IRMA-International.org: Creator of Knowledge
Information Resources Management Association
Advancing the Concepts & Practices of Information Resources Management in Modern Organizations

Develop a Hybrid Ensemble Transfer-Based Residual Multi-Resolution CNN for Classification of Land Cover in Remote Sensing Images

Develop a Hybrid Ensemble Transfer-Based Residual Multi-Resolution CNN for Classification of Land Cover in Remote Sensing Images
View Sample PDF
Author(s): M. Suresh Anand (Department of Computing Technologies, SRM Institute of Science and Technology, Kattankulathur, India), R. Anto Arockia Rosaline (Department of Computing Technologies, SRM Institute of Science and Technology, Kattankulathur, India), G. Padmapriya (Department of Computing Technologies, SRM Institute of Science and Technology, Kattankulathur, India), Prithi Samuel (Department of Computational Intelligence, SRM Institute of Science and Technology, Kattankulathur, India)and P. Kirubanantham (Department of Computing Technologies, SRM Institute of Science and Technology, Kattankulathur, India)
Copyright: 2025
Pages: 26
Source title: Harnessing AI in Geospatial Technology for Environmental Monitoring and Management
Source Author(s)/Editor(s): Froilan D. Mobo (Philippine Merchant Marine Academy, Philippines)
DOI: 10.4018/979-8-3693-8104-5.ch005

Purchase


Abstract

A reliable field covering identification from distant sensed Images is important for numerous programs, like ecological surveillance, city creation, and asset administration. Conventional approaches typically fell low in managing the intricacy and variety of distant sensed information. Present a novel method to enhance the robustness and accuracy of land cover classification through ensemble transfer-based Residual Multi-Resolution Convolutional Neural Network (RMRCNN) models. The proposed system uses the residual and multi-resolution architecture of multiple pre-trained RMRCNNs to capture multi-scale features. We train these neural networks on targeted satellite imagery sets using transfer learning, which enables the algorithms to leverage pre-learned characteristics from massive Image libraries. Using the complementing strengths of each RMRCNN and reducing its drawbacks, the ensemble method combines the output of different RMRCNNs to improve classification performance. Large tests were conducted on renowned distant recognizing information sets to assess the proposed method. This leads to suggest that the group transfer-based RMRCNN algorithms significantly surpass conventional single-model addresses and other developed methods, accomplishing greater categorization reliability, exactness, and remember. The excellent applicability of the proposed strategy over various kinds of land cover and geographical regions highlights its capacity for expansion and resilience. This study offers a viable path for further research in this area by demonstrating the effectiveness of ensemble transfer learning techniques with RMRCNN algorithms for improving surface area categorization in satellite imagery.

Related Content

Frederic Andres. © 2027. 14 pages.
Kalsoom Safdar, Khairul Najmy Abdul Rani, Mohd Aminudin Jamlos, Siti Julia Rosli, Muhammad Usman Younus, Zanab Safdar. © 2027. 27 pages.
Bani Adam, Binastya Anggara Sekti, Muhammad Adi Zacky Zahran. © 2027. 24 pages.
Swetha Margaret T. A., Renuka Devi D.. © 2027. 31 pages.
Maurice Saluschke, Michael Schulz. © 2027. 30 pages.
Mirjam Sepesy Maučec, Gregor Donaj. © 2027. 16 pages.
Jorge A. Ruiz-Vanoye, Ocotlan Diaz-Parra, Ricardo A. Barrera-Cámara, Alejandro Fuentes-Penna, Francisco R. Trejo-Macotela, Jaime Aguilar-Ortiz, Eric Simancas-Acevedo. © 2027. 21 pages.
Body Bottom