Deep Learning Inspired Nonlinear Classification Methodology for Handwritten Digits Recognition Using DSR Encoder
- Title
- Deep Learning Inspired Nonlinear Classification Methodology for Handwritten Digits Recognition Using DSR Encoder
- Creator
- Singh D.; Bano S.; Samanta D.; Mekala M.S.; Islam S.H.
- Description
- The overlapped handwritten digit classification is a global challenge and a significant measure to assess the network recognition ability ratio. Most efficient models have been designed based on convolutional neural networks (CNN) for effective image classification and digit identification. Subsequently, multiple CNN models have inadequate accuracy because of high degree parameter dimensions that lead to abnormal digit detection error rates and computation complexity. We propose a Deep Digit Recognition Network (DDRNet) based on Deep ConvNets to minimize the number of parameters and features to keep the model light while maximizing the accuracy with an adaptive voting (AV) scheme for digit recognition. The individual digit is identified by CNN, and uncertain digits or strings are identified by Deep Convolutional Network (DCN) with AV scheme through Voting-Weight Conditional Random Field (VWCRF) strategy. These methods originated with the YOLO algorithm. The simulations show that our DDRNet approach achieves an accuracy of 99.4% without error fluctuations, in a stable state with less than 15 epochs contrast with state-of-art approaches. Additionally, specific convolution techniques (SqueezeNet, batch normalization) and image augmentation techniques (dropout, back-propagation, and an optimum learning rate) were examined to assess the system performance based on MNIST dataset (available at: http://yann.lecun.com/exdb/mnist/). 2022, King Fahd University of Petroleum & Minerals.
- Source
- Arabian Journal for Science and Engineering, Vol-48, No. 2, pp. 1385-1397.
- Date
- 2023-01-01
- Publisher
- Institute for Ionics
- Subject
- Adaptive voting scheme; Batch normalization; DDRNet approach; MaxPooling; SqueezeNet
- Coverage
- Singh D., Department of Computer Science, CHRIST (Deemed to be University), Karnataka, Bangalore, 560029, India; Bano S., Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Andhra Pradesh, Vaddeswaram, India; Samanta D., Department of Computer Science, CHRIST (Deemed to be University), Karnataka, Bangalore, 560029, India; Mekala M.S., Department of Information and Communication Engineering, Yeungnam University, Gyeongsan, 38544, South Korea, RLRC for Autonomous Vehicle Parts and Materials Innovation, Yeungnam University, Gyeongsan, 38544, South Korea; Islam S.H., Department of Computer Science and Engineering, Indian Institute of Information Technology Kalyani, Kalyani, West Bengal, 741235, India
- Rights
- Restricted Access
- Relation
- ISSN: 2193567X
- Format
- Online
- Language
- English
- Type
- Article
Collection
Citation
Singh D.; Bano S.; Samanta D.; Mekala M.S.; Islam S.H., “Deep Learning Inspired Nonlinear Classification Methodology for Handwritten Digits Recognition Using DSR Encoder,” CHRIST (Deemed To Be University) Institutional Repository, accessed February 25, 2025, https://archives.christuniversity.in/items/show/14424.