In this paper we present a two-stage classifier for the recognition of freight train ID numbers under outdoor environment. This kind of recognition is different from ordinary OCR in that images of numerals are usually corrupted by "noises" caused by dirty materials, varying outdoor illumination, and so on. We focused our attention on developing a robust classifier against such noises. The two-stage classifier is constructed by cascade connection of a main classifier and an auxiliary classifier. Since ID numbers are printed with relatively small size and shape variations, we adopted a template-matching technique in the main classifier. Three top-scored candidates from the main classifier are given to the auxiliary classifier which consists of 10 expert neural networks. These neural nets exploit local characteristics of each numeral. We collected several hundreds of sample images to test our algorithm, and some promising results were obtained.
[1]
Jin Wang,et al.
Resolving multifont character confusion with neural networks
,
1993,
Pattern Recognit..
[2]
Harvey A. Cohen,et al.
Targetting Number Plates Effectively Using Sparse/Full Templates and Coarse/Fine Template Matching
,
1990,
MVA.
[3]
Richard P. Lippmann,et al.
An introduction to computing with neural nets
,
1987
.
[4]
Henry S. Baird,et al.
Document image defect models
,
1995
.
[5]
Kazuhiko Yamamoto,et al.
Structured Document Image Analysis
,
1992,
Springer Berlin Heidelberg.
[6]
N. Otsu.
A threshold selection method from gray level histograms
,
1979
.