力湄公河生物多样性热点区域河流障碍物检测及证实严重生境破碎化》所使用的代码与数据库。其中`FCOS`文件夹存储了障碍物检测模型（FCOS ResNext-101-FPN）的Python代码，该模型基于MMDetection框架搭建，用于从遥感影像中识别河流障碍物。`R_script`文件夹包含本论文使用的R脚本文件：`Coordinate.R`用于从每张TIF格式图像的边界框中提取坐标信息；`CAFI.R`用于计算各子流域的CAFI指数。`Barrier Database`文件夹包含模型训练过程中使用的「训练验证数据库」，以及本研究生成的「湄公河河流障碍物数据库」。如需了解MMDetection框架的更多信息，请参阅以下GitHub仓库：https://github.com/open-mmlab/mmdetection