1,a),b) 3,c) 1,d) 3 Cappelli Bazen Cappelli Bazen Cappelli 1.,,,,,.,,,,.,,.,,,,.,, 1 Department of Electrical Electronic and Communication Engineering Faculty of Science and Engineering Chuo University 1-13-7 Kasuga Bunkyo-ku 11-8551Tokyo Japan Department of Mathematics Faculty of Science Josai University Keyakidai 1-1 Sakado Japan 3 Advanced Industrial Science and Technology Umezono Tsukuba-shi Ibaraki 305-8568 Japan a) ueda-shusei@imailab.jp b) inuma@josai.ac.jp c) otsuka@ni aist.go.jp d) h-imai@imailab.jp., Ground Truth. Ground Truth,.,,., Ground Truth, Ground Truth, M( ).,, Ground Truth., Ground Truth.,,,., Ground Truth,,, Ground Truth.,,,, c 014 Information Processing Society of Japan 1
,, Ground Truth,,,.,, Ground Truth 3 3., 3,, 1 ( 1 ) 3 ϕ 1 ( ) ϕ Ground Truth,.,.,, 001 R. Cappelli [1],,.,, a, b, c 3, a,, b, c,. a c Bazen [] Thin-Plate Spline Thin-Plate Spline Ground Truth,, Optical Flow.,, Optical Flow, Optical Flow,,,,,, [1],, [1], 11.95, [1],.. 00 R. Cappelli [1],.,...,, a, b, c 3. 1 3. ( ). ([1] ) c 014 Information Processing Society of Japan
a, 3,. b c, 3. a, a, a. ( dx v = (x, y), d = dy ) ( ) sin θ cos θ, R θ = cos θ sin θ (a) (b) 4 3 (a), (b). ([1] ) R. Cappelli, [1], a a. 0 a, Shapedist a (v) = dist a a, (1) dist a, dist a, dist ell dist ell (v) = (v C e ) T A 1 (v C e ) 1 C e = ( Cx C y ) anda x = ( S x 0 0 S y C e, xy, A x, (S x, S y ). c, 4 a,, 4 a, c., 4 a c b. b, distortion(v), c. distortion(v) = v + (v)brake(shapedist a (v), k) 0 if t < 0 brake(t, k) = 1 t π (1 cos( k )) if 0 t k 1 otherwise ) 3. 3.1 SecuGen Plus.,., PC,, TANITA.,.. 3., 5. 0g. 6. 60g,, 0g, 400g 18. 1., 18 10, 360., 1 Optical Flow,. 3.3 Optical Flow 1 18, Optical Flow, c 014 Information Processing Society of Japan 3
R = { (x k j 1, y k j 1) (x k j, y k j ) 1 k n, j 18} 5. R 11. 7 R.,, S x R, S x U.,,,, Optical Flow. 6 (x, y)., 18. Optical Flow, ( [8]). 7. 3.4 (x k j, yk j ) k j (60 + 0(j 1) g ), S k = {(x k 1, y k 1 ), (x k, y k ),..., (x k 18, y k 18)}.. U = S 1 S S 3... S n {R k } (k = 1... n) R k = { (x k j 1, y k j 1) (x k j, y k j ) 1 k n, j 18} 7 (max[r k ]) S k Optical Flow U,. R 3.5 3.4 U 8., [1] 8., a, c a,,.,,.,, 8. [1] c (x, y),. 60g 400g, c., c (x, y). c (x, y) Pi w = (x 1, y 1,..., x i, y i ), c 014 Information Processing Society of Japan 4
A. : 60g c (x, y), P 0 i Pi 400 : 400g c (x, y), () Pi 0 400 Pi, 1 n i AP 0 i P 400 i (3) AP 0 i P 400 i (4) 9 10. 8 1. 10..,,. Cappelli [1], Affine. 8 Capplli a Affine a 10 5., Capplli [1].,, Capplli [1],,., [1],,. 4. 9 1., 8,.,, JSPS ( (A)540017) [1] R. Cappelli, D. Maio, D. Maltoni, Modelling plastic distortion in fingerprint images, in: Proceedings of ICAPR001, Second International Conference on Advances in Pattern Recognition, Rio de Janeiro, 001. [] Asker M. Bazen*, Sabih H. Gerez, Fingerprint match- c 014 Information Processing Society of Japan 5
ing by thin-plate spline modelling of elastic deformations [3] F. Bookstein, Principal warps: Thin-plate splines and the decomposition of deformations, IEEE Trans. Pattern Anal. Mach. Intell. 11 (6) (1989) 567-585. [4] X. Jiang, W. Yau, Fingerprint minutiae matching based onthe local and global structures, in: Proceedings of ICPR000, 15th International Conference on Pattern Recognition, Vol., Barcelona, Spain, 00, pp. 104-1045. [5],,,, 63-607, C(1997), 881-888. [6] / /, 011, 44-59 [7] R. Cappelli, Synthetic Fingerprint generation, in D. Maltoni, D. Maio, A.K. Jain and S. Prabhakar, Handbook of Fingerprint Recognition (Second Edition), Springer, 009 [8] Horn, Berthold K., and Brian G. Schunck. Determining optical flow. 1981 Technical Symposium East. International Society for Optics and Photonics, 1981. c 014 Information Processing Society of Japan 6