Efficient wildlife monitoring: Deep learning-based detection and counting of green turtles in coastal areas

Drones have recently been used to assess wildlife populations and their abundance. The automatic detection of target animals in drone footage enables efficient abundance estimation. However, accurately detecting animals remains challenging, especially in complex field environments. Moreover, automating the tracking of individuals across consecutive images and counting them along transect lines is necessary to apply drones to line-transect surveys. In this dataset, we prepared images obtained at Japanese coastal areas from drones for deep-learning-based models to automatically detect green turtles (Chelonia mydas). The images contain training and validation data for the YOLOv7 model. In addition, for testing the BoT-SORT object-tracking algorithm to track green turtles, we prepared eight drone footage clips.