Centroid-UNet: Detecting Centroids in Aerial Images

TL;DR

This paper introduces Centroid-UNet, a deep learning model based on U-Net architecture, designed to accurately locate object centroids in aerial images for applications where object shape details are unnecessary.

Contribution

The study adapts the U-Net architecture for centroid detection in satellite images, demonstrating its effectiveness and simplicity in two case studies.

Findings

Achieved high accuracy in centroid detection for buildings and coconut trees.

Compared favorably with existing methods in terms of accuracy and simplicity.

Provided open-source code and models for community use.

Abstract

In many applications of aerial/satellite image analysis (remote sensing), the generation of exact shapes of objects is a cumbersome task. In most remote sensing applications such as counting objects requires only location estimation of objects. Hence, locating object centroids in aerial/satellite images is an easy solution for tasks where the object's exact shape is not necessary. Thus, this study focuses on assessing the feasibility of using deep neural networks for locating object centroids in satellite images. Name of our model is Centroid-UNet. The Centroid-UNet model is based on classic U-Net semantic segmentation architecture. We modified and adapted the U-Net semantic segmentation architecture into a centroid detection model preserving the simplicity of the original model. Furthermore, we have tested and evaluated our model with two case studies involving aerial/satellite images. Those two case studies are building centroid detection case study and coconut tree centroid detection case study. Our evaluation results have reached comparably good accuracy compared to other methods, and also offer simplicity. The code and models developed under this study are also available in the Centroid-UNet GitHub repository: https://github.com/gicait/centroid-unet