DIGI-Net: a deep convolutional neural network for multi-format digit recognition

Arun, Madakannu; Arivazhagan, S.

doi:10.1007/s00521-019-04632-9

Cited by 22 publications

(9 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The notable point is that for digit recognition of various datasets and different languages, the training of the proposed network was performed in 20 training epochs, while for character datasets of Chars74K, the training was performed in 50 training epochs. However, the training epochs of the proposed network for digit recognition experiments are fewer than other compared methods such as [72] which reported their number of training epochs.…”

Section: Discussionmentioning

confidence: 92%

Pattern Separation Network Based on the Hippocampus Activity for Handwritten Recognition

et al. 2020

View full text Add to dashboard Cite

Reaching high accuracy in handwritten character recognition is an essential challenge since it is widely used in many fields such as signature analysis and forgery detection. Recently, deep learning has demonstrated efficiency in this field. The problem with deep learning is that it uses a vast number of parameters that require a large dataset for training. To overcome this problem, an intelligent network is proposed in this study, based on the computational function of the dentate gyrus of the brain's hippocampus. The ability to separate patterns with high overlapping is a task that is referred to as the dentate gyrus. Handwritten character images have high overlapping due to various writers' styles, or even one writer's style under different conditions. Therefore, proposing a network based on the dentate gyrus' functional computation can be useful in this field. One of the prominent features of the proposed network is employing two excitation steps and two inhibition steps, augmenting the accuracy of recognizing handwritten characters. The proposed network was evaluated with six datasets of digits and characters from five languages. Experiments on all of the used datasets showed promising results. Moreover, a comparative and detailed analysis of the proposed network with other SOM-based and deep learning methods is provided. Experimental results show a significant boost in accuracy. While the character error rate (CER) was smaller than 1.85% for all the experiments, the smallest CER of 0.6% was achieved by the MNIST dataset. Moreover, in recognizing patterns with high noise, the proposed network showed satisfactory results.

show abstract

Section: Discussionmentioning

confidence: 92%

Pattern Separation Network Based on the Hippocampus Activity for Handwritten Recognition

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The proposed handwritten character recognition model is simulated using MATLAB (2018a) environment. In this research, the effectiveness of the proposed UFS-MSVM model performance is validated by comparing with benchmark models; FLM-FFNN [15], DIGI Net model [16] and context aware model [19] and adapted deep hybrid transfer model [20] on chars74K and MADbase digits datasets. In addition, the performance of the proposed UFS-MSVM model is evaluated in terms of accuracy, MCC, f-score, sensitivity and specificity.…”

Section: Simulation Resultsmentioning

confidence: 99%

“…However, the hybrid neural network model showed comparable performance in real scene dataset (camera captured text image), because it was affected from environmental noise, background complexity, and deformations. Madakannu and Selvaraj [16] introduced DIGI Net model for learning feature vector and recognizing printed font, natural images and handwritten images. In this study, DIGI Net model attained effective performance on three benchmark datasets like Chars74K, MNIST and CVL single digit datasets.…”

Section: Related Workmentioning

confidence: 99%

“…The simulation results showed that the FLM-FFNN model achieved 95% of accuracy in English hand-Madakannu and Selvaraj[16] has implemented a novel DIGI-Net CNN model for automatic feature learning, and recognizing English printed font, natural images, and handwritten images. Experimental analysis showed that the developed DIGI-Net CNN model obtained 85% of classification accuracy in handwritten character recognition on chars74K dataset.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Handwritten Character Recognition Using Unsupervised Feature Selection and Multi Support Vector Machine Classifier

2021

IJIES

View full text Add to dashboard Cite

“…This data set is used for handwritten word alteration detection experiments using pen ink analysis. It has been observed that CNNbased approaches outperform other approaches for variety of classification tasks, like medical image classification [37], [38], plant disease classification [39], and optical character recognition (OCR) [40]- [42]. Training a CNN model requires huge amount of annotated data.…”

Section: Volume X Yyyymentioning

confidence: 99%

Generation of Synthetic Data for Handwritten Word Alteration Detection

2021

View full text Add to dashboard Cite

Fraudsters often alter handwritten contents in a document in order to achieve illicit purposes. At times, this may result in financial and mental loss to an individual or an organization. Hence, ink analysis is necessary to identify such an alteration. Convolution Neural Network (CNN) can be used to identify such cases of alteration, as CNN has emerged as a monumental success in the field of computer vision for varieties of classification tasks. But, CNN requires large amount of labeled data for training. Hence, there is a need to generate a large dataset for the experiments relating to handwritten word alteration detection. Collection, digitization, and cropping of a large number of altered and unaltered handwritten words are tedious and time consuming. To overcome such an issue, an approach for synthetic word data generation is presented in this paper for handwritten word alteration detection experiments. This scheme is designed in such a way that the synthetically generated words are very similar to the original ones. In order to achieve this, handwritten character data set is prepared using 10 blue and 10 black pens. These handwritten characters are used for creating synthetic word alteration data set. The presented approach uses relatively less number of handwritten character images to create a huge word alteration data set. Further, deep learning models are trained on the synthetically generated data set for word alteration detection.

show abstract

DIGI-Net: a deep convolutional neural network for multi-format digit recognition

Cited by 22 publications

References 20 publications

Pattern Separation Network Based on the Hippocampus Activity for Handwritten Recognition

Pattern Separation Network Based on the Hippocampus Activity for Handwritten Recognition

Handwritten Character Recognition Using Unsupervised Feature Selection and Multi Support Vector Machine Classifier

Generation of Synthetic Data for Handwritten Word Alteration Detection

Contact Info

Product

Resources

About