A survey of active learning algorithms for supervised remote sensing image classification

Abstract

Defining an efficient training set is one of the most delicate phases for the success of remote sensing image classification routines. The complexity of the problem, the limited temporal and financial resources, as well as the high intraclass variance can make an algorithm fail if it is trained with a suboptimal dataset. Active learning aims at building efficient training sets by iteratively improving the model performance through sampling. A user-defined heuristic ranks the unlabeled pixels according to a function of the uncertainty of their class membership and then the user is asked to provide labels for the most uncertain pixels. This paper reviews and tests the main families of active learning algorithms: committee, large margin and posterior probability-based. For each of them, the most recent advances in the remote sensing community are discussed and some heuristics are detailed and tested. Several challenging remote sensing scenarios are considered, including very high spatial resolution and hyperspectral image classification. Finally, guidelines for choosing the good architecture are provided for new and/or unexperienced user.

References in corpus (2)

• Active Learning with Multiple Views
• Active Learning with Statistical Models

Cited by in corpus (23)

• AID: A Benchmark Dataset for Performance Evaluation of Aerial Scene Classification
• Remote Sensing Image Scene Classification Meets Deep Learning: Challenges, Methods, Benchmarks, and Opportunities
• Towards Better Exploiting Convolutional Neural Networks for Remote Sensing Scene Classification
• Advances in Hyperspectral Image Classification: Earth monitoring with statistical learning methods
• Recent Advances in Domain Adaptation for the Classification of Remote Sensing Data
• Galaxy Zoo: Probabilistic Morphology through Bayesian CNNs and Active Learning
• Detecting animals in African Savanna with UAVs and the crowds
• Half a Percent of Labels is Enough: Efficient Animal Detection in UAV Imagery using Deep CNNs and Active Learning
• Semisupervised Manifold Alignment of Multimodal Remote Sensing Images
• Active Learning Methods for Efficient Hybrid Biophysical Variable Retrieval
• Optical Remote Sensing Image Understanding with Weak Supervision: Concepts, Methods, and Perspectives
• Incremental Import Vector Machines for Classifying Hyperspectral Data
• Integration of LiDAR and multispectral images for exposure and earthquake vulnerability estimation. Application in Lorca, Spain
• Active Ensemble Deep Learning for Polarimetric Synthetic Aperture Radar Image Classification
• Learning User's confidence for active learning
• Retrieving Quantum Information with Active Learning
• Improving Interpretability of Deep Active Learning for Flood Inundation Mapping Through Class Ambiguity Indices Using Multi-spectral Satellite Imagery
• Better, Not Just More: Data-Centric Machine Learning for Earth Observation
• Soft Computing Techniques for Change Detection in remotely sensed images : A Review
• On Creating Benchmark Dataset for Aerial Image Interpretation: Reviews, Guidances and Million-AID
• Minimum-Margin Active Learning
• A Survey on Spatial and Spatiotemporal Prediction Methods
• Active Learning in Physics: From 101, to Progress, and Perspective