A Comparison of FAST, SURF, Eigen, Harris, and MSER Features

Engr Farman, Ali, Engr Sajid, Ullah Khan, Engr Muhammad, Zarrar Mahmudi, Rahmat Ullah, Sarhad University Peshawar
2016 International Journal of Computer Engineering and Information Technology   unpublished
Precise, successful in desire target, strong healthy and self loading image registration is critical task in the field of computer vision. The most require key steps of image alignment/ registration are: Feature matching, Feature detection, , derivation of transformation function based on corresponding features in images and reconstruction of images based on derived transformation function. This is also the aim of computer vision in many applications to achieve an optimal and accurate image,
more » ... ch depends on optimal features matching and detection. The investigation of this paper summarize the coincidence among five different methods for robust features/interest points (or landmarks) detector and indentify images which are (FAST), Speed Up Robust Features (SURF), (Eigen),(Harris) & Maximally Stable Extremal Regions (MSER). This paper also focuses on the unique extraction from the images which can be used to perform good matching on different views of the images/objects/scenes.