介绍了一种基于windows2 0 0 0操作系统的高速物流图像数据采集与实时剔除物流中异物技术的实现原理与方法。Windows2 0 0 0本身不是实时操作系统 ,难于实现大数据量的连续采集与实时处理 ,文中介绍了解决这一问题的方法。考虑到要进行高速数据采集和处理的连续性和实时性 ,提出了内存数据直接处理和双缓存及多缓存处理技术 ,并介绍了这一技术在实时在线烟草异物检测与剔除系统中的成功应用。 This paper introduces a high-speed logistics image data acquisition system based on Windows 2000 operating system and real-time culling of foreign objects in the logistics technology to achieve the principle and method. Windows2 0 0 0 itself is not a real-time operating system, it is difficult to achieve continuous acquisition of large amounts of data and real-time processing, the paper describes the method to solve this problem. Considering the continuity and real-time of high-speed data acquisition and processing, this paper proposes the direct processing of memory data and the double-buffer and multi-buffer processing technologies. The successful application of this technology in the real-time online tobacco foreign body detection and rejection system .