Conjoint Cross-Attention Modeling and Joint Feature Calibrating for Remote Sensing Image Change Detection via a Triple-Double Network

Remote sensing (RS) image change detection (CD) based on deep learning (DL), has received increasing attention recently. However, the general independent learning of bi-temporal images ignores the relationship between them, falling short in learning the change information. In this article, a triple-double (TD) framework with the ability of conjoint cross-attention (CCA) modeling and joint feature calibrating is proposed for CD. Specifically, the TD framework composed of a triple-branch encoder and double-branch decoder is constructed to extract diverse features and acquire changed maps with the guidance of original edge cues. To enhance the perception of the connection between the bi-temporal features, the multiscale difference guidance (MDG) module and CCA module are designed for the dual-branch encoder, wherein the CCA introduces a novel and efficient rule for modeling the affinity in spatial and channel dimensions simultaneously. Furthermore, a joint feature calibration (JFC) module is introduced to enhance the expression of feature diversity in the joint features within the single-branch encoder. Experimental results on three public datasets demonstrate the superiority of the proposed method compared to the state-of-the-art (SOTA) methods.