Tree Enumeration Using Image Analytics

ABSTRACT--Accurate tree enumeration is essential for ecological monitoring, urban planning, and forest management. Traditionally, this task has been carried out manually, which is not only time-consuming but also prone to human error. This project aims to simplify and automate the process of tree counting using image analytics. By leveraging a deep learning-based object detection model—YOLOv8—we trained the system to detect and count trees from aerial and landscape images. The approach involves curating a custom dataset, annotating it using tools like Roboflow, and training the model on Google Colab. A simple web interface was developed using Flask to allow users to upload an image and receive real-time results showing the number and type of trees detected. The model performed well on various test images, showing a high detection accuracy. This system not only reduces manual effort but also provides a scalable and efficient solution for large-scale environmental data collection Keywords: Tree enumeration, Image analytics, YOLOv8, YOLOv9, YOLOv10, Deep learning, Object detection, Streamlit, Python, OpenCV, TensorFlow, Environmental monitoring, Forest management.