影像配准和变化检测对遥感时序信息的提取与分析至关重要。当前，深度学习方法通常将配准和变化检测视为独立任务，缺乏联合处理与显式配准机制。同时，现有变化检测方法缺乏差异特征空间信息与语义信息的协同交互，难以有效构建双时相影像间的实质性空谱差异。此外，二者联合模型的轻量化，对资源受限设备部署、大规模影像批处理至关重要。为此，本文提出一种联合遥感影像配准与变化检测的轻量级网络。首先，利用MobileNet V3 Large提取用于配准和变化检测的多尺度特征。其次，利用空间一致性模块实现半密集特征点匹配，建立跨尺度的空间变换模型，使得不同尺度特征图相互对齐。然后，不同尺度特征图通过时空差异协同模块，增强双时相特征的时空异质性。最后，对多尺度差异特征进行融合，生成变化检测结果。选取SVCD、SYSU-CD和SECOND数据集进行试验，与当前主流的变化检测网络进行了对比。结果表明：本文方法能有效构建待配准影像间的空间变换关系，在定量分析和定性分析方面均优于其它方法，并在网络复杂度方面具有一定优势。

Abstract: Image registration and change detection are two fundamental procedures for extracting and interpreting multitemporal remote sensing information. However, most deep learning approaches treat them as separate tasks, and existing joint frameworks either lack an explicit and interpretable registration mechanism or neglect the collaborative interaction between spatial and semantic difference features. This study aims to develop a lightweight unified network that can align misregistered bi-temporal optical remote sensing images and accurately identify real land-cover changes under geometric offsets and pseudo-changes induced by spatio-spectral variations. A lightweight joint registration and change detection network, named LJRCDNet, is proposed. MobileNetV3-Large is first adopted as a shared encoder to extract four-scale features for both registration and change detection. A spatial consistency module is then designed to perform semi-dense keypoint detection and local descriptor construction. Guided by self-supervised correspondences and knowledge distillation, this module estimates a homography-based spatial transformation and aligns multi-scale feature maps of the pre-event image with those of the post-event image. In the aligned feature space, a spatio-temporal difference collaboration module is introduced. It combines spatial cross-attention and channel cross-attention to model the coupling relationship between spatial difference features and semantic difference features, while depthwise separable convolution and efficient channel attention are used to control computational cost. Finally, multi-scale collaborative difference features are fused through a scale-adaptive perception module and upsampled to generate the final change map. Experiments are conducted on the SVCD, SYSU-CD and SECOND datasets under both co-registered and misregistered settings, and the proposed method is compared with thirteen representative change detection networks. In the co-registered setting, the model without the registration branch still achieves competitive performance with only 5.75 M parameters and 2.90 GFLOPs, indicating the effectiveness of the difference collaboration design. In the misregistered setting, the complete LJRCDNet obtains IoU scores of 79.62%, 65.09% and 52.24% on SVCD, SYSU-CD and SECOND, respectively, outperforming the second-best methods by absolute gains of 4.52%, 5.56% and 5.66%. Qualitative results show that the proposed method suppresses false alarms caused by geometric offsets, illumination changes, seasonal variations and land-object appearance differences, while preserving more complete boundaries of changed objects such as vehicles, impervious surfaces, cultivated land and buildings. Ablation experiments further verify that SCM and STDCM are both necessary: removing SCM reduces IoU by 1.87%, 3.03% and 2.16%, removing STDCM reduces IoU by 1.54%, 3.71% and 1.29%, and replacing SCM with SIFT also degrades performance. Matching experiments show that SCM achieves the best 3-pixel matching accuracy on all datasets. LJRCDNet integrates explicit spatial registration and spatio-temporal difference enhancement into an efficient end-to-end framework. The spatial consistency module provides reliably aligned features for subsequent change detection, and the spatio-temporal difference collaboration module enhances true spatio-spectral differences while suppressing pseudo-changes. The network achieves higher accuracy and better robustness than existing methods with relatively low complexity, making it suitable for large-scale remote sensing image processing and deployment on resource-limited devices. Future work will extend the unified framework to heterogeneous remote sensing imagery and strengthen the robustness of the matching module when large-area changes reduce the number of valid correspondences and degrade their spatial distribution.