Urbanization has caused drastic changes in landscapes in many developing countries. This study examines the effects of urbanization on Land use and air quality in Kharagpur, a developing industrial city in West Bengal, India. Land use dynamics are analyzed using satellite data and advanced classification techniques from 2010 to 2022. Cellular automata -Markov chain-based future Land use scenarios have been analyzed. Air Quality Index (AQI) data from various land uses are collected and analyzed to examine correlations between air quality and LULC using three Generalized Linear Models (GLMs). The third Model considering Business as Usual and lockdown scenarios, highlights that urban areas have a significant positive correlation with AQI. In contrast, vegetation, agricultural land, and water bodies correlate negatively. The findings show that industrial operations, transportation emissions, and urban development affect air quality. This research aids land use planning and air quality management decisions.

The road is a requisite asset to urban infrastructure for generating nations' economic growth and development. Therefore, road distress mapping is paramount for maintenance planning. Fusing deep learning methods with GIS techniques, provides insight to obtaining new opportunities using satellite imageries through spatial, temporal, and spectral resolutions with data integration. In this context, the research proposes implementing an end-to-end convolutional neural network architecture for extracting the road networks and its application towards road distress mapping using very high-resolution (VHR) remote sensing images. Understanding the limitations of existing methods, the proposed model is tweaked to extract finer features and, thus, increase the likelihood of accurate prediction.

To reconcile social and economic progress with environmental conservation, natural resource management (NRM) focuses on the long-term sustainability of natural resources. Current and future generations must preserve major natural resource wherein, mangroves are one of the world's most productive and economically beneficial forests, growing in marshes, shorelines, coastline regions, and coastal zone of river deltas in tropical, subtropical, and few temperate coasts. In this study, we analyze the impact of the change in land cover under mangroves across India between 1997 and 2019 using the Google Earth Engine (GEE) platform and assess the biophysical indicators. The findings of our research reveal that the area covered by mangrove forests has expanded in the last two decades

This research employs fully convolutional neural networks, followed by the transfer learning method to extract buildings. The model was developed by utilizing layers of down sampling and upsampling. Two convolution layers and a ReLU activation function make up this model. To minimize overfitting the dropout layer is included. The outputs establish that the FCN model adequately predicts the pixels that correspond to buildings, but it also incorrectly predicts many non-building pixels as building pixels. The methodology of transfer learning using U-Net and the pre-trained model is utilized to improve the precision. The segmentation model library is employed, which provides access to about 25 encoders that can be used with the U-Net model. To improve outcomes, we used three alternative encoders. The study${}^{\prime}\mathrm{s}$findings show that the model performs better with Inception-V3 and U-Net than the other two encoders. The accuracy of the fully convolutional neural networks model was 89.86 while the accuracy of the Inception-V3 and U-Net architecture was 96.39 percent.

We propose the implementation of deep learning-based architecture for extracting the built-up and application towards solar photovoltaic potential estimation. The U-Net model proposed for the extraction of building rooftops using medium resolution satellite data. Building footprints from OpenStreetMap is used to generate labels necessary for training the model. The analysis of study region establishes that about 1.89 Giga Watt could be generated from building roofs of the study region. The roof area estimation is validated using polygonal building outlines from 25 km2of the city locality. Analysis of this kind could help in reaching the targets set for implementation of visualized scale of programs by the governing bodies.

To overcome the limitations of traditional land use modelling techniques, this paper introduces genetic algorithm (GA) to find optimum solution for the factors affecting the change in land use within the search space through developing and Agent Based Model(ABM). Optimized values of agents were then introduced to a goal-oriented environment where interactions are programmed to be self-constrained ABM. GA-ABM clearly outperformed the traditional CA approach in terms of Kappa indices of agreement, therefore, reducing the simulation uncertainty, improving the overall model accuracy. Modelled output for the year 2025 insists planning authorities for considering rapid expansion of paved spaces as a challenge to achieve sustainable growth, to fulfil sustainable development goals by the year 2030.

Digital Surface Model (DSM) provides the detailed structure and geometry of an urban environment. This paper proposes an approach of using a type of image-to-image translation deep learning model called cycle consistent adversarial networks for reconstructing DSM from monocular aerial imagery. The cycleGAN architecture consisted of two generators with an encoder-decoder network with skip connections and two discriminators that punishes structures at the scale of patches. The cycleGAN objective function was adapted for training on paired images. The evaluation was performed using mean square error (MSE) and zero normalized cross-correlation (ZNCC) for errors in reconstruction. cGAN model was considered as a baseline model for comparison of the proposed approach. The results using the proposed approach confirmed a higher reconstruction accuracy than previous studies that utilized conditional GAN.

Visualizing the pattern of urbanization and correlating it with Land Surface Temperature (LST) serves as vital information for understanding the phenomena of urban heat island and other heat-related issues. LST is an important variable to define microclimate, ecology, bio-geo-chemical and biodiversity of the region. The foremost objective of this study is to forecast LST using Artificial Neural Network (ANN) and geospatial technology as a tool. Temporal land use and LST for 1991, 2000, 2009 and 2017 along with the elevation details were used to define the pattern followed by deriving relationship to forecast LST. The results obtained signifies a relationship between rise in concrete area and reduction in open and vegetated spaces with rising surface temperatures. The forecasting equation developed from the model shows good accuracy for prediction. Outcomes of the study demonstrated the capability and proficiency of ANN models to forecast surface temperature considering various parameters for the complex and dynamic physical environment.

Mapping of urban environments is a challenging task because of the dense and heterogeneous nature. Constantly developing infrastructures necessitate frequent updating of the urban database. The spatial pattern of built-up land use indicates the growth of cities and incorporating vertical components in the calculation of built-up volume is essential to plan for a sustainable future. In this study, the urban built-up volume is estimated through built-up area that is derived from high-resolution LISS IV data using deep learning techniques and height data is derived from Cartosat-1 stereo images using photogrammetric methods. The results indicate that the eastern part of the city has urban volume development compared to the central business district. This result can be used as a reference indication for characterizing economic growth, proxy measure to various parameters such as population, traffic conditions, power consumption, telecom network planning, etc.

Multi temporal land use information were derived using two decades remote sensing data and simulated for 2012 and 2020 with Cellular Automata (CA) considering scenarios, change probabilities (through Markov chain) and Multi Criteria Evaluation (MCE). Agents and constraints were considered for modeling the urbanization process. Agents were normalized through fuzzyfication and priority weights were assigned through Analytical Hierarchical Process (AHP) pairwise comparison for each factor (in MCE) to derive behavior-oriented rules of transition for each land use class. Simulation shows a good agreement with the classified data. Fuzzy and AHP helped in analyzing the effects of agents of growth clearly and CA-Markov proved as a powerful tool in modelling and helped in capturing and visualizing the spatiotemporal patterns of urbanization. This provided rapid land evaluation framework with the essential insights of the urban trajectory for effective sustainable city planning.