• Home
  • Search
  • Deep Learning Network Intensification for Preventing Noisy-Labeled Samples for Remote Sensing Classification
  • Open Access IconOpen Access
  • Cite Icon9
  • https://doi.org/10.3390/rs13091689Copy DOI Icon

Deep Learning Network Intensification for Preventing Noisy-Labeled Samples for Remote Sensing Classification

Show More
  • Abstract
  • Highlights & Summary
  • PDF
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

The deep-learning-network performance depends on the accuracy of the training samples. The training samples are commonly labeled by human visual investigation or inherited from historical land-cover or land-use maps, which usually contain label noise, depending on subjective knowledge and the time of the historical map. Helping the network to distinguish noisy labels during the training process is a prerequisite for applying the model for training across time and locations. This study proposes an antinoise framework, the Weight Loss Network (WLN), to achieve this goal. The WLN contains three main parts: (1) the segmentation subnetwork, which any state-of-the-art segmentation network can replace; (2) the attention subnetwork (λ); and (3) the class-balance coefficient (α). Four types of label noise (an insufficient label, redundant label, missing label and incorrect label) were simulated by dilate and erode processing to test the network’s antinoise ability. The segmentation task was set to extract buildings from the Inria Aerial Image Labeling Dataset, which includes Austin, Chicago, Kitsap County, Western Tyrol and Vienna. The network’s performance was evaluated by comparing it with the original U-Net model by adding noisy training samples with different noise rates and noise levels. The result shows that the proposed antinoise framework (WLN) can maintain high accuracy, while the accuracy of the U-Net model dropped. Specifically, after adding 50% of dilated-label samples at noise level 3, the U-Net model’s accuracy dropped by 12.7% for OA, 20.7% for the Mean Intersection over Union (MIOU) and 13.8% for Kappa scores. By contrast, the accuracy of the WLN dropped by 0.2% for OA, 0.3% for the MIOU and 0.8% for Kappa scores. For eroded-label samples at the same level, the accuracy of the U-Net model dropped by 8.4% for OA, 24.2% for the MIOU and 43.3% for Kappa scores, while the accuracy of the WLN dropped by 4.5% for OA, 4.7% for the MIOU and 0.5% for Kappa scores. This result shows that the antinoise framework proposed in this paper can help current segmentation models to avoid the impact of noisy training labels and has the potential to be trained by a larger remote sensing image set regardless of the inner label error.

Loading PDF

Similar Papers
  • Research Article
  • Citations702

Deep learning with noisy labels: Exploring techniques and remedies in medical image analysis.

  • Jun 20, 2020
  • Medical image analysis
  • Davood Karimi +3
  • Research Article
  • Citations29

BadLabel: A Robust Perspective on Evaluating and Enhancing Label-Noise Learning.

  • Jun 01, 2024
  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • Jingfeng Zhang +6
  • Conference Article
  • Citations173

Distilling Effective Supervision From Severe Label Noise

  • Jun 01, 2020
  • Zizhao Zhang +4
  • PDF
  • Research Article
  • Citations2

Uncertainty is not sufficient for identifying noisy labels in training data for binary segmentation of building footprints

  • Jan 10, 2023
  • Frontiers in Remote Sensing
  • Hannah Ulman +2
  • Research Article
  • Citations24

Two Wrongs Don’t Make a Right: Combating Confirmation Bias in Learning with Label Noise

  • Jun 26, 2023
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Mingcai Chen +5
  • Conference Article
  • Citations47

EvidentialMix: Learning with Combined Open-set and Closed-set Noisy Labels

  • Jan 01, 2021
  • Ragav Sachdeva +4
  • Research Article
  • Citations31

Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery

  • Dec 06, 2020
  • International Journal of Environment and Geoinformatics
  • Fırat Erdem +1
  • Research Article
  • Citations15

Multi-category classification with label noise by robust binary loss

  • Jan 19, 2022
  • Neurocomputing
  • Defu Liu +4
  • Research Article
  • Citations22

Using machine learning and satellite data from multiple sources to analyze mining, water management, and preservation of cultural heritage

  • Jul 15, 2023
  • Geo-spatial Information Science
  • Joaquim J Sousa +13
  • PDF
  • Research Article
  • Citations18

Evaluation of residual plastic film pollution in pre-sowing cotton field using UAV imaging and semantic segmentation.

  • Sep 09, 2022
  • Frontiers in Plant Science
  • Zhiqiang Zhai +6
  • Research Article
  • Citations70

Robust Medical Image Classification From Noisy Labeled Data With Global and Local Representation Guided Co-Training.

  • Jun 01, 2022
  • IEEE Transactions on Medical Imaging
  • Cheng Xue +4
  • PDF
  • Research Article
  • Citations16

Mapping Debris-Covered Glaciers Using High-Resolution Imagery (GF-2) and Deep Learning Algorithms

  • Jun 07, 2024
  • Remote Sensing
  • Xin Yang +9
  • Research Article
  • Citations5

Label noise detection under the noise at random model with ensemble filters

  • Sep 05, 2022
  • Intelligent Data Analysis
  • Kecia G Moura +2
  • Conference Article
  • Citations1

Ensemble Methods for Label Noise Detection Under the Noisy at Random Model

  • Oct 01, 2018
  • Kecia G De Moura +2
  • Research Article
  • Citations1

Graph Neural Networks with Coarse- and Fine-Grained Division for mitigating label noise and sparsity.

  • Jul 01, 2025
  • Neural networks : the official journal of the International Neural Network Society
  • Shuangjie Li +5
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.