Pruning Distorted Images in MNIST Handwritten Digits

TL;DR

This paper introduces a two-stage deep learning method to improve handwritten digit recognition on MNIST by filtering out distorted images, resulting in over 99.5% accuracy.

Contribution

The study presents a novel filtering approach that enhances MNIST digit classification accuracy by removing distorted images before retraining the model.

Findings

Achieved over 99.5% accuracy on MNIST test set.

Filtering distorted images improves model confidence and reduces errors.

Method effectively mitigates issues of underfitting and overfitting.

Abstract

Recognizing handwritten digits is a challenging task primarily due to the diversity of writing styles and the presence of noisy images. The widely used MNIST dataset, which is commonly employed as a benchmark for this task, includes distorted digits with irregular shapes, incomplete strokes, and varying skew in both the training and testing datasets. Consequently, these factors contribute to reduced accuracy in digit recognition. To overcome this challenge, we propose a two-stage deep learning approach. In the first stage, we create a simple neural network to identify distorted digits within the training set. This model serves to detect and filter out such distorted and ambiguous images. In the second stage, we exclude these identified images from the training dataset and proceed to retrain the model using the filtered dataset. This process aims to improve the classification accuracy and confidence levels while mitigating issues of underfitting and overfitting. Our experimental results demonstrate the effectiveness of the proposed approach, achieving an accuracy rate of over 99.5% on the testing dataset. This significant improvement showcases the potential of our method in enhancing digit classification accuracy. In our future work, we intend to explore the scalability of this approach and investigate techniques to further enhance accuracy by reducing the size of the training data.