Abstract:Because of its wide potential applications, remote sensing (RS) image semantic segmentation has attracted increasing research interest in recent years. Until now, deep semantic segmentation network (DSSN) has achieved a certain degree of success on semantic segmentation of RS imagery and can obviously outperform the traditional methods based on hand-crafted features. As a classic data-driven technique, DSSN can be trained by an end-to-end mechanism and is competent for employing low-level and mid-level cues (i.e., the discriminative image structure) to understand RS images. However, its interpretability and reliability are poor due to the nature weakness of the data-driven deep learning methods. By contrast, human beings have an excellent inference capacity and can reliably interpret RS imagery with the basic RS domain knowledge. Ontological reasoning is an ideal way to imitate and employ the domain knowledge of human beings. However, it is still rarely explored and adopted in the RS domain. As a solution of the aforementioned critical limitation of DSSN, this study proposes a collaboratively boosting framework (CBF) to combine the data-driven deep learning module and knowledge-guided ontology reasoning module in an iterative manner. The deep learning module adopts the DSSN architecture and takes the integration of the original image and inferred channels as the input of the DSSN. In addition, the ontology reasoning module is composed of intra- and extra-taxonomy reasoning. More specifically, the intra-taxonomy reasoning directly corrects misclassifications of the deep learning module based on the domain knowledge, which is the key to improve the classification performance. The extra-taxonomy reasoning aims to generate the inferred channels beyond the current taxonomy to improve the discriminative performance of DSSN in the original RS image space. On the one hand, benefiting from the referred channels from the ontology reasoning module, the deep learning module using the integration of the original image and referred channels can achieve better classification performance than only using the original image. On the other hand, better classification results from the deep learning module further improve the performance of the ontology reasoning module. As a whole, the deep learning and ontology reasoning modules are mutually boosted in the iterations. Extensive experiments on two publicly open RS datasets such as UCM and ISPRS Potsdam show that our proposed CBF can outperform the competitive baselines with a large margin.

Unsupervised Deep Features for Remote Sensing Image Matching via Discriminator Network

Robust Local Feature Descriptor for Multisource Remote Sensing Image Registration

TextRS: Deep Bidirectional Triplet Network for Matching Text to Remote Sensing Images

StateNet: Deep State Learning for Robust Feature Matching of Remote Sensing Images

A Novel Neural Network for Remote Sensing Image Matching

Unsupervised SAR and Optical Image Matching Using Siamese Domain Adaptation

Deep Semantic-Aware Remote Sensing Image Deblurring.

Optical and SAR Image Matching Using Pixelwise Deep Dense Features

Multimodal Remote Sensing Image Matching via Learning Features and Attention Mechanism

A Robust Multisource Remote Sensing Image Matching Method Utilizing Attention and Feature Enhancement Against Noise Interference

A Deep Semantic Alignment Network for the Cross-Modal Image-Text Retrieval in Remote Sensing

Combining deep learning and ontology reasoning for remote sensing image semantic segmentation

Proxy-Based Rotation Invariant Deep Metric Learning for Remote Sensing Image Retrieval

Self-Supervised Keypoint Detection and Cross-Fusion Matching Networks for Multimodal Remote Sensing Image Registration

RS3Mamba: Visual State Space Model for Remote Sensing Image Semantic Segmentation

RSDehamba: Lightweight Vision Mamba for Remote Sensing Satellite Image Dehazing

Self-Distillation Feature Learning Network for Optical and SAR Image Registration

Multi-Class Remote Sensing Image Retrieval Using Optimized Convolution Neural Network with Weighted Distances

Unsupervised Deep Feature Extraction for Remote Sensing Image Classification

A Light-weight Transformer-based Self-supervised Matching Network for Heterogeneous Images

A Hierarchical Consensus Attention Network for Feature Matching of Remote Sensing Images