Review of Optical Devanagari Character Recognition Techniques

Singh, Sukhjinder; Garg, Naresh Kumar

doi:10.1007/978-981-15-5400-1_11

Cited by 15 publications

(9 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some have provided an examination of the many methods investigated for Devanagari word and difficult-toread character recognition. Additionally provided are insights into several OCR challenges [3]. The detection of restricted handwritten Devanagari characters has been studied.…”

Section: Related Workmentioning

confidence: 99%

Evaluation of Deep Learning CNN Model for Recognition of Devanagari Digit

Bhosle¹,

Musande

2023

AIA

View full text Add to dashboard Cite

Devanagari character and digit recognition is a difficult undertaking because writing style depends on a person's traits and differs from person to person. We get more precise results in digit recognition thanks to deep learning convolutional neural networks (CNN), which function similarly to the human brain. In this study, the CNN method was put into practise and contrasted with the feed-forward neural network and random forest approaches. In comparison to previous methods, CNN has reportedly provided an accuracy rating of up to 99.2%. CNN is effective with both organised and unstructured data, including pictures, video, and audio.

show abstract

Section: Related Workmentioning

confidence: 99%

Evaluation of Deep Learning CNN Model for Recognition of Devanagari Digit

Bhosle¹,

Musande

2023

AIA

View full text Add to dashboard Cite

show abstract

“…Generally, the workflow consists of three main parts. Firstly, the technique of object detection [18] together with optical character recognition [19] are applied to transform the raw image into discrete element representations as shown in Fig. 6.a.…”

Section: Figureqa Datasetmentioning

confidence: 99%

Graph-based Heuristic Search for Module Selection Procedure in Neural Module Network

Nakayama

2020

Preprint

View full text Add to dashboard Cite

Neural Module Network (NMN) is a machine learning model for solving the visual question answering tasks. NMN uses programs to encode modules' structures, and its modularized architecture enables it to solve logical problems more reasonably. However, because of the nondifferentiable procedure of module selection, NMN is hard to be trained end-to-end. To overcome this problem, existing work either included ground-truth program into training data or applied reinforcement learning to explore the program. However, both of these methods still have weaknesses. In consideration of this, we proposed a new learning framework for NMN. Graph-based Heuristic Search is the algorithm we proposed to discover the optimal program through a heuristic search on the data structure named Program Graph. Our experiments on FigureQA and CLEVR dataset show that our methods can realize the training of NMN without ground-truth programs and achieve superior efficiency over existing reinforcement learning methods in program exploration.

show abstract

“…It is one of the important steps for an optical character recognition (OCR) system [2]. Further, separation of text and non-text regions are useful for spotting information about symbols, words, logos, graphics, scripts, pictures, tabular structures, etc [3].…”

Section: Introductionmentioning

confidence: 99%