• Home
  • Search
  • Semi-Supervised Deep Learning Using Pseudo Labels for Hyperspectral Image Classification.
  • Cite Icon490
  • https://doi.org/10.1109/tip.2017.2772836Copy DOI Icon

Semi-Supervised Deep Learning Using Pseudo Labels for Hyperspectral Image Classification.

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

Deep learning has gained popularity in a variety of computer vision tasks. Recently, it has also been successfully applied for hyperspectral image classification tasks. Training deep neural networks, such as a convolutional neural network for classification requires a large number of labeled samples. However, in remote sensing applications, we usually only have a small amount of labeled data for training because they are expensive to collect, although we still have abundant unlabeled data. In this paper, we propose semi-supervised deep learning for hyperspectral image classification-our approach uses limited labeled data and abundant unlabeled data to train a deep neural network. More specifically, we use deep convolutional recurrent neural networks (CRNN) for hyperspectral image classification by treating each hyperspectral pixel as a spectral sequence. In the proposed semi-supervised learning framework, the abundant unlabeled data are utilized with their pseudo labels (cluster labels). We propose to use all the training data together with their pseudo labels to pre-train a deep CRNN, and then fine-tune using the limited available labeled data. Further, to utilize spatial information in the hyperspectral images, we propose a constrained Dirichlet process mixture model (C-DPMM), a non-parametric Bayesian clustering algorithm, for semi-supervised clustering which includes pairwise must-link and cannot-link constraints-this produces high-quality pseudo-labels, resulting in improved initialization of the deep neural network. We also derived a variational inference model for the C-DPMM for efficient inference. Experimental results with real hyperspectral image data sets demonstrate that the proposed semi-supervised method outperforms state-of-the-art supervised and semi-supervised learning methods for hyperspectral classification.

Similar Papers
  • Research Article
  • Citations39

Predicting the non-linear response of composite materials using deep recurrent convolutional neural networks

  • May 23, 2023
  • International Journal of Solids and Structures
  • Bassam El Said
  • Conference Article
  • Citations17

Deep Learning Based Emotion Classification Using Mel Frequency Magnitude Coefficient

  • Mar 04, 2023
  • Siba Prasad Mishra +2
  • Research Article
  • Citations59

DnRCNN: Deep Recurrent Convolutional Neural Network for HSI Destriping.

  • Jul 01, 2023
  • IEEE transactions on neural networks and learning systems
  • Juntao Guan +4
  • Research Article
  • Citations28

Glaucoma Expert-Level Detection of Angle Closure in Goniophotographs With Convolutional Neural Networks: The Chinese American Eye Study

  • Feb 09, 2021
  • American Journal of Ophthalmology
  • Michael Chiang +13
  • Research Article

Brain Tumor Classification Using Relief Algorithm Based Convolutional Neural Network

  • Apr 24, 2021
  • Turkish Journal of Computer and Mathematics Education (TURCOMAT)
  • R Karthik
  • PDF
  • Research Article
  • Citations25

Divide-and-Conquer Dual-Architecture Convolutional Neural Network for Classification of Hyperspectral Images

  • Feb 27, 2019
  • Remote Sensing
  • Jie Feng +4
  • Research Article
  • Citations50

Deep convolutional recurrent neural network with transfer learning for hyperspectral image classification

  • Jun 11, 2018
  • Journal of Applied Remote Sensing
  • Bing Liu +3
  • PDF
  • Research Article
  • Citations65

Semi-Supervised Deep Learning Classification for Hyperspectral Image Based on Dual-Strategy Sample Selection

  • Apr 08, 2018
  • Remote Sensing
  • Bei Fang +3
  • Research Article
  • Citations7

A plant disease image using convolutional recurrent neural network procedure intended for big data plant classification

  • Aug 10, 2022
  • Journal of Intelligent & Fuzzy Systems
  • S Gopinath +2
  • Research Article
  • Citations45

Deep Feature Alignment Neural Networks for Domain Adaptation of Hyperspectral Data

  • Oct 01, 2018
  • IEEE Transactions on Geoscience and Remote Sensing
  • Xiong Zhou +1
  • Conference Article
  • Citations1053

DeepVO: Towards end-to-end visual odometry with deep Recurrent Convolutional Neural Networks

  • May 01, 2017
  • Sen Wang +3
  • Book Chapter

Accurate Detection of Brain Tumor Using Compound Filter and Deep Neural Network

  • Jan 01, 2023
  • Praveen Kumar Ramtekkar +2
  • Book Chapter
  • Citations1

Real Time 3D Pose Estimation of Both Human Hands via RGB-Depth Camera and Deep Convolutional Neural Networks

  • Jun 06, 2019
  • Geon Gi +6
  • Research Article
  • Citations228

Using Deep and Convolutional Neural Networks for Accurate Emotion Classification on DEAP Data

  • Feb 11, 2017
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Samarth Tripathi +4
  • Research Article
  • Citations21

Multi‐channel Sliced Deep RCNN with Residual Network for Text Classification

  • Sep 01, 2020
  • Chinese Journal of Electronics
  • Chuanhua Zhou +4
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.