Advancing Weakly-Supervised Change Detection in Satellite Images via Adversarial Class Prompting

2025-08-23

AI Summary

Weakly supervised change detection (WSCD) in remote sensing imagery often misclassifies background variations—such as illumination and weather changes—as genuine object-level changes. To address this, we propose Adversarial Class Prompting Mechanism (ACPM), which employs adversarial perturbations to uncover confounding background features. Aiming to robustly distinguish true changes from co-occurring noisy samples, ACPM integrates erroneous one-hot label activation with exponential moving average to construct an online global prototype for adaptive noise identification and correction. This work is the first to introduce adversarial prompting into weakly supervised remote sensing change detection, significantly enhancing model discriminative robustness under complex scene conditions. The method is architecture-agnostic, seamlessly compatible with CNNs, Transformers, and SAM-based backbones. Extensive experiments on multiple benchmarks demonstrate consistent suppression of background false positives, yielding mIoU improvements of 3.2–5.7%. It exhibits strong generalization capability, and the source code is publicly available.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Weakly-Supervised Change Detection (WSCD) aims to distinguish specific object changes (e.g., objects appearing or disappearing) from background variations (e.g., environmental changes due to light, weather, or seasonal shifts) in paired satellite images, relying only on paired image (i.e., image-level) classification labels. This technique significantly reduces the need for dense annotations required in fully-supervised change detection. However, as image-level supervision only indicates whether objects have changed in a scene, WSCD methods often misclassify background variations as object changes, especially in complex remote-sensing scenarios. In this work, we propose an Adversarial Class Prompting (AdvCP) method to address this co-occurring noise problem, including two phases: a) Adversarial Prompt Mining: After each training iteration, we introduce adversarial prompting perturbations, using incorrect one-hot image-level labels to activate erroneous feature mappings. This process reveals co-occurring adversarial samples under weak supervision, namely background variation features that are likely to be misclassified as object changes. b) Adversarial Sample Rectification: We integrate these adversarially prompt-activated pixel samples into training by constructing an online global prototype. This prototype is built from an exponentially weighted moving average of the current batch and all historical training data. Our AdvCP can be seamlessly integrated into current WSCD methods without adding additional inference cost. Experiments on ConvNet, Transformer, and Segment Anything Model (SAM)-based baselines demonstrate significant performance enhancements. Furthermore, we demonstrate the generalizability of AdvCP to other multi-class weakly-supervised dense prediction scenarios. Code is available at https://github.com/zhenghuizhao/AdvCP

Problem

• Reducing background noise misclassification in weakly-supervised change detection
• Addressing co-occurring noise in satellite image change analysis
• Improving object change identification with limited supervision labels

Innovation

• Adversarial Prompt Mining to activate erroneous features
• Adversarial Sample Rectification using global prototype
• Seamless integration without additional inference cost