Feature Point Extraction Method of X-Ray Image Based on Scale Invariant
Feature Point Extraction Method of X-ray Image Based on Scale Invariant is proposed in this paper for industrial X-ray image with low contrast and some artifacts. First of all, the scale transformation of original image is adopted by the Gaussian kernel to building the DOG multi-scale pyramid. Then, the location and scale of the key points is fixed by the three-dimensional quadratic function. Finally, the Simply SIFT descriptor illustrates the key points. Experimental results show that the algorithm has good stability in translation, rotation and affine transformation, especially with 10 percent normalized Gaussian noise, this algorithm can still be detected feature points accuracy.
Junpeng Shao and Xianli Liu
Y. W. Wang et al., "Feature Point Extraction Method of X-Ray Image Based on Scale Invariant", Applied Mechanics and Materials, Vol. 274, pp. 667-670, 2013