Significance: Indocyanine green-based dynamic contrast-enhanced fluorescence imaging (DCE-FI) can objectively assess bone perfusion intraoperatively. However, it is susceptible to motion artifact due to patients' involuntary respiration and mechanical disturbance. Reducing motion artifacts would significantly improve DCE-FI for orthopedic surgical guidance. Aim: Our primary objective is to develop an automated correction method to reduce motion artifacts in DCE-FI and improve the accuracy of bone perfusion assessment. Approach: We developed an automated motion correction approach based on frame-by-frame mutual information (MI) and validated the effectiveness of this approach in various phantom studies and patient images from 45 imaging sessions of fifteen amputees. Results: The MI-based correction reduced motion artifacts by 93% for mechanical disturbances and 76% for simulated respiration in phantom studies. Patient images show improved alignment, improved kinetic curves, and restored bone perfusionrelated parameters with an average correction of 4.3 and 9.6 mm in x- and y-axes per session. Conclusions: The automated MI-based motion correction was able to eliminate motion artifacts effectively and significantly improved the quantitative assessment of bone perfusion by DCE-FI. [ABSTRACT FROM AUTHOR]