Abstract:Artificial intelligence (AI) began to make its way into geoinformation science several decades ago and since then has constantly brought forth new cutting-edge approaches for diverse geographic use cases. AI and deep learning methods have become essential approaches for land use and land cover (LULC) classifications, which are important in urban planning and regional management. While the extraction of LULC information from multispectral satellite images has been a well-studied part of past and present research, only a few studies emerged about the recovery of spectral properties from LULC information. Estimates of the spectral characteristics of LULC categories could enrich LULC forecasting models by providing necessary information to delineate vegetation indices or microclimatic parameters. We train two identical Conditional Generative Adversarial Networks (CGAN) to synthesize a multispectral Sentinel-2 image based on different combinations of open-source LULC data sets. Large-scale synthetic multispectral satellite images of the administrative region of Bonn and Rhein-Sieg in Germany are generated with a Euclidean distance-based patch-fusion method. The approach generated a realistic-looking satellite image without noticeable seams between patch borders. Based on several metrics, such as difference calculations, the spectral information divergence (SID), and the Fréchet inception distance (FID), we evaluate the resulting images. The models reach mean SIDs as low as 0.026 for urban fabrics and forests and FIDs below 90 for bands B2 and B5 showing that the CGAN is capable of synthesizing distinct synthetic features matching with features typical for respective LULC categories and manages to mimic multispectral signatures. The method used in this paper to generate large-scale synthetic multispectral satellite images can be used as an approach to support scenario-oriented sustainable urban planning.

What problem does this paper attempt to address?

### What problems does this paper attempt to solve? This paper aims to explore how to use Conditional Generative Adversarial Networks (CGAN) to synthesize seamless, large - scale multispectral satellite images from open - source Land Use and Land Cover (LULC) data. Specifically, the research mainly focuses on the following aspects: 1. **Conversion of LULC information to multispectral images**: - Investigate whether open - source LULC data can be used as input to generate images with features corresponding to the underlying LULC categories through CGAN. 2. **Reliable synthesis of multispectral properties**: - Explore whether CGAN can reliably synthesize multispectral characteristics and ensure that the spectral characteristics of the generated images in different bands are similar to those of real images. 3. **Performance differences in different LULC categories**: - Analyze whether there are differences in the performance of CGAN in different LULC categories, such as urban structures, forests, etc. 4. **Performance differences in different image channels**: - Examine whether there are significant differences in the performance of CGAN in different image channels (such as B2 and B5 bands). ### Method overview To achieve the above goals, the authors adopted the following methods: - **Data preparation**: Use open - source LULC datasets from different federal states in western Germany as training data, and select data from Bonn and Rhein - Sieg regions for testing. - **Model architecture**: Adjust based on the Pix2Pix network architecture, including a generator with a U - NET structure and a discriminator with a Patch GAN structure. - **Post - processing**: Adopt a patch - fusion method based on Euclidean distance to generate seamless large - scale multispectral satellite images. - **Evaluation metrics**: Use multiple evaluation metrics, including Root Mean Square Error (RMSE), Spectral Information Divergence (SID), Fréchet Inception Distance (FID), etc., to evaluate the quality of the generated images. ### Conclusion Through experiments, the authors have proven that CGAN can successfully synthesize realistic multispectral satellite images from LULC data to a certain extent, especially performing well in some specific LULC categories and image channels. This provides new possibilities for future applications in urban planning and regional management.

Synthesis and evaluation of seamless, large-scale, multispectral satellite images using Generative Adversarial Networks on land use and land cover and Sentinel-2 data

Synthesizing Optical and SAR Imagery From Land Cover Maps and Auxiliary Raster Data

Generation of synthetic generative adversarial network-based multispectral satellite images with improved sharpness

Generating Synthetic Multispectral Satellite Imagery from Sentinel-2

Predicting Landscapes from Environmental Conditions Using Generative Networks

Semantic Segmentation of Medium-Resolution Satellite Imagery using Conditional Generative Adversarial Networks

Multispectral Band-Aware Generation of Satellite Images across Domains Using Generative Adversarial Networks and Contrastive Learning

Synthesis of Multispectral Optical Images From SAR/Optical Multitemporal Data Using Conditional Generative Adversarial Networks

Single Remote Sensing Image Super-Resolution Via a Generative Adversarial Network with Stratified Dense Sampling and Chain Training

Multispectral Satellite Image Generation Using StyleGAN3

Towards CGAN-based Satellite Image Synthesis with Partial Pixel-Wise Annotation

Advancements in urban scene segmentation using deep learning and generative adversarial networks for accurate satellite image analysis

Generating Synthetic Satellite Imagery for Rare Objects: An Empirical Comparison of Models and Metrics

Enhancing Sentinel-2 Image Resolution: Evaluating Advanced Techniques based on Convolutional and Generative Neural Networks

Attention GANs: Unsupervised Deep Feature Learning for Aerial Scene Classification.

Deep Snow: Synthesizing Remote Sensing Imagery with Generative Adversarial Nets

Conditional Generative Adversarial Networks for Data Augmentation and Adaptation in Remotely Sensed Imagery

Multi-Spectral Image Synthesis for Crop/Weed Segmentation in Precision Farming

Design of CGAN Models for Multispectral Reconstruction in Remote Sensing

SatSynth: Augmenting Image-Mask Pairs through Diffusion Models for Aerial Semantic Segmentation

Deep Learning for Feature-Level Data Fusion: Higher Resolution Reconstruction of Historical Landsat Archive