• Home
  • Search
  • Classification of Remote Sensing scenes using Semi-Supervised Domain Adaptation based on Entropy Adversarial Optimization
  • Cite Icon1
  • https://doi.org/10.1145/3454127.3456610Copy DOI Icon

Classification of Remote Sensing scenes using Semi-Supervised Domain Adaptation based on Entropy Adversarial Optimization

  • Apr 1, 2021
  • Tariq Lasloum +2 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

In this paper, we present a new method for semi-supervised domain adaptation in remote sensing scene classification. The method is based on a pre-trained Convolutional Neural Network (CNN) model for the extraction of highly discriminative features, followed by a fully connected layer with softmax activation function that is responsible for the classification task. The weights of the fully connected layer represent prototype feature vectors for each class. These weights are divide by a temperature parameter for normalization. The whole network is trained on both the labeled and unlabeled target samples. First, the whole network is trained on the labeled source and target samples using the standard cross entropy loss to predict their correct classes. At the same time, the model is trained to learn domain invariant features using another loss function based on entropy computed over the unlabeled target samples. Unlike the standard cross entropy loss, the novel entropy loss function is computed on the predicted probabilities of the model and does not need the true labels. The proposed model combines the standard cross entropy loss and the new unlabeled samples entropy loss and optimizes them jointly. However, the new entropy loss function needs to be maximized with respect to the classification layer to learn features that are domain invariant (hence removing the data shift), and at the same time, it should be minimized with respect to the CNN feature extractor to learn discriminative feature that are clustered around the class prototypes (in other words reducing intra-class variance). To accomplish this maximization and minimization processes at the same time, we use an adversarial training approach, where we alternate between the two processes. This type of approach is called minmax entropy and the new proposed method is called Domain Adaptation CNN with MinMax Entropy (DACNN-MME). The proposed method is tested on three RS scene datasets, namely UC Merced, AID, and NWPU. The preliminary experimental results demonstrate the potential of the proposed method. Its performance is already better than several state-of-the-art methods including RevGard, ADDA, Siamese-GAN, and MSCN. With more analysis and fine-tuning of the method even better results can be achieved in the future.

Similar Papers
  • PDF
  • Research Article
  • Citations39

SSDAN: Multi-Source Semi-Supervised Domain Adaptation Network for Remote Sensing Scene Classification

  • Sep 27, 2021
  • Remote Sensing
  • Tariq Lasloum +3
  • Conference Article
  • Citations2

Towards Realizing the Value of Labeled Target Samples: A Two-Stage Approach for Semi-Supervised Domain Adaptation

  • Jun 04, 2023
  • Mengqun Jin +4
  • Research Article
  • Citations183

Self-Attention-Based Deep Feature Fusion for Remote Sensing Scene Classification

  • Feb 11, 2020
  • IEEE Geoscience and Remote Sensing Letters
  • Ran Cao +3
  • Research Article
  • Citations1

RDPN: Tackling the Data Shift Problem in Remote Sensing Scene Classification

  • Jan 01, 2022
  • IEEE Geoscience and Remote Sensing Letters
  • Xiang Zhang +4
  • PDF
  • Research Article
  • Citations18

A Deformable Convolutional Neural Network with Spatial-Channel Attention for Remote Sensing Scene Classification

  • Dec 14, 2021
  • Remote Sensing
  • Di Wang +1
  • Book Chapter
  • Citations4

Convolutional Neural Network with Discriminant Criterion for Input of Each Neuron in Output Layer

  • Jan 01, 2018
  • Hidenori Ide +1
  • Research Article
  • Citations2

Convolutional neural network for sapphire ingots defect detection and classification

  • Sep 01, 2021
  • Optical Materials
  • Euphrem Mugisha Rwagasore +9
  • Research Article
  • Citations48

Searching for CNN Architectures for Remote Sensing Scene Classification

  • Jan 01, 2022
  • IEEE Transactions on Geoscience and Remote Sensing
  • Clifford Broni-Bediako +3
  • Research Article
  • Citations212

A Feature Aggregation Convolutional Neural Network for Remote Sensing Scene Classification

  • Oct 01, 2019
  • IEEE Transactions on Geoscience and Remote Sensing
  • Xiaoqiang Lu +2
  • PDF
  • Research Article
  • Citations68

Homo–Heterogenous Transformer Learning Framework for RS Scene Classification

  • Jan 01, 2022
  • IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
  • Jingjing Ma +5
  • Research Article
  • Citations137

EMTCAL: Efficient Multiscale Transformer and Cross-Level Attention Learning for Remote Sensing Scene Classification

  • Jan 01, 2022
  • IEEE Transactions on Geoscience and Remote Sensing
  • Xu Tang +5
  • Conference Article
  • Citations137

Cross-Domain Adaptive Clustering for Semi-Supervised Domain Adaptation

  • Jun 01, 2021
  • Jichang Li +3
  • Research Article
  • Citations16

Performance Improvement Of Pre-trained Convolutional Neural Networks For Action Recognition

  • Jun 15, 2020
  • The Computer Journal
  • Tayyip Ozcan +1
  • Research Article
  • Citations31

Convolutional neural network based heterogeneous transfer learning for remote-sensing scene classification

  • May 15, 2019
  • International Journal of Remote Sensing
  • Huizhen Zhao +3
  • Conference Article
  • Citations4

Resformer: Bridging Residual Network and Transformer for Remote Sensing Scene Classification

  • Jul 17, 2022
  • Mingteng Li +5
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.