Training Neocognitron to Recognize Handwritten Digits in the Real World
PAS '97 Proceedings of the 2nd AIZU International Symposium on Parallel Algorithms / Architecture Synthesis
Active and Adaptive Vision: Neural Network Models
BMVC '00 Proceedings of the First IEEE International Workshop on Biologically Motivated Computer Vision
Training Neocognitron to Recognize Handwritten Digits in the Real World
PAS '97 Proceedings of the 2nd AIZU International Symposium on Parallel Algorithms / Architecture Synthesis
| Hi-index | 0.00 |
Using a large scale real-world database ETL-1, we show that the neocognitron trained by unsupervised learning with a winner-take-all process can recognize handwritten digits with a recognition rate higher than 97%. We use the technique of dual thresholds for feature-extracting S-cells, and higher threshold values are used in the learning than in the recognition phase. We discuss how the threshold values affect the recognition rate. The learning method for the cells of the highest stage of the network has been modified from the conventional one, in order to reconcile the unsupervised learning with the use of information of the category names of the training patterns.