| Title | Adaptive Fingerprint Image Enhancement with Emphasis on Preprocessing of Data |
| Authors | Josef Ström Bartunek, Mikael Nilsson, Benny Sällberg, Ingvar Claesson |
| Full-text | Available as PDF, Restricted Access |
| Alternative Location | http://dx.doi.org/10.1109/T..., Restricted Access |
| Publication | IEEE Transactions on Image Processing |
| Year | 2013 |
| Volume | 22 |
| Issue | 2 |
| Pages | 644 - 656 |
| Document type | Article |
| Status | Published |
| Quality controlled | Yes |
| Language | eng |
| Publisher | IEEE |
| Abstract English | This article proposes several improvements to an<br> adaptive fingerprint enhancement method that is based on<br> contextual filtering. The term adaptive implies that parameters<br> of the method are automatically adjusted based on the input<br> fingerprint image. Five processing blocks comprise the adaptive<br> fingerprint enhancement method, where four of these blocks are<br> updated in our proposed system. Hence, the proposed overall<br> system is novel. The four updated processing blocks are; preprocessing,<br> global analysis, local analysis and matched filtering.<br> In the pre-processing and local analysis blocks, a nonlinear<br> dynamic range adjustment method is used. In the global analysis<br> and matched filtering blocks, different forms of order statistical<br> filters are applied. These processing blocks yield an improved<br> and new adaptive fingerprint image processing method. The<br> performance of the updated processing blocks is presented in the<br> evaluation part of this paper. The algorithm is evaluated towards<br> the NIST developed NBIS software for fingerprint recognition on<br> FVC databases. |
| Keywords | successive mean quantization transform, Fourier transform, image processing, directional filtering, spectral feature estimation, |
| ISBN/ISSN/Other | ISSN: 1941-0042 (online) |
Questions: webmaster
Last update: 2013-04-11
Centre for Mathematical Sciences, Box 118, SE-22100, Lund. Telefon: +46 46-222 00 00 (vx)