基于移动平台的SIFT算法优化

An Optimization of SIFT Algorithm Based on Mobile Platform

本文通过快速构建图像的低频信息图像,并完成其上的特征点检测,简化SIFT(Scale Invariant Feature Transform)图像匹配算法高斯空间构造的复杂度,以此来提高特征点检测效率,提高算法在移动设备上的运行速度,并在Android平台上将优化算法和SIFT算法进行对比。实验结果表明,优化的算法在多种场景下保证配准效果的同时提高了特征点检测速度,尤其是在旋转变化条件下效果更好。

Through the rapid construction of low-frequency information and the feature point detection of the image,simplified SIFT（ Scale Invariant Feature Transform） image matching algorithm for a Gaussian spatial structure complexity to improve the efficiency of feature point detection and improve the algorithm running speed in mobile devices. Finally,the optimization algorithm and SIFT algorithm are compared in the Android platform. The experimental results show that the optimization algorithm ensure the registration effect and improves the detection speed of the feature points in a variety of scenarios,especially the effect of the rotation change is better.