2019/10/17 by Maria Papadomanolaki, Papadomanolaki, Maria, Sagar Verma +7 · 3 citations
Earth and Planetary Sciences · Engineering · Environmental Science · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Land Use and Ecosystem Services #Remote Sensing and Land Use #Remote-Sensing Image Classification #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1910.07778
openalex publication_date 2019/10/17 · openalex created_date 2022/10/11 · openalex updated_date 2026/07/28
beginabstract The advent of multitemporal high resolution data, like the\nCopernicus Sentinel-2, has enhanced significantly the potential of monitoring\nthe earth's surface and environmental dynamics. In this paper, we present a\nnovel deep learning framework for urban change detection which combines\nstate-of-the-art fully convolutional networks (similar to U-Net) for feature\nrepresentation and powerful recurrent networks (such as LSTMs) for temporal\nmodeling. We report our results on the recently publicly available bi-temporal\nOnera Satellite Change Detection (OSCD) Sentinel-2 dataset, enhancing the\ntemporal information with additional images of the same region on different\ndates. Moreover, we evaluate the performance of the recurrent networks as well\nas the use of the additional dates on the unseen test-set using an ensemble\ncross-validation strategy. All the developed models during the validation phase\nhave scored an overall accuracy of more than 95%, while the use of LSTMs and\nfurther temporal information, boost the F1 rate of the change class by an\nadditional 1.5%.\n