Abstract:In this paper, we focus on tackling the problem of automatic semantic annotation of high resolution (HR) optical satellite images, which aims to assign one or several predefined semantic concepts to an image according to its content. The main challenges arise from the difficulty of characterizing complex and ambiguous contents of the satellite images and the high human labor cost caused by preparing a large amount of training examples with high-quality pixel-level labels in fully supervised annotation methods. To address these challenges, we propose a unified annotation framework by combining discriminative high-level feature learning and weakly supervised feature transferring. Specifically, an efficient stacked discriminative sparse autoencoder (SDSAE) is first proposed to learn high-level features on an auxiliary satellite image data set for the land-use classification task. Inspired by the motivation that the encoder of the prelearned SDSAE can be regarded as a generic high-level feature extractor for HR optical satellite images, we then transfer the learned high-level features to semantic annotation. To compensate the difference between the auxiliary data set and the annotation data set, the transferred high-level features are further fine-tuned in a weakly supervised scheme by using the tile-level annotated training data. Finally, the fine-tuning process is formulated as an ultimate optimization problem, which can be solved efficiently with our proposed alternate iterative optimization method. Comprehensive experiments on a publicly available land-use classification data set and an annotation data set demonstrate the superiority of our SDSAE-based high-level feature learning method and the effectiveness of our weakly supervised semantic annotation framework compared with state-of-the-art fully supervised annotation methods.

Satellite and Instrument Entity Recognition Using a Pre-Trained Language Model with Distant Supervision.

Named entity recognition based on semi-supervised ensemble learning with the improved tri-training algorithm

Extraction of Traditional Chinese Medicine Entity: Design of a Novel Span-Level Named Entity Recognition Method With Distant Supervision

Distantly-Supervised Named Entity Recognition with Adaptive Teacher Learning and Fine-grained Student Ensemble

Distantly Supervised NER with Partial Annotation Learning and Reinforcement Learning.

Distantly Supervised Named Entity Recognition Using Positive-Unlabeled Learning.

Joint Representation Learning of Cross-lingual Words and Entities via Attentive Distant Supervision

Using Domain Knowledge for Low Resource Named Entity Recognition

Semantic Annotation of High-Resolution Satellite Images Via Weakly Supervised Learning.

Debiased and Denoised Entity Recognition from Distant Supervision.

Towards Automatic Satellite Images Captions Generation Using Large Language Models

Knowledge-Augmented Language Model and its Application to Unsupervised Named-Entity Recognition

Label Refinement Via Contrastive Learning for Distantly-Supervised Named Entity Recognition.

Remote Sensing Vision-Language Foundation Models without Annotations via Ground Remote Alignment

Method and Dataset Entity Mining in Scientific Literature: A CNN + Bi-LSTM Model with Self-attention

A Large Model Assisted Remote Sensing Image Scene Understanding Algorithm Based on Object Detection

Distantly Supervised Entity Relation Extraction with Adapted Manual Annotations

Named entity recognition based on a machine learning model

A Semi-Supervised Deep Rule-Based Approach for Complex Satellite Sensor Image Analysis

Satellite Captioning: Large Language Models to Augment Labeling

An Internet Resource Extracting Relation Model based on Distant Supervision and Deep Learning