This research paper introduces a Federated Learning (FL) framework tailored for multimodal data privacy in disaster analysis, seamlessly integrating geospatial data privacy concerns, privacy preservation techniques, and disaster management applications. The framework adeptly addresses the ethical challenges posed by the usage of geospatial data and the delicate balance between open research and individual privacy. The proposed framework takes advantage of FL's decentralized nature to enable secure and privacy-preserving analysis of multimodal data, including textual content from social media and geospatial information from satellites. Employing advanced privacy preservation techniques, including anonymization and encryption, the framework ensures the protection of individual privacy rights while enhancing disaster management applications [6]. The intersection of these components is pivotal, emphasizing the careful equilibrium between data utility and privacy protection. Furthermore, the paper presents compelling results derived from the application of the proposed FL framework, showcasing its effectiveness in disaster analysis. The results highlight the successful integration of decentralized multimodal data, preserving privacy while contributing valuable insights to enhance disaster management strategies. This work underscores the importance of striking a thoughtful balance between data utility and individual privacy in advancing disaster research and response capabilities [7].

The unprecedented growth in internet users and various social media platforms increases storage and transfer of digital data and hence leads towards security of this data. The data security ensures the ownership of the multimedia content. Watermarking, steganography, and encryption are the key technologies used to preserve multimedia content ownership. The multispectral images consist of many spectral bands of the images where watermarking is challenging to apply. This paper presents the multispectral image watermarking using the Archimedes Optimization algorithm (AoA) and Discrete Cosine Transform (DCT). The AoA algorithm selects the spectral band with the highest variability as a cover image for watermarking. The DCT is used to watermark the secret logo in the cover image. The performance of the suggested watermarking scheme is estimated based on peak signal-to-noise ratio (PSNR), mean square error (MSE), and structural similarity index (SSIM).

Geospatial data is obtained from a variety of sources, including satellite sensors, GPS, and observations that are mostly based on the ground or aerial sensors. Through a variety of consistent, reliable indications that square measures available for storing, managing, and sharing spatial information, attention towards security, access management, and privacy policies of spatial information are extremely low. The user is sharing its location perpetually in several Geo-social applications in terms of latitude and longitude to urge information regarding their encompassing with the assistance of other locations and their recommendations. But users' information is exploited by the pursuit of users' activities if there are no privacy preservation techniques applied to the current geo spatial information. It protects the spatial data from attackers while permitting licensed users to issue spatial queries expeditiously by the service supplier. The target of this paper is to develop techniques to produce protection and security at dissemination for geospatial information through error-based transformation.

Spatio-temporal data harmonization is crucial for effectively managing soil health in sustainable agriculture. In the case of Indian soil, known for its diversity and complexity, integrating and analyzing different soil data sources presents significant challenges. This research proposes a novel approach that utilizes knowledge graphs to address these challenges in soil health management.This paper presents a knowledge graph that captures domain-specific information about Indian soil, such as soil properties, climatic conditions, land use patterns, and agricultural practices. This knowledge graph serves as a unified framework for integrating diverse data sources, such as soil survey lab reports, sensor data, and agricultural databases. An architecture for spatiotemporal data harmonization is presented in this paper, which includes data acquisition, pre-processing, knowledge graph construction, embedding techniques, alignment, harmonization, and integration. In the context of soil health management, it facilitates the integration and analysis of diverse soil data sources.

Geospatial technology holds immense potential for transforming the overall business and educational sector in India. Recently, a National Geospatial Policy was framed and adapted in India. Although, policy claims to foster plethora of opportunities. However, there are few challenges hindering the effective implementation of geospatial policy in various business sector as well as in education field. This paper presents overview of National Geospatial Policy of India with various provisions to upboost business, research, and education in India. Also, the impact, opportunities, and challenges for adaptation of this policy with respect to education are addressed.

Classification of land cover in remote sensing images is a very important area of research as it has various applications like mapping of land cover and land use, detecting changes on the specified area over a period, disaster management, city planning, etc. Remote sensing image classification is a complex task because of the presence of various artifacts in RS images. In recent years deep convolution networks have outperformed other approaches in computer vision Recently, fully convolutional networks-based techniques have reached state-of-the-art performance in semantic segmentation and made dense pixel-wise classification possible, with the U-Net architecture being one of the most prominent models. This paper compares Unet, ResNet34, Efficientnetb7, DenseNet201. Dataset of aerial images of Dubai captured by MBRSC satellites are used for model training and testing. Images are of different sizes hence the images are cropped to the nearest size divisible by 256. The models were compared in terms of accuracy, Jacard coefficient, IOU and training validation loss. Total params and training params of all models are also analyzed.

Due to unprecedented growth in freely and easily available remote sensing data, it is being used in plenty of applications and creating opportunities to contribute in this field. At the same time at the educational perspectives, very few countries are paying attention to remote sensing education at wider scale. Many of the developing countries still do not have exposure to remote sensing education. Therefore, it is required to have a learning platform to reach out to the learners using simple and easily available technologies. We are proposing mobile based remote sensing learning platform which will focus understanding, training, and research in remote sensing domain and have maximum outreach to all interested learners.

Geographic Information System (GIS) and Remote Sensing (RS) are domains that are recognized globally with lot of on-going research world-wide. In India, the scenario is just the opposite. These domains are less widespread in the field of higher education. This paper focuses on the issues that are prevalent in GIS and RS education in India. Literature survey states that out of 789 Universities, 11,443 Stand-alone institutions and 658 Autonomous Colleges only 40 institutions offer GIS and RS courses. An online survey was conducted to study the issues faced by the instructors and learners. The study of the responses endorses the fact that in India it is required to create awareness of these courses by including them in the University curriculum and providing necessary training.