tree counting from UAV based aerial image with image processing

nayim sayad roshan¹

1) Faculty of Mechanical Engineering Noshirvani University of Technology, Babol, Mazandaran, Iran

Publication : 5th. International Congress On Engineering, Technology and Innovation(eticong.com)
Abstract :
presents a novel approach for tree counting using unmanned aerial vehicles (UAVs) and image processing techniques. The increasing availability and affordability of UAVs have made them a popular tool for capturing high-resolution aerial imagery. However, the manual counting of trees from these images can be time-consuming and prone to human error. Therefore, automated methods for tree counting are essential for efficient and accurate analysis of large-scale tree populations. we propose a workflow that combines UAV-based aerial imagery acquisition with image processing algorithms to automatically detect and count trees. The first step involves acquiring high-resolution aerial images using a UAV equipped with a high-resolution camera. These images are then pre-processed to enhance the visibility of tree crowns and remove any noise or unwanted objects. a series of image processing algorithms are applied to segment the tree crowns from the background and extract their features. These features include shape, color, and texture characteristics, which are used to distinguish trees from other objects in the image. a tree counting algorithm is developed to identify and count individual trees based on the extracted features. This algorithm takes into account the spatial distribution and arrangement of tree crowns to avoid over-counting or under-counting.
Keywords : UAV – tree counting – image processing