Handwritten digit recognition with a back-propagation network
Advances in neural information processing systems 2
Object Recognition from Local Scale-Invariant Features
ICCV '99 Proceedings of the International Conference on Computer Vision-Volume 2 - Volume 2
ASIFT: A New Framework for Fully Affine Invariant Image Comparison
SIAM Journal on Imaging Sciences
Automatic change detection of driving environments in a vision-based driver assistance system
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
The paper presents a method of static graphical pattern identification inspired by human perception. Cooperation of visual cortex regions is thought to play a major role in the perception, that is why in our approach this cooperation is modelled by joining algorithms responsible for shape and color processing. In order to obtain more stable set of characteristic points, the SIFT algorithm has been modified. To acquire information about shapes Harris operator and Hough transform are considered. The proposed method achieves about 28% less number of incorrect identification comparing to the results obtained by the classical SIFT algorithm.