Track XIII

Advances and Frontier Application in Intelligent Remote Sensing Interpretation (Submission Deadline: October 30, 2026) 遥感智能解译方法前沿进展与领域应用

Chair:	Co-chairs:

Jianing Wang	Di Zhang	Lin Li
‌Xidian University, China	Northwest Normal University, China	Northwestern Polytechnical University, China

Topics:
Multimodal feature extraction for remote sensing data (遥感数据的多模态特征提取) High-efficiency multimodal and multisource remote sensing data processing and model design (高效多模态遥感数据处理方法) Continual learning in remote sensing interpretation (持续学习方法在遥感解译算法中的应用) Multimodal and multisource remote sensing model design (多模态与多源遥感模型设计) Brain-inspired multimodal feature extraction and model design (脑启发的多模态特征提取与模型设计) Lightweight network design for remote sensing processing (轻量化网络在遥感数据与算法设计中的处理与应用) The optimization algorithm design for remote sensing interpretation model (遥感处理模型的优化算法设计) Multimodal remote sensing for change detection, classification, anomaly detection, feature fusion, object detection and tracking (多模态遥感在变化监测、地物分类、异常检测、特征提取、目标检测与追踪等任务中的应用) Application of hyperspectral remote sensing in tumor tissue diagnosis, ecological and environmental monitoring, agriculture, forest protection, fire warning and surveillance. water pollution monitoring and lunar mineral detection (高光谱遥感在肿瘤组织诊断、生态环境监测、农业、林区保护、火灾预防与监测、水污染监测、灾害检测与月球矿物探测) Few-shot and Zero-shot learning methods for remote sensing interpretation (小样本与零样本学习技术在遥感解译方法中的应用) Spike neural network for remote sensing image classification, segmentation, fusion, detection and recognition (脉冲神经网络在遥感图像分类、分割、融合、检测和识别中的应用)

Summary:
With the rapid development of remote sensing technology, remote sensing data has shown a trend of high resolution, multi-modal, and massive volume. Intelligent interpretation has become the core means for extracting remote sensing information. Focusing on cutting-edge methods such as deep learning, multi-modal fusion, self-supervised learning, few-shot and zero-shot learning, continuous learning methods, and model optimization, this special topic mainly explores and exchanges the innovative applications and original research results of related frontier technologies and methods in fields such as national land survey, ecological environment monitoring, disaster response, agricultural assessment, and urban planning. It also discusses the challenges and application prospects of remote sensing interpretation technology from method to application implementation.

随着对地观测技术的飞速发展，遥感数据呈现出高分辨率、多模态、海量化的趋势，智能解译成为遥感信息提取的核心手段。围绕深度学习、多模态融合、自监督学习、小样本与零样本学习、持续学习方法以及模型优化等前沿方法，本专题主要探讨交流相关前沿技术与方法在国土调查、生态环境监测、灾害应急、农业评估、城市规划等领域的创新应用与原创性研究成果，探讨遥感解译技术从方法到应用落地的挑战与应用展望。