Hybrid feature extraction and feature selection for improving recognition accuracy of handwritten numerals
Authors:
- P. ZhangPZP. Zhang
- T.D. BuiTBT.D. Bui
- C.Y. SuenCSC.Y. Suen
- 1 January 2005
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 136-140 Vol. 1
Abstract
The recognition of handwritten numerals is a challenging task in pattern recognition. It can be considered as one of the benchmarks in evaluating feature extraction methods and the performance of classifiers. In this paper, we propose a new method to improve the recognition accuracy of handwritten numerals by using hybrid feature extraction and random feature selection. First, we present seven feature extraction methods. A novel multi-class divergence criterion for large scale feature analysis is proposed and a random feature selection strategy is used to congregate three new hybrid feature sets. The new congregated features are complementary as they are formed from different original feature sets extracted by different means. Experiments conducted on MNIST database show that our proposed method can increase the recognition accuracy.Keywords
This publication has 9 references indexed in Scilit:
- Extraction of Hybrid Complex Wavelet Features for the Verification of Handwritten Numerals2004
- Feature dimensionality reduction for the verification of handwritten numeralsPattern Analysis and Applications, 2004
- Handwritten digit recognition: investigation of normalization and feature extraction techniquesPattern Recognition, 2004
- Multiresolution recognition of handwritten numerals with wavelet transform and multilayer cluster neural network2002
- Handwritten numeral recognition using gradient and curvature of gray scale imagePattern Recognition, 2002
- Image processing with complex waveletsPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 1999
- The random subspace method for constructing decision forestsIEEE Transactions on Pattern Analysis and Machine Intelligence, 1998
- Floating search methods in feature selectionPattern Recognition Letters, 1994
- Computer recognition of unconstrained handwritten numeralsProceedings of the IEEE, 1992