ChessMix: Spatial Context Data Augmentation for Remote Sensing Semantic Segmentation

  title={ChessMix: Spatial Context Data Augmentation for Remote Sensing Semantic Segmentation},
  author={Matheus B. Pereira and Jefersson Alex dos Santos},
  journal={2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)},
  • M. B. Pereira, J. A. D. Santos
  • Published 26 August 2021
  • Computer Science, Environmental Science
  • 2021 34th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)
Labeling semantic segmentation datasets is a costly and laborious process if compared with tasks like image classification and object detection. This is especially true for remote sensing applications that not only work with extremely high spatial resolution data but also commonly require the knowledge of experts of the area to perform the manual labeling. Data augmentation techniques help to improve deep learning models under the circumstance of few and imbalanced labeled samples. In this work… 
1 Citations

Figures and Tables from this paper

SuperpixelGridCut, SuperpixelGridMean and SuperpixelGridMix Data Augmentation
A novel approach of data augmentation based on irregular superpixel decomposition is proposed. This approach called SuperpixelGridMasks permits to extend original image datasets that are required by


ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data
Class Attention Network for Semantic Segmentation of Remote Sensing Images
  • Zhibo Rao, Mingyi He, Yuchao Dai
  • Computer Science, Environmental Science
    2020 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)
  • 2020
This paper proposes a novel class attention module and decomposition-fusion strategy to cope with imbalanced labels in remote sensing images, and introduces salient detection, which decomposes semantic segmentation into multi-class salient detection and then fuses them to produce a segmentation map.
Efficient Patch-Wise Semantic Segmentation for Large-Scale Remote Sensing Images
A novel patch-wise semantic segmentation method with a new training strategy based on fully convolutional networks is presented to segment common land resources and outperforms other state-of-the-art methods in remote sensing image segmentation.
Dynamic Multicontext Segmentation of Remote Sensing Images Based on Convolutional Networks
A novel technique to perform semantic segmentation of remote sensing images that exploits a multicontext paradigm without increasing the number of parameters while defining, in training time, the best patch size is proposed.
Deep Learning for Multilabel Land Cover Scene Categorization Using Data Augmentation
A data augmentation technique that can drastically increase the size of a smaller data set to copious amounts is proposed that outperforms the current state of the art by more than 6% in terms of the F-score metric.
Fully Convolutional Networks for Dense Semantic Labelling of High-Resolution Aerial Imagery
  • J. Sherrah
  • Environmental Science, Computer Science
  • 2016
To make better use of image features, a pre-trained CNN is fine-tuned on remote sensing data in a hybrid network context, resulting in superior results compared to a network trained from scratch.
Segmentation of Imbalanced Classes in Satellite Imagery using Adaptive Uncertainty Weighted Class Loss
A loss function which leverages the per class uncertainty of the model during training together with median frequency balancing of the class pixels is introduced for the training of deep Convolutional Neural Networks focusing on land use and land cover classification in remote sensed data.
GridMask Data Augmentation
This paper proposes a novel data augmentation method `GridMask', which is based on the deletion of regions of the input image and outperforms the latest AutoAugment, which is way more computationally expensive due to the use of reinforcement learning to find the best policies.
SiftingGAN: Generating and Sifting Labeled Samples to Improve the Remote Sensing Image Scene Classification Baseline In Vitro
Experiments on the well-known aerial image data set demonstrate that the proposed SiftingGAN method can not only effectively improve the performance of the scene classification baseline that is achieved without data augmentation but also significantly excels the comparison methods based on traditional geometric/radiometric transformation operations.
Perspective Transformation Data Augmentation for Object Detection
A novel data augmentation framework based on perspective transformation that automatically generates new annotated data without extra manual labeling, thus effectively extends the inadequate dataset.