Shaziya Noorien, Syeda Muskaan Begum, Syeda Zehra Fatima Razvi, Ms. Sadaf Jahan 02 04 To address the challenges of fewshot aerial image semantic segmentation where unseencategory objects in query aerial images need to be parsed with only a few annotated support images we propose a novel approach by integrating a UNet architecture with Efficient Net Typically in fewshot segmentation category prototypes are extracted from support samples to segment query images in a pixelwise matching process However the arbitrary orientations and distribution of aerial objects in such images often result in significant feature variations making conventional methods which do not account for orientation changes ineffective The rotation sensitivity of aerial images causes substantial feature distortions leading to low confidence scores and misclassification of samecategory objects with different orientations To overcome these limitations we propose an enhanced solution by combining UNet for robust semantic segmentation with Efficient Net for efficient feature extraction and scale adaptability This architecture which we refer to as Efficient UNet introduces rotationinvariant feature extraction to handle the varying orientations of aerial objects By leveraging Efficient Nets scalable Convolutional layers for feature extraction we ensure that the network can capture orientationvarying yet categoryconsistent information from support images This approach enhances the segmentation accuracy by aligning samecategory objects irrespective of their orientation thereby minimizing the oscillation of confidence scores and improving the detection of rotated semantic objects This Efficient UNet model provides a scalable rotationinvariant solution to the fewshot segmentation of aerial images Copyright (c) Journal of Science Engineering Technology and Management Science. All rights reserved 2026