Abstract:Land cover classification is one of the most fundamental tasks in the field of remote sensing. In recent years, fully supervised fully convolutional network (FCN)-based semantic segmentation models have achieved state-of-the-art performance in the semantic segmentation task. However, creating pixel-level annotations is prohibitively expensive and laborious, especially when dealing with remote sensing images. Weakly supervised learning methods from weakly labeled annotations can overcome this difficulty to some extent and achieve impressive segmentation results, but results are limited in accuracy. Inspired by point supervision and the traditional segmentation method of seeded region growing (SRG) algorithm, a weakly towards strongly (WTS) supervised learning framework is proposed in this study for remote sensing land cover classification to handle the absence of well-labeled and abundant pixel-level annotations when using segmentation models. In this framework, only several points with true class labels are required as the training set, which are much less expensive to acquire compared with pixel-level annotations through field survey or visual interpretation using high-resolution images. Firstly, they are used to train a Support Vector Machine (SVM) classifier. Once fully trained, the SVM is used to generate the initial seeded pixel-level training set, in which only the pixels with high confidence are assigned with class labels whereas others are unlabeled. They are used to weakly train the segmentation model. Then, the seeded region growing module and fully connected Conditional Random Fields (CRFs) are used to iteratively update the seeded pixel-level training set for progressively increasing pixel-level supervision of the segmentation model. Sentinel-2 remote sensing images are used to validate the proposed framework, and SVM is selected for comparison. In addition, FROM-GLC10 global land cover map is used as training reference to directly train the segmentation model. Experimental results show that the proposed framework outperforms other methods and can be highly recommended for land cover classification tasks when the pixel-level labeled datasets are insufficient by using segmentation models.

Weakly Supervised Learning for Target Detection in Remote Sensing Images

Weakly Supervised Learning for Airplane Detection in Remote Sensing Images

Object Detection in Optical Remote Sensing Images Based on Weakly Supervised Learning and High-Level Feature Learning

Automatic Weakly Supervised Object Detection from High Spatial Resolution Remote Sensing Images Via Dynamic Curriculum Learning

Negative Bootstrapping for Weakly Supervised Target Detection in Remote Sensing Images

Remote Sensing Object Detection in Disturbed Environment Based on Improved RTMDet Network

Point-Based Weakly Supervised Learning for Object Detection in High Spatial Resolution Remote Sensing Images

MOL: Towards Accurate Weakly Supervised Remote Sensing Object Detection Via Multi-view Noisy Learning

Self-Training-Based Semantic-Balanced Network for Weakly Supervised Object Detection in Remote-Sensing Images

WTS: A Weakly towards Strongly Supervised Learning Framework for Remote Sensing Land Cover Classification Using Segmentation Models

Weakly Supervised Few-Shot Object Detection with DETR

Deep Active Learning for Remote Sensing Object Detection

Weakly-semi-supervised object detection in remotely sensed imagery

Few-Shot Object Detection of Remote Sensing Images via Two-Stage Fine-Tuning

Automatic Labelling Framework for Optical Remote Sensing Object Detection Samples in a Wide Area Using Deep Learning

A Training-free, One-shot Detection Framework For Geospatial Objects In Remote Sensing Images

Hierarchical fusion and divergent activation based weakly supervised learning for object detection from remote sensing images

Target Discrimination Based on Weakly Supervised Learning for High-Resolution SAR Images in Complex Scenes

Multiscale Image Splitting Based Feature Enhancement and Instance Difficulty Aware Training for Weakly Supervised Object Detection in Remote Sensing Images

Debiased Learning for Remote Sensing Data

Weakly Supervised Object Detection for Remote Sensing Images: A Survey