Optical iconic filters for large class recognition

David Casasent; Abhijit Mahalamobis

doi:10.1364/AO.26.002266

Applied Optics
Vol. 26,
Issue 11,
pp. 2266-2273
(1987)
•https://doi.org/10.1364/AO.26.002266

Optical iconic filters for large class recognition

David Casasent and Abhijit Mahalamobis

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
David Casasent and Abhijit Mahalamobis, "Optical iconic filters for large class recognition," Appl. Opt. 26, 2266-2273 (1987)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Abstract

Approaches are advanced for pattern recognition when a large number of classes must be identified. Multilevel encoded multiple-iconic filters are considered for this problem. Hierarchical arrangements of iconic filters and/or preprocessing stages are described. A theoretical basis for the sidelobe level and noise effects of filters designed for large class problems is advanced. Experimental data are provided for an optical character recognition case study.

Full Article | PDF Article

More Like This

Model-based knowledge-based optical processors

David Casasent and Suzanne A. Liebowitz
Appl. Opt. 26(10) 1935-1942 (1987)

Rule-based symbolic processor for object recognition

David Casasent and Abhijit Mahalanobis
Appl. Opt. 26(22) 4795-4802 (1987)

Minimum average correlation energy filters

Abhijit Mahalanobis, B. V. K. Vijaya Kumar, and David Casasent
Appl. Opt. 26(17) 3633-3640 (1987)

Previous Article Next Article

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (5)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (6)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (12)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Number of training images N_T (5/class)	Correlation plane PSR
10	2.04
20	1.98
40	1.76
60	1.48
100	1.52
200	0.98
400	0.04
930	0.006

N_T	S_F (scatter)	Maximum pixel value
2	0.02	0.05
15	0.03	0.06
25	0.18	0.10
35	0.35	0.22
75	0.78	0.82
115	0.87	0.95
130	0.89	0.96
150	0.92	1.29
170	1.05	1.62
190	1.16	1.66
248	1.33	2.31
930	18.10	9.90

Input test character	Response for filters F₁–F₄

	F1	F2	F3	F4
E	0.3299	0.6601	0.6600	0.6599
T	0.6600	0.3299	0.3301	0.9899
h	0.6599	0.6600	0.9899	0.6599
1	0.9900	0.3299	0.3300	0.9901
6	0.3301	0.3299	0.6599	0.3300

Input test character	Response (and error) for filters F1–F4

	F1	F2	F3	F4
E	0.24(0.09)	0.55(0.11)	0.62(0.04)	0.75(0.24)*
T	0.57(0.09)	0.28(0.05)	0.29(0.04)	0.91(0.08)
W	0.58(0.08)	0.35(0.02)	0.60(0.06)	1.03(0.04)
h	0.68(0.02)	0.60(0.06)	0.89(0.10)	0.62(0.04)
u	0.62(0.04)	0.85(0.14)	0.92(0.07)	0.90(0.09)
1	0.90(0.09)	0.24(0.09)	0.28(0.05)	0.91(0.08)
3	0.28(0.05)	0.20(0.13)	0.34(0.01)	0.56(0.10)
6	0.27(0.06)	0.29(0.04)	0.57(0.09)	0.30(0.03)
9	0.28(0.05)	0.58(0.08)	0.29(0.04)	0.31(0.02)

Input test character	Response (and error) for filters F1–F4

	F1	F2	F3	F4
Q	0.37(0.04)	0.91(0.08)	0.95(0.04)	0.90(0.09)
R	0.33(0.33)	0.31(0.02)	0.38(0.05)	0.32(0.01)
V	0.58(0.08)	0.31(0.02)	0.62(0.04)	0.70(0.04)
t	1.17(0.18)	0.39(0.06)	0.20(0.13)	0.54(0.12)
u	0.91(0.08)	0.34(0.01)	0.37(0.04)	0.96(0.03)
x	1.03(0.04)	0.34(0.01)	0.60(0.06)	0.91(0.08)
5	0.41(0.08)	0.27(0.06)	0.62(0.04)	0.97(0.02)
6	0.21(0.12)	0.28(0.05)	1.02(0.03)	0.31(0.02)
9	0.38(0.05)	0.68(0.02)	0.35(0.02)	0.32(0.01)
0	0.37(0.04)	0.32(0.01)	0.39(0.06)	0.27(0.06)

Input test character	Font	Response (and error) for filters F1–F4

		F1	F2	F3	F4
r		0.48(0.18)	0.14(0.15)	0.98(0.01)	0.92(0.07)
S		0.56(0.10)	0.31(0.02)	0.28(0.05)	0.67(0.01)
s	Times	0.97(0.02)	0.30(0.03)	0.30(0.03)	0.36(0.03)
2		0.37(0.04)	0.32(0.01)	0.36(0 03)	1.01(0.02)
4		0.31(0.02)	0.33(0.00)	0.69(0.03)	0.31(0.02)

Abstract

Cited By

Figures (5)

Tables (6)

Equations (12)

Applied Optics