Combinational neural network using Gabor filters for the classification of handwritten digits

TL;DR

This paper introduces a hybrid neural network and k-nearest neighbors classifier utilizing Gabor filters for feature extraction, achieving faster training times on the MNIST dataset and aiming for integration into object recognition systems.

Contribution

The paper presents a novel combination of neural networks and k-NN with Gabor filters, reducing training time and improving feature extraction for handwritten digit classification.

Findings

Reduced training time significantly on MNIST dataset

Effective feature extraction with Gabor filters

Potential for integration into object recognition software

Abstract

A classification algorithm that combines the components of k-nearest neighbours and multilayer neural networks has been designed and tested. With this method the computational time required for training the dataset has been reduced substancially. Gabor filters were used for the feature extraction to ensure a better performance. This algorithm is tested with MNIST dataset and it will be integrated as a module in the object recognition software which is currently under development.