MSER Based Object Character Recognition Technique

Abstract

Recognition of detected text from any image is challenging but it is essential to do so. In this paper, we propose an effective and accurate algorithm namely advanced maximally stable extremal regions (msers) as main character candidates and these character candidates are filtered by stroke width variation for removing regions where the stroke width exhibits too much variation. For the detection of text regions, firstly some pre-processing is applied to the natural image and then after detecting msers, an intersection of canny edge and mser region is produced to locate regions that are even more likely to belong to text. Finally, to select the text region we take it as an input of a novel optical character recognition (ocr) technique so that the text is usable. The evaluation result substantiates 70% accuracy.

Authors and Affiliations

Bhaskar B, Manjushree R M, Meeth Kumar R Kantesaria, Nagendra S

Keywords

Related Articles

A Review: Forgery Image Detection in Forensics

The proposed system investigate changed spaces, spoke to by picture illuminant maps to propose a techniques for selecting complementary forms of characterizing visual properties for an effective and automated detection...

A Review Paper on Cryptography for Data Security

Cryptography is the technique with which we can convert the message into secret text which is not understandable to unauthorized users. It is the technique for hiding data and information from unauthorized users. When w...

Online Food Order System

Nowadays people don’t have much time to spend in restaurant by just there and waiting for the waiter to take their order. Many customer visits the restaurant in their lunch break and recess so they have limited time to...

Study of Xylanolytic Microorganisms by using Xylan Extracted from Natural Waste

The novel approach used in current research is use of extracted xylan from natural wastes like wheat bran and Pistia sp for screening of xylanolytic microorganisms. Xylan recovered from extraction of wheat bran was 22.2...

slugFeasibility Evaluation of Symmetric Key Encryption Techniques for Wireless Channel and Disk Storage

Network security has become very important and challenging issue in this world. Cryptography is a technique that provides security for the sender and receiver to exchange their information over a network. There are basi...

Download PDF file
  • EP ID EP24688
  • DOI -
  • Views 405
  • Downloads 14

How To Cite

Bhaskar B, Manjushree R M, Meeth Kumar R Kantesaria, Nagendra S (2017). MSER Based Object Character Recognition Technique. International Journal for Research in Applied Science and Engineering Technology (IJRASET), 5(6), -. https://europub.co.uk/articles/-A-24688