This paper proposes a new rotation correction based method to extract regions of interest (ROI) from non-contact finger vein images. Firstly, finger median lines and image center points are used to make rotation correction. Then the arc diameter of fingertips is introduced to locate and determine sizes of ROI .Finally, do the size normalization. It is proved by experiments that the algorithm can effectively eliminate the finger vein image rotation and translation to some extent, and it is still efficient when the light is uneven to a certain degree or images are not clear enough. That is to say, this algorithm has high accuracy and is robust at the same time. © Springer International Publishing 2013.
CITATION STYLE
Zuo, C., Wang, K., & Song, X. (2013). A new ROI extraction method of non-contact finger vein images. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8232 LNCS, pp. 288–297). https://doi.org/10.1007/978-3-319-02961-0_36
Mendeley helps you to discover research relevant for your work.