In recent decades, rapid urbanization in India, fueled by population growth, has spurred the construction of new cities and the expansion of existing urban centers, extending into larger peripheral areas—a trend common in developing nations globally. Despite its widespread occurrence, accurately mapping human settlements and building distributions remains a challenge. State authorities and private enterprises, including Microsoft and Google, have sought to comprehensively capture this data. While existing initiatives offer a global perspective, challenges persist, especially in precision, for countries like India where cities boast highly dense and mixed urban development. This study explores the potential of Sentinel-2 images for building footprint mapping and change detection at 2.5 m spatial resolution, focusing on the dynamic Delhi National Capital Region (NCR). Recent works demonstrate sub-pixel accuracy in deriving building footprint maps through deep learning on Sentinel-2 imagery. The research aims to develop on existing findings taking into account seasonal variations and using improved training labels to further extend these findings to India.

This study presents a novel pretraining approach for self-supervised learning on optical Earth observation satellite data based on the masked autoencoder paradigm. Unlike typical methods limited to a single sensor’s data, our method operates across various sensors by encoding physical sensor parameters into the learning step, to account for the unique differences among sensor designs. This enables merging training datasets acquired with different sensors as well as performing inference in a sensor-independent manner. Successful encoding of the sensor parameters through our approach is shown through testing on a downstream land-cover mapping task, where baseline models are outperformed by up to 6 points for the F1-score.

In this work, we investigate two critical design decisions that arise when adapting the concept of contrastive learning to optical earth observation data. We work within the framework of SimCLR in order to pre-train neural network architectures and test their respective applicability on downstream datasets over various common remote sensing tasks. During the training, we focus in detail on the concept of the creation of positive and negative pairs due to the introduction of different batch sampling strategies and color-related augmentations. We report all performance metrics as a function of the available training data and discuss underlying mechanisms in order to drive the understanding of contrastive learning in optical Earth observation forward.

In the field of urban monitoring, accurate mapping of building structures and the corresponding changes is one of the most essential pieces of information. In many cases, this problem is approached with very high-resolution images, needed due to the spatial complexity in urban environments. And still, many binary change detection (CD) methods cannot segregate the changes introduced by the generation of new building structures from seasonal changes or other semantic changes. In this paper, we investigate a simple approach for building CD from freely available Sentinel-2 images, that neither depends on high-resolution imagery nor is prone to be negatively affected by seasonal changes. The proposed approach is simple and mainly based on the utilization of Sentinel-2 and existing corresponding building footprint information. We discuss in detail all the necessary steps to produce sub-resolution CD maps and shine a light on the critical influence of georeferencing correction.

Change detection (CD) is one of the most researched areas in remote sensing. However, most CD methods assume that the pre-change and post-change images are acquired by the same sensor, having the same set of spectral bands and same spatial resolution. This severely limits the applicability of CD methods. It is not trivial to apply the existing CD methods in multisensor scenario. Towards this direction, we propose an unsupervised CD method that can handle large differences in spatial resolution and can work with completely different set of spectral bands. The proposed method uses a self-supervised super-resolution strategy to upsample the lower resolution image, thus mitigating differences in spatial resolution. To mitigate spectral differences, a self-supervised learning strategy is used that ingests both images as input and trains a network using self-supervised loss accounting for the spectral differences in both images. Once trained this network is used in deep change vector analysis framework for change detection. We validated the proposed method in an experimental setup where the pre-change and post-change images have different spatial resolution (10m and 20 m/pixel) and completely disjoint set of spectral bands.