Chaudhary et al. (2025) Comprehensive Analysis of State-of-the-Art Deep Learning-Based Image Fusion Systems in the Context of Remote Sensing Images

Identification

• Journal: Lecture notes in networks and systems
• Year: 2025
• Date: 2025-10-19
• Authors: Anita Chaudhary, Navdeep Kaur, Bal Ram Singh, Steve Gill, Navjot Singh Talwandi
• DOI: 10.1007/978-3-032-04222-4_29

Research Groups

• Department of Computer Science, Sri Guru Granth Sahib World University, Fatehgarh Sahib, Punjab, India
• Department of Zoology, Akal College of Basic Sciences, Eternal University, Baru Sahib, Himachal Pradesh, India
• Department of Computer Science and Engineering AIT-CSE, Chandigarh University, Gharuan, Punjab, India

Short Summary

This paper provides a comprehensive review and analysis of state-of-the-art deep learning-based image fusion systems specifically applied to remote sensing images, aiming to enhance image quality by combining spatial, spectral, and temporal information. It summarizes reported deep learning techniques used across various remote sensing datasets to improve image detail and quality through fusion.

Objective

• To summarize and thoroughly study the most advanced deep learning techniques reported for image fusion in remote sensing, with the goal of improving image quality and detail.

Study Configuration

• Spatial Scale: Review of techniques applicable to various remote sensing images, covering diverse geographical areas and resolutions depending on the datasets analyzed in the reviewed literature.
• Temporal Scale: Review of existing deep learning methodologies for image fusion, not a temporal study of a specific phenomenon.

Methodology and Data

• Models used: Deep learning-based image fusion systems (e.g., Convolutional Neural Networks, multimodal deep networks, self-supervised networks, fusion transformers, etc., as referenced in the literature reviewed).
• Data sources: Various remote sensing image datasets (e.g., Sentinel-1, Sentinel-2, hyperspectral, multispectral, LiDAR data, multimodal imagery) as utilized by the deep learning techniques surveyed.

Main Results

• The study presents a thorough analysis of current deep learning-based image fusion solutions, highlighting their efficacy in enhancing the quality of remote sensing images.
• It demonstrates that deep learning approaches offer a significant advantage over conventional methods by continuously adjusting and optimizing fused image quality through training processes.
• The paper serves as a valuable resource for professionals working on visual challenges in remote sensing applications by consolidating information on advanced fusion techniques.

Contributions

• Provides a comprehensive, state-of-the-art overview and analysis of deep learning-based image fusion systems specifically tailored for remote sensing images.
• Consolidates knowledge on how deep learning addresses the challenge of integrating spatial, spectral, and temporal information for improved image quality.
• Offers a foundational reference for researchers and practitioners seeking to apply or develop advanced image fusion techniques in remote sensing.

Funding

• Not explicitly mentioned in the provided text.

Citation

@article{Chaudhary2025Comprehensive,
  author = {Chaudhary, Anita and Kaur, Navdeep and Singh, Bal Ram and Gill, Steve and Talwandi, Navjot Singh},
  title = {Comprehensive Analysis of State-of-the-Art Deep Learning-Based Image Fusion Systems in the Context of Remote Sensing Images},
  journal = {Lecture notes in networks and systems},
  year = {2025},
  doi = {10.1007/978-3-032-04222-4_29},
  url = {https://doi.org/10.1007/978-3-032-04222-4_29}
}