Scholarly open access journals, Peer-reviewed, and Refereed Journals, Impact factor 8.14 (Calculate by google scholar and Semantic Scholar | AI-Powered Research Tool) , Multidisciplinary, Monthly, Indexing in all major database & Metadata, Citation Generator, Digital Object Identifier(DOI)
High-resolution RS images play a crucial role in providing a sound geospatial analysis in its use in areas like urban planning, environmental surveillance and land-cover mapping. But it is still difficult to semantically segment such images because of the high level of spatial complexity, botanic textures, and poor models of long range contexts of traditional convolution-related networks. The paper presents Hybrid Spectral-Context Fusion Framework, which is a combination of Frequency Adaptive State Space Attention (FASSA) and Edge Calibrated Multiscale Orientation Filtering (ECMOF) in order to enhance the performance of segmentation. The suggested Vaibhav Malhotra Department of Computer Science and Engineering SRM Institute of Science and Technology, Inida vm3820@srmist.edu.in A.Thiruneelakandan Assistantt Professor, Department of Computer Science and Engineering SRM Institute of Science and Technology, Inida thirunea@srmist.edu.in that can then be interpreted into a detailed representation of complex land surface. Proper segmentation allows the generation of geographic meaningful information automatically, which helps the decision-maker in developing infrastructure facilities, protecting the environment, and risks management. Nevertheless, the spatial resolution of remote sensing images is also increasing thus causing new computational and modeling issues because of the occurrence of complex textures, varying sizes of objects and terribly heterogeneous landscapes. architecture integrates spatial domain convolution, spectral domain representation as well as effective global reasoning using state-space modeling. A Boundary-Aware Contrastive Refinement Module is proposed to further improve the dissimilarity between classes along the object edges, whereas a Dual-Domain Residual Fusion Strategy allows custom responsiveness between spectral and spatial abilities. Experimental test of the high-resolution remote sensing data show better segmentation accuracy, promotional boundary delineation, as well as strong functionality in obtruse terrain and luminance conditioning.
Keywords:
Remote Sensing Semantic Segmentation, Hybrid Spectral Context Fusion, State Space Attention, Frequency Domain Representation, Boundary Aware Feature Learning, Multiscale Orientation Filtering, High Resolution Geospatial Analysis.
Cite Article:
"GF-DeepSeg: A Gabor–Fourier Integrated Hybrid Deep Learning Model for Precise Remote Sensing Image Segmentation ", International Journal for Research Trends and Innovation (www.ijrti.org), ISSN:2456-3315, Vol.11, Issue 4, page no.b639-b645, April-2026, Available :http://www.ijrti.org/papers/IJRTI2604223.pdf
Downloads:
000205505
ISSN:
2456-3315 | IMPACT FACTOR: 8.14 Calculated By Google Scholar| ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 8.14 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator