面向移动平台的二进制特征的性能研究

Performance study of binary features for mobile platform

针对移动平台时间空间局限性,研究二进制特征在移动平台的性能。扩展已有的评测标准,以BRIEF、ORB、BRISK和FREAK为研究对象,SIFT、SURF非二进制特征作为参照;使用通用的图片库,研究各个描述子在PC平台和移动平台的时空性能;并研究在旋转变化、尺度变化、模糊处理、照度变化和透视变化下各特征的鲁棒性。研究表明二进制特征描述子具有很高的时空效率,平均提高约10倍;有不同程度的鲁棒性,与SIFT具有可比性,整体性能很好。

Concerning the limitations of mobile platform, the performance of binary features was researched. Evaluation system was extended with new metrics. Then BRIEF （Binary Robust Independent Elementary Feature）, ORB （ORiented BRIEF）, BRISK （Binary Robust Invariant Scalable Keypoint） and FREAK （Fast Retina Keypoint） were selected as the representatives while SIFT （Scale-Invariant Feature Transform） and SURF （Speeded Up Robust Feature） as references. The space and time performance was studied both on mobile device and PC. Also, the robustness of binary features under scale change, rotation, blur, illumination change and perspective change was researched. The results show that binary features have superior performance in time and space that the efficiencies are improved about 10 times. Also they have very good robustness overall which is comparable with SIFTs.