Handprinted digit recognition using spatiotemporal connectionist models

We describe a connectionist model for recognizing unconstrained handprinted digits. Instead of treating the input as a static signal, the image is scanned over time and converted into a time-varying signal. The temporalized image is processed by a spatiotemporal connectionist network. The resulting system o ers shift-invariance along the temporalized axis, a reduction in the number of free parameters, and the ability to process images of arbitrary length. Results for handprinted digit recognition are among the best reported to date on a set of real-world ZIP code digit images, provided by the United Stated Postal Service. The system achieved a 99.1% recognition rate on the training set and a 96.0% recognition rate on the test set with no rejections. A 99.0% recognition rate on the test set was achieved when 14.6% of the images were rejected.