Primitive Feature-Based Optical Character Recognition of the Devanagari Script

Sharma, Richa; Mudgal, Tarun

doi:10.1007/978-981-13-0224-4_23

Cited by 6 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…al. [14] [17] present a paper on Primitive Feature-Based Optical Character Recognition of the Devanagari Script. The presented procedure consist with representation of each character image on the basic of presence and location of primitive features like vertical lines, the frequency, and location of the intersections and the frequency of intersections of character body with Shirorekha of a Devanagari character.…”

Section: Bmentioning

confidence: 99%

Classification Techniques for Handwritten Character Recognition

Sinha¹,

Sharma²,

Roshan³

2020

IJEAST

View full text Add to dashboard Cite

In this paper, we present various classification methods for printedoptical character identification (POCR) similarly handwritten optical character identification (HOCR). Observation response of various method illustrate that which scheme produce batter recognition trueness in printed optical character identification (POCR) similarly handwritten optical character identification(HOCR). This article illustrate analysis of previous paper, and also distinguish the most important once out of the diversity of superior existing classification and feature extraction techniques and we will standardize the techniques by their feature circumstances and dataset used by different authors. It bring us to the performance of the algorithms produced to the expected efficiency.Feature withdrawal supporting to examine the shape controlled in the outline. While a quantity of feature taking out and categorization techniques are accessible, other than the picking of an exceptionallysuperior; technique decides the high degree of recognition correctness. A batch of author present investigation in this field and designnovel techniques of extraction and categorization. The corepurpose of this proposed article is to re-examine these techniques, so that the group of these techniques can be comprehended.

show abstract

Section: Bmentioning

confidence: 99%

Classification Techniques for Handwritten Character Recognition

Sinha¹,

Sharma²,

Roshan³

2020

IJEAST

View full text Add to dashboard Cite

show abstract

“…Any alphabet can be divided into the features like vertical bar, intersections, shirorekha touch frequency and number of endpoints and slopes of ending curves [25]. Preprocessing is a very important stage in character recognition which contains smoothening, binarization, thinning and spur removal from the character.…”

Section: Fig 2 Devanagari Vowels and Consonantsmentioning

confidence: 99%

A Survey on Recent Methodologies in Multilingual Character Detection and Recognition

Gaikwad¹,

Nalbalwar²

2019

IJEAST

View full text Add to dashboard Cite

Content discovery and recognition has risen as a significant issue in the previous couple of years. Headways in the field of computer vision, Artificial Intelligence and some constant applications dependent on content identification and recognition has taken more consideration of researchers. Different workshops and conferences are being sorted out on global level giving further ascent to advancements in field of character detection and recognition. Multilingual character detection and recognition from video subtitles, scenes and documents is additionally getting high consideration on this subject. Lot of work has been done for English language but still the issue of handwritten Hindi character recognition isn't yet comprehended sufficient. Different optical character recognition systems perform good for English characters but the accuracy for Hindi character recognition is not up to the mark. This paper surveys different recent ongoing strategies for English and Hindi character recognition. The methodologies include recent features extraction and classification techniques. Moreover, this study includes accepted and published work by research community in recent years. This study also benefits its readers by talking about limitations of existing systems in this field.

show abstract