Multi-Scale Harris Corner Detection Based on B-Spline

Wenqiu Zhu, Keke Xu

Abstract


 The existing Harris corner detection algorithm using mostly Gaussian low-pass filter to smooth image, and there are some phenomena about loss of information and location of the corner offset in images, at the same time, the single-scale Harris corner detection algorithm does not have the scale invariance. B-spline function converges to a Gaussian function, therefore we combined B-spline wavelet multi-scale theory and Harris, and proposed multi-scale Harris corner detection method based on B-spline. Firstly, we used B-spline function to smooth filter image at different scales. Secondly image and B-spline convolution template were calculated by convolution operator. Finally,we extracted alternative corner from the different scale images, and searching for the extreme value of scale space as the location and characteristics scale of the feature points within the search window template that has belonged to a fixed size at the center of Harris corner. The experiments show that the proposed method not only maintains the good performance of Harris operator, but also has scale invariance.

 

DOI : http://dx.doi.org/10.11591/telkomnika.v12i4.4797


Keywords


B-spline function; multi-scale; feature detect; invariance

Full Text:

PDF

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics