天基光学遥感动目标检测旨在对遥感卫星视频中具有连续运动特性的目标进行定位和分类，比如遥感视频卫星中的运动车辆、舰船和飞机。随着遥感视频卫星技术和深度学习技术的快速发展，基于模型驱动的传统遥感动目标检测方法正朝着基于数据驱动的深度学习方法进行演变，以完成高可靠、高时效、高性能的天基光学遥感图像动目标检测。本文介绍了光学遥感视频卫星的发展现状，并对基于模型驱动和基于数据驱动的光学遥感动目标检测方法进行了总结，梳理和分析了光学遥感动目标检测技术的发展历程。最后，在此基础上对光学遥感动目标检测的未来发展趋势进行了展望。

The goal of spaceborne optical remote sensing for moving target detection is to locate and classify targets with continuous motion characteristics in satellite video, such as moving vehicles, ships, and aircraft. With the rapid development of spaceborne optical remote sensing satellite technology and deep learning techniques, traditional model-driven methods for remote sensing moving target detection are evolving towards data-driven deep learning methods to achieve high reliability, efficiency, and performance. This paper introduces the current status of optical remote sensing satellite systems and provides a summary of model-driven and data-driven approaches for optical remote sensing moving target detection. The development process of optical remote sensing moving target detection technology is reviewed, analyzed, and discussed. Finally, based on this foundation, future trends in optical remote sensing moving target detection are proposed.