目的 将FAST特征点检测算法与FREAK特征点描述算法相结合用于图像匹配， 以改善图像识别算法在移动终端上对印刷品图像的识别性能。方法 匹配的过程需要不断对摄像头获得的图像进行实时处理， 但由于手机设备的处理器、 内存等硬件条件有限， 因此匹配算法的速度和效率是需要首要考量的指标。借鉴ORB的FAST特征点检测算法的思想， 将FAST与FREAK特征点描述算法相结合用于图像匹配， 然后与ORB算法的匹配速度和匹配精确度进行比较。结果 结合了FAST特征点检测算法的FREAK算法， 与ORB算法相比， 匹配速度有了一定的提升， 匹配精确度也基本可以满足纸质印刷品图像匹配的需求。结论 在移动终端进行印刷品图像识别与匹配时， 文中的研究能够在保证图像识别准确性的基础上使识别算法的运算速度得到一定的提升。

The FAST feature point′ s detection algorithm and the FREAK feature point′ s description algorithm were combined and applied in image matching, to improve the image recognition performance of the image recognition algorithm in the mobile phone. Due to the constant need for real-time computation of the images obtained by the camera, and the limited hardware capabilities of the mobile phone, the matching algorithm′ s accuracy, speed and efficiency should be emphasized. Referring to the thought of the FAST feature point′ s detection algorithm of the ORB, this paper combined FAST with the FREAK feature point′ s description algorithm to accomplish the function of image matching. And then the matching speed and matching accuracy were compared to those of the ORB algorithm. Experimental results showed that the FREAK algorithm combining the FAST feature points detection algorithm had higher matching speed than the ORB algorithm, and the matching accuracy could basically meet with the demand of image matching in paper printing. In terms of image recognition and matching using mobile phone, the research of this paper could improve the speed of the matching algorithm on the basis of satisfying the accuracy of image recognition.