Using Generative Models for Handwritten Digit Recognition
IEEE Transactions on Pattern Analysis and Machine Intelligence
Representation and Recognition of Handwritten Digits Using Deformable Templates
IEEE Transactions on Pattern Analysis and Machine Intelligence
IEEE Transactions on Pattern Analysis and Machine Intelligence
Affine-Invariant Recognition of Gray-Scale Characters Using Global Affine Transformation Correlation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Efficient Pattern Recognition Using a New Transformation Distance
Advances in Neural Information Processing Systems 5, [NIPS Conference]
Handwritten Character Recognition Using Piecewise Linear Two-Dimensional Warping
ICDAR '01 Proceedings of the Sixth International Conference on Document Analysis and Recognition
Hi-index | 0.00 |
This paper addresses the problem of to what extentlinear transformation can alleviate nonlinear distortion.We investigate a technique of global affinetransformation (GAT) correlation to absorb lineardistortion between gray-scale images. Features used inGAT correlation are occurrence probabilities of blackpixels or gradients. Experiments using the handwrittennumeral database IPTP CDROM1B show that theentropy of GAT-superimposed images decreases byaround 15%. Furthermore, gray-level-based GATcorrelation improves the recognition rate from 85.78% to91.01%, while gradient-based GAT correlation improvesthe recognition rate from 91.80% to 94.02%. Theseresults show that GAT correlation has a marked effect ofimproving both shape matching and discriminationabilities by extracting linear distortion from nonlinearone.