Abstract:Traditional systems of handwriting recognition have relied on handcrafted features and a large amount of prior knowledge. Training an Optical character recognition (OCR) system based on these prerequisites is a challenging task. Research in the handwriting recognition field is focused around deep learning techniques and has achieved breakthrough performance in the last few years. Still, the rapid growth in the amount of handwritten data and the availability of massive processing power demands improvement in recognition accuracy and deserves further investigation. Convolutional neural networks (CNNs) are very effective in perceiving the structure of handwritten characters/words in ways that help in automatic extraction of distinct features and make CNN the most suitable approach for solving handwriting recognition problems. Our aim in the proposed work is to explore the various design options like number of layers, stride size, receptive field, kernel size, padding and dilution for CNN-based handwritten digit recognition. In addition, we aim to evaluate various SGD optimization algorithms in improving the performance of handwritten digit recognition. A network's recognition accuracy increases by incorporating ensemble architecture. Here, our objective is to achieve comparable accuracy by using a pure CNN architecture without ensemble architecture, as ensemble architectures introduce increased computational cost and high testing complexity. Thus, a CNN architecture is proposed in order to achieve accuracy even better than that of ensemble architectures, along with reduced operational complexity and cost. Moreover, we also present an appropriate combination of learning parameters in designing a CNN that leads us to reach a new absolute record in classifying MNIST handwritten digits. We carried out extensive experiments and achieved a recognition accuracy of 99.87% for a MNIST dataset.

Handwritten Digit String Recognition Using Convolutional Neural Network

Handwritten Digit String Recognition by Combination of Residual Network and RNN-CTC.

Residual Crnn And Its Application To Handwritten Digit String Recognition

DenseNet-CTC: An end-to-end RNN-free architecture for context-free string recognition

A Sequence Labeling Convolutional Network and Its Application to Handwritten String Recognition.

Handwritten String Recognition Based on YOLOv4 and CRNN on Different Deep Learning Frameworks

Handwritten character recognition with sequential convolutional neural network

On usage of an end-to-end deep neural architecture for handwritten digit string recognition

A New Hybrid-Parameter Recurrent Neural Network for Online Handwritten Chinese Character Recognition

Improved Handwritten Digit Recognition Using Convolutional Neural Networks (CNN)

Offline Handwritten Chinese Text Recognition with Convolutional Neural Networks

Convolutional Multi-directional Recurrent Network for Offline Handwritten Text Recognition

Low Resolution Handwritten Digit String Recognition based on Object Detection Network

Online Chinese Handwriting Recognition with Time Sequence Information

A Handwritten Digit Recognizer Using Ensemble Method

On the Use of Attention Mechanism in a Seq2Seq Based Approach for Off-Line Handwritten Digit String Recognition

Learning Spatial-Semantic Context with Fully Convolutional Recurrent Network for Online Handwritten Chinese Text Recognition

Drawing and Recognizing Chinese Characters with Recurrent Neural Network

On Study of the Reliable Fully Convolutional Networks with Tree Arranged Outputs (TAO-FCN) for Handwritten String Recognition

Intelligent character recognition using fully convolutional neural networks

Multilayer Self-Organizing Impulse Neural Network for Handwritten Digit Recognition