Distinctive Image Features from Scale-Invariant Keypoints
International Journal of Computer Vision
IEEE Transactions on Image Processing
Hi-index | 0.00 |
Referring to the less correlative of Infrared and visible image registration, which lead to low matching accuracy rate. An improved registration of infrared and visible images approach based on SIFT operator is proposed. Firstly, Difference-of-Gaussian scale-spaces were built that based on Sub-band of the wavelet decomposition. The scale-invariant feature points were detected in each scale. Then, SIFT vector values were constrained through given threshold, and angle between the key-points descriptors is treated as the similarity measure. Finally, the least squares method was employed to find optimal solutions to affine transform equations. Experimental results show that the proposed increase the anti-jamming and approach matching accuracy rate.