Effective Handwritten Digit Recognition using Deep Convolution Neural Network

This paper proposed a simple neural network approach towards handwritten digit recognition using convolution. With machine learning algorithms like KNN, SVM/SOM, recognizing digits is considered as one of the unsolvable tasks due to its distinctiveness in the style of writing. In this paper, Convolution Neural Networks are implemented with an MNIST dataset of 70000 digits with 250 distinct forms of writings. The proposed method achieved 98.51% accuracy for real-world handwritten digit prediction with less than 0.1 % loss on training with 60000 digits while 10000 under validation.

PDF Abstract

Datasets


Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Handwritten Digit Recognition MNIST CNN Accuracy 96.95% # 1

Methods