A method for task-oriented examination of SURF keypoint detector accuracy is presented in the paper. It consists of generating test images, based on a given exemplar, processed by affine transformations: random rotation and scaling, and varying degree of degradations: darkening, blurring, noising, and compression. Details of applied degradation procedure are presented, followed by essentials of SURF-based images matching. A distance metric is adapted for SURF accuracy assessment. It consists of calculation of Euclidean distances between anchor points in the exemplar and degraded images. Finally, results of tests are presented, and impact of degradation level on SURF accuracy is discussed.
Autorzy
Informacje dodatkowe
- Kategoria
- Aktywność konferencyjna
- Typ
- materiały konferencyjne indeksowane w Web of Science
- Język
- angielski
- Rok wydania
- 2013
