Authors : Zinah Mohsin Arkah, Reem Ibrahim Hasan, Sif. K. Ebis, Haider A. Alwzwazy and Mohammed A. Fadhel
Abstract: It is important to improve the performance of images matching, so that, many algorithms have been studied to achieve this goal. However, finding best one is still a very interested field of study because the performance of matching is a wide term containing many factors like rotating invariant, scaling invariant, noise invariant and etc. In this study, we compare the performance of the algorithms by calculating the score of matching which is number of key points and number of matching. The descriptor should study with all cases such as rotation with different angels and scaling of different sizes. We have implemented the Scale Invariant Feature Transform (SIFT) as a first algorithm then we applied Affine Scale Invariant Feature Transform (ASIFT), finally we applied moment invariants. Therefore by evaluating the results of these algorithms we can know which one has good matches. ASIFT has shown a great performance in term of image matching.
Zinah Mohsin Arkah, Reem Ibrahim Hasan, Sif. K. Ebis, Haider A. Alwzwazy and Mohammed A. Fadhel, 2019. Image Matching Performance Comparison using SIFT, ASIFT and Moment Invariants. Journal of Engineering and Applied Sciences, 14: 7656-7662.