Abstract: In research on the change detection of remote sensing images, the ubiquitous presence of noise in the images represents a key factor interfering with the detection accuracy. The available SFCM-EMD framework based on spatial fuzzy C-means clustering (SFCM) and the earth mover's distance (EMD) can effectively enhance the robustness of change detection in the presence of noise. However, the EMD algorithm exhibits high computational complexity, thereby struggling to achieve rapid change detection and to improve detection efficiency while remaining noise-robust. By integrating the SFCM with integrated region matching (IRM), this study established the SFCM-IRM framework for noise-robust rapid change detection. Under noisy conditions, the SFCM-IRM framework can ensure high detection accuracy while significantly reducing algorithm runtime. The experimental results indicate that compared to the SFCM-EMD framework, the SFCM-IRM framework delivered similar detection accuracy, with a difference of merely 0.004 4 in average Kappa coefficients. Meanwhile, the SFCM-IRM algorithm exhibited an average runtime of only 0.98 seconds in contrast to the 10.29 seconds of the SFCM-EMD algorithm, signaling a significant improvement in computational efficiency. These results confirm the theoretical value and application potential of the SFCM-IRM framework in rapid noise-resistant change detection.