Ohnishi Lab. |
Computer Vision Research Theme |
Recognition of Handwritten Numerals by Multilayer Perceptron |
An OCR system is proposed that can recognize hand-written numerals regardless of changes in rotation and scale. The system consists of two phases. In the first phase, a binary input image is transformed with complex-log mapping followed by the Fourier transform into a rotation-and-scale invariant image. Then the transformed image is fed into a multi-layer neural network, the weights of which are modified by the error-back-propagation algorithm to absorb slight shape distortions. The system was implemented and tested using hand-written numerals. As results, high recognition rates of 90 to 95 % are obtained. One method for improving performance is suggested, too. |