FRACTIONAL-POWER SYNTHETIC DISCRIMINANT FUNCTIONS

被引:18
作者
BRASHER, JD
KINSER, JM
机构
关键词
SYNTHETIC DISCRIMINANT FUNCTION; FRACTIONAL POWER; COEFFICIENT ROOTING; POWER SPECTRAL DENSITY; GENERALIZATION; DISCRIMINATION;
D O I
10.1016/0031-3203(94)90038-8
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The standard synthetic discriminant function (SDF) generalizes well. That is, it recognizes objects represented by, but not included in, the training set from which it is synthesized. However, it also correlates with objects not represented by the training set. That is, it does not discriminate well. Conversely, the minimum average correlation energy (MACE) SDF discriminates, but does not generalize, well. By using a power spectrum normalization procedure, a parametric SDF which generalizes better than the MACE SDF and discriminates better than the standard SDF is obtained
引用
收藏
页码:577 / 585
页数:9
相关论文
共 9 条
[1]   MULTIVARIANT TECHNIQUE FOR MULTICLASS PATTERN-RECOGNITION [J].
HESTER, CF ;
CASASENT, D .
APPLIED OPTICS, 1980, 19 (11) :1758-1761
[2]  
KINSER JM, 1992, P SOC PHOTO-OPT INS, V1701, P188, DOI 10.1117/12.138327
[3]   MINIMUM-VARIANCE SYNTHETIC DISCRIMINANT FUNCTIONS [J].
KUMAR, BVKV .
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION, 1986, 3 (10) :1579-1584
[4]   TUTORIAL SURVEY OF COMPOSITE FILTER DESIGNS FOR OPTICAL CORRELATORS [J].
KUMAR, BVKV .
APPLIED OPTICS, 1992, 31 (23) :4773-4801
[5]   PERFORMANCE-MEASURES FOR CORRELATION FILTERS [J].
KUMAR, BVKV ;
HASSEBROOK, L .
APPLIED OPTICS, 1990, 29 (20) :2997-3006
[6]   MINIMUM AVERAGE CORRELATION-ENERGY FILTERS [J].
MAHALANOBIS, A ;
KUMAR, BVKV ;
CASASENT, D .
APPLIED OPTICS, 1987, 26 (17) :3633-3640
[7]  
PRATT WK, 1978, DIGITAL IMAGE PROCES, P326
[8]   FILTER DESIGN FOR OPTICAL-PATTERN RECOGNITION - MULTICRITERIA OPTIMIZATION APPROACH [J].
REFREGIER, P .
OPTICS LETTERS, 1990, 15 (15) :854-856
[9]  
SULLIVAN DJ, 1992, P SOC PHOTO-OPT INS, V1701, P178, DOI 10.1117/12.138326