• Home
  • Search
  • Instance-Level Scaling and Dynamic Margin-Alignment Knowledge Distillation for Remote Sensing Image Scene Classification
  • Cite Icon1
  • https://doi.org/10.3390/rs16203853Copy DOI Icon

Instance-Level Scaling and Dynamic Margin-Alignment Knowledge Distillation for Remote Sensing Image Scene Classification

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Remote sensing image (RSI) scene classification aims to identify semantic categories in RSI using neural networks. However, high-performance deep neural networks typically demand substantial storage and computational resources, making practical deployment challenging. Knowledge distillation has emerged as an effective technique for developing compact models that maintain high classification accuracy in RSI tasks. Existing knowledge distillation methods often overlook the high inter-class similarity in RSI scenes, leading to low-confidence soft labels from the teacher model, which can mislead the student model. Conversely, overly confident soft labels may discard valuable non-target information. Additionally, the significant intra-class variability in RSI contributes to instability in the model’s decision boundaries. To address these challenges, we propose an efficient method called instance-level scaling and dynamic margin-alignment knowledge distillation (ISDM) for RSI scene classification. To balance the target and non-target class influence, we apply an entropy regularization loss to scale the teacher model’s target class at the instance level. Moreover, we introduce dynamic margin alignment between the student and teacher models to improve the student’s discriminative capability. By optimizing soft labels and enhancing the student’s ability to distinguish between classes, our method reduces the effects of inter-class similarity and intra-class variability. Experimental results on three public RSI scene classification datasets (AID, UCMerced, and NWPU-RESISC) demonstrate that our method achieves state-of-the-art performance across all teacher–student pairs with lower computational costs. Additionally, we validate the generalization of our approach on general datasets, including CIFAR-100 and ImageNet-1k.

Similar Papers
  • Research Article
  • Citations19

Spatial Information Considered Network for Scene Classification

  • May 29, 2020
  • IEEE Geoscience and Remote Sensing Letters
  • Chao Tao +3
  • Research Article
  • Citations34

An Open Set Domain Adaptation Algorithm via Exploring Transferability and Discriminability for Remote Sensing Image Scene Classification

  • Jan 01, 2022
  • IEEE Transactions on Geoscience and Remote Sensing
  • Jun Zhang +5
  • Research Article

A Hierarchical ViT With Dynamic Window Shift Unit and Curriculum Learning for Remote Sensing Image Scene Classification

  • Jan 01, 2025
  • IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
  • Yi Liu +5
  • Conference Article
  • Citations6

Remote Scene Image Scene Classification Based on Adaptive Segmentation and Dynamic Graph Convolution

  • Jul 11, 2021
  • Yuqun Yang +5
  • PDF
  • Research Article
  • Citations21

Project Gradient Descent Adversarial Attack against Multisource Remote Sensing Image Scene Classification

  • Jun 12, 2021
  • Security and Communication Networks
  • Yan Jiang +3
  • Research Article
  • Citations105

Learning Deep Cross-Modal Embedding Networks for Zero-Shot Remote Sensing Image Scene Classification

  • Jan 14, 2021
  • IEEE Transactions on Geoscience and Remote Sensing
  • Yansheng Li +3
  • Research Article
  • Citations118

Vision Transformer: An Excellent Teacher for Guiding Small Networks in Remote Sensing Image Scene Classification

  • Jan 01, 2022
  • IEEE Transactions on Geoscience and Remote Sensing
  • Kejie Xu +2
  • PDF
  • Research Article
  • Citations51

Hierarchical Attention and Bilinear Fusion for Remote Sensing Image Scene Classification

  • Jan 01, 2020
  • IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
  • Donghang Yu +5
  • PDF
  • Research Article
  • Citations7

Evaluation of Ten Deep-Learning-Based Out-of-Distribution Detection Methods for Remote Sensing Image Scene Classification

  • Apr 24, 2024
  • Remote Sensing
  • Sicong Li +4
  • PDF
  • Research Article
  • Citations22

Attack Selectivity of Adversarial Examples in Remote Sensing Image Scene Classification

  • Jan 01, 2020
  • IEEE Access
  • Li Chen +7
  • Research Article
  • Citations34

LaST: Label-Free Self-Distillation Contrastive Learning With Transformer Architecture for Remote Sensing Image Scene Classification

  • Jan 01, 2022
  • IEEE Geoscience and Remote Sensing Letters
  • Xuying Wang +6
  • Research Article
  • Citations121

Remote Sensing Image Scene Classification With Self-Supervised Paradigm Under Limited Labeled Samples

  • Dec 21, 2020
  • IEEE Geoscience and Remote Sensing Letters
  • Chao Tao +4
  • Research Article
  • Citations47

Combining Multilevel Features for Remote Sensing Image Scene Classification With Attention Model

  • Nov 22, 2019
  • IEEE Geoscience and Remote Sensing Letters
  • Jinsheng Ji +4
  • PDF
  • Research Article
  • Citations15

Aggregating Features From Dual Paths for Remote Sensing Image Scene Classification

  • Jan 01, 2022
  • IEEE Access
  • Donghang Yu +5
  • Research Article
  • Citations33

A Lightweight and Multiscale Network for Remote Sensing Image Scene Classification

  • May 22, 2021
  • IEEE Geoscience and Remote Sensing Letters
  • Lin Bai +5
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.