Classification methods for handwritten digit recognition: A survey

Introduction/purpose: This paper provides a survey of handwritten digit recognition methods tested on the MNIST dataset. Methods: The analyzes, synthesizes and compares development different classifiers applied to problem, from linear convolutional neural networks. Results: Handwritten classification accuracy dataset while using training testing sets is now higher than 99.5% most successful method network. Conclusions: problem with numerous real-life applications. Accurate various handwriting styles, specifically digits task studied for decades this summarizes achieved results. best results have been networks worst are classifiers. give better if expended data augmentation.