Analysis of CNN Digit Classifier Parameters
- 26 August 2022
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2022 2nd Asian Conference on Innovation in Technology (ASIANCON)
Abstract
Convolutional Neural Networks (CNN’s) are widely being used for various image processing applications such as denoising, classification, de-hazing and super-resolution. In this paper, CNN image classifier to classify the digits is designed with the convolutional units, batch normalization units and the rectified linear units. The classification accuracy variation by changing different CNN parameters such as learning rate, convolutional filters, convolutional layers and training images is being analyzed. The accuracy saturates or degrades with the increment in number of convolutional layers. Selection of number of filters and learning rate are important hyper parameters impacting classifier accuracy.Keywords
This publication has 14 references indexed in Scilit:
- Convolutional neural networks in medical image understanding: a surveyEvolutionary Intelligence, 2021
- COVID-19 image classification using deep features and fractional-order marine predators algorithmScientific Reports, 2020
- State‐of‐art analysis of image denoising methods using convolutional neural networksIET Image Processing, 2019
- Explaining Convolutional Neural Networks using Softmax Gradient Layer-wise Relevance PropagationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2019
- A Survey of Transfer Learning for Convolutional Neural NetworksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2019
- Wider or Deeper: Revisiting the ResNet Model for Visual RecognitionPattern Recognition, 2019
- Research on image classification model based on deep convolution neural networkEURASIP Journal on Image and Video Processing, 2019
- Training Faster by Separating Modes of Variation in Batch-Normalized ModelsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2019
- A new image classification method using CNN transfer learning and web data augmentationExpert Systems with Applications, 2018
- Going deeper with convolutionsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2015