Satellite Imageries for Detection of Bangladesh’s Rural and Urban Areas Using YOLOv5 and CNN

Abstract

In recent years, there have been significant advancements in object identification in natural photos. However, when applying natural image object recognition techniques directly to satellite images, the results are often unsatisfactory. This is primarily due to inherent disparities in the object scale and orientation caused by the omniscient viewpoint of satellite imagery. The distinguishing factors between rural and urban areas lie in the objects that cover them. Furthermore, the complex backdrop of satellite photos poses challenges in accurately extracting features, leading to the omission of small objects in many regions. The performance of object detection, which is crucial for area identification, is also affected by dense object overlap and occlusion. To address these aforementioned issues, we made modifications to the generalized one-stage detector YOLOv5, specifically tailored for satellite photos. For this research, we manually collected data from Google Earth, meticulously labeling them and subsequently verifying them with human annotators. We then preprocessed the data using computer vision techniques, such as resizing and normalization. Next, we employed YOLOv5 and transfer learning-based CNN architectures of InceptionV3, DenseNet201, and Xception to compare their performances. The goal was to accurately identify rural and urban areas from remote sensing images.