This paper puts forward a speeded up robust nonlinear scale invariant feature ( SURNSIF ) . Noise is wiped off and edge response is guaranteed through the fast solving of nonlinear scale space. Adaptive selection of number of scale space and the Adaptive and Generic corner detection based on the accelerated segment test ( AGAST) , combined with frame Laplace filter via removing edge response take account of the detection accuracy and real⁃time performance. Constructing descriptor overlap, introduction of gauge derivatives and the constraint of feature point in the nonlinear scale space location enhance the accuracy. Comparing to scale invariant feature trans⁃form(SIFT), speeded up robust features(SURF), KAZE, binary robust invariant scalable keypoints(BRISK), AGAST and fast⁃Hessian experiments, the SURNSIF reveals stronger robustness with 5 kinds of changes, and its registration speed is faster. Compared with KAZE, comprehensive robustness is increased about 10.87%, and the speed is increased about 47%.%提出了一种快速鲁棒性非线性尺度不变的特征匹配算子( speeded up robust nonlinear scale in⁃variant feature,SURNSIF),通过检测子非线性尺度空间的快速求解去除了噪声,同时保证了图像边缘细节,并将自适应选取尺度空间组数、adaptive and generic corner detection based on the accelerated seg⁃ment test( AGAST)与框状拉普拉斯滤波器去除边缘响应相结合,兼顾了检测的准确性与实时性;描述子交叠带的构建、规范微分响应与非线性尺度空间约束的引入增强了描绘准确性。通过与scale in⁃variant feature transform ( SIFT )、speeded up robust features ( SURF )、KAZE、binary robust invariant scalable keypoints(BRISK)、AGAST以及快速海森(fast⁃Hessian)的实验对比,SURNSIF的5种变换鲁棒性均较强,同时速度也更快,综合性能较KAZE提高约10.87%,速度提高约47%。
展开▼