AKHCRNet: Bengali Handwritten Character Recognition Using Deep Learning

• URL: http://arxiv.org/abs/2008.12995v3
• Date: Sun, 20 Sep 2020 22:48:57 GMT
• Title: AKHCRNet: Bengali Handwritten Character Recognition Using Deep Learning
• Authors: Akash Roy
• Abstract summary: I propose a state of the art deep neural architectural solution for handwritten character recognition for Bengali alphabets, compound characters, and numerical digits. I propose an HCR network that is trained from the scratch on Bengali characters without the "Ensemble Learning" that can outperform previous architectures.
• Score: 0.228438857884398
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: I propose a state of the art deep neural architectural solution for handwritten character recognition for Bengali alphabets, compound characters as well as numerical digits that achieves state-of-the-art accuracy 96.8% in just 11 epochs. Similar work has been done before by Chatterjee, Swagato, et al. but they achieved 96.12% accuracy in about 47 epochs. The deep neural architecture used in that paper was fairly large considering the inclusion of the weights of the ResNet 50 model which is a 50 layer Residual Network. This proposed model achieves higher accuracy as compared to any previous work & in a little number of epochs. ResNet50 is a good model trained on the ImageNet dataset, but I propose an HCR network that is trained from the scratch on Bengali characters without the "Ensemble Learning" that can outperform previous architectures.

Related papers

• Classification of Non-native Handwritten Characters Using Convolutional Neural Network [0.0]
  The classification of English characters written by non-native users is performed by proposing a custom-tailored CNN model. We train this CNN with a new dataset called the handwritten isolated English character dataset. The proposed model with five convolutional layers and one hidden layer outperforms state-of-the-art models in terms of character recognition accuracy.
  arXiv  Detail & Related papers  (2024-06-06T21:08:07Z)
• Data Engineering for Scaling Language Models to 128K Context [98.41554785106902]
  We study the continual pretraining recipe for scaling language models' context lengths to 128K. We find that naively upsampling longer data on certain domains like books, a common practice of existing work, gives suboptimal performance. Our recipe outperforms strong open-source long-context models and closes the gap to frontier models like GPT-4 128K.
  arXiv  Detail & Related papers  (2024-02-15T18:19:16Z)
• Efficient Self-supervised Learning with Contextualized Target Representations for Vision, Speech and Language [60.12197397018094]
  data2vec is a learning objective that generalizes across several modalities. We do not encode masked tokens, use a fast convolutional decoder and amortize the effort to build teacher representations. Experiments on ImageNet-1K image classification show that data2vec 2.0 matches the accuracy of Masked Autoencoders in 16.4x lower pre-training time.
  arXiv  Detail & Related papers  (2022-12-14T22:13:11Z)
• Kurdish Handwritten Character Recognition using Deep Learning Techniques [26.23274417985375]
  This paper attempts to design and develop a model that can recognize handwritten characters for Kurdish alphabets using deep learning techniques. A comprehensive dataset was created for handwritten Kurdish characters, which contains more than 40 thousand images. The tested results reported a 96% accuracy rate, and training accuracy reported a 97% accuracy rate.
  arXiv  Detail & Related papers  (2022-10-18T16:48:28Z)
• Efficient approach of using CNN based pretrained model in Bangla handwritten digit recognition [0.0]
  Handwritten digit recognition is essential for numerous applications in various industries. Due to the complexity of Bengali writing in terms of variety in shape, size, and writing style, researchers did not get better accuracy usingSupervised machine learning algorithms to date. We propose a novel CNN-based pre-trained handwritten digit recognition model which includes Resnet-50, Inception-v3, and EfficientNetB0 on NumtaDB dataset of 17 thousand instances with 10 classes.
  arXiv  Detail & Related papers  (2022-09-19T15:58:53Z)
• Lexically Aware Semi-Supervised Learning for OCR Post-Correction [90.54336622024299]
  Much of the existing linguistic data in many languages of the world is locked away in non-digitized books and documents. Previous work has demonstrated the utility of neural post-correction methods on recognition of less-well-resourced languages. We present a semi-supervised learning method that makes it possible to utilize raw images to improve performance.
  arXiv  Detail & Related papers  (2021-11-04T04:39:02Z)
• HCR-Net: A deep learning based script independent handwritten character recognition network [5.8067395321424975]
  Handwritten character recognition (HCR) remains a challenging pattern recognition problem despite decades of research. We have proposed a script independent deep learning network for HCR research, called HCR-Net, that sets a new research direction for the field.
  arXiv  Detail & Related papers  (2021-08-15T05:48:07Z)
• Effective Model Sparsification by Scheduled Grow-and-Prune Methods [73.03533268740605]
  We propose a novel scheduled grow-and-prune (GaP) methodology without pre-training the dense models. Experiments have shown that such models can match or beat the quality of highly optimized dense models at 80% sparsity on a variety of tasks.
  arXiv  Detail & Related papers  (2021-06-18T01:03:13Z)
• Post-training deep neural network pruning via layer-wise calibration [70.65691136625514]
  We propose a data-free extension of the approach for computer vision models based on automatically-generated synthetic fractal images. When using real data, we are able to get a ResNet50 model on ImageNet with 65% sparsity rate in 8-bit precision in a post-training setting.
  arXiv  Detail & Related papers  (2021-04-30T14:20:51Z)
• Improved Residual Networks for Image and Video Recognition [98.10703825716142]
  Residual networks (ResNets) represent a powerful type of convolutional neural network (CNN) architecture. We show consistent improvements in accuracy and learning convergence over the baseline. Our proposed approach allows us to train extremely deep networks, while the baseline shows severe optimization issues.
  arXiv  Detail & Related papers  (2020-04-10T11:09:50Z)
• Performance Evaluation of Advanced Deep Learning Architectures for Offline Handwritten Character Recognition [0.6445605125467573]
  The system utilizes advanced multilayer deep neural network by collecting features from raw pixel values. Two state of the art deep learning architectures were used which includes Caffe AlexNet and GoogleNet models in NVIDIA DIGITS. The accuracy level achieved with AlexNet was 77.77% and 88.89% with Google Net.
  arXiv  Detail & Related papers  (2020-03-15T11:17:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.