Detecting and Localizing Color Text in Natural Scene Images Using Region Based & Connected Component Method

Journal Title: International Journal of Modern Engineering Research (IJMER) - Year 2013, Vol 3, Issue 1

Abstract

 Abstract: Large amounts of information are embedded in natural scenes which are often required to be automatically recognized and processed. This requires automatic detection, segmentation and recognition of visual text entities in natural scene images. In this paper, we present a hybrid approach to detect color texts in natural scene images. The approaches used in this project are region based and connected component based approach. A text region detector is designed to estimate the probabilities of text position and scale, which helps to segment candidate text components with an efficient local binarization algorithm. To combine unary component properties and binary contextual component relationships, a conditional random field (CRF) model with supervised parameter learning is proposed. Finally, text components are grouped into text lines/words with a learning-based energy minimization method. In our proposed system, a selective metricbased clustering is used to extract textual information in real-world images, thus enabling the processing of character segmentation into individual components to increase final recognition rates. This project is evaluated on natural scene image dataset.

Authors and Affiliations

Mohanabharathi. R

Keywords

Related Articles

 A ZVS Interleaved Boost AC/DC Converter Using Super Capacitor Power for Hybrid Electrical Vehicles

 This paper presents the supercapacitor power management method for the Hybrid Electricals Vehicles.A simple zero-voltage switching(ZVS) interleaved boost power factor correction(PFC) AC/DC converter is used to char...

 Performance Evaluation of Nine Level Modified CHB Multilevel Inverter for Various PWM Strategies

 In this paper nine level Modified Cascaded H-Bridge Multilevel Inverter (CHB-MLI) is analyzed for the various multi-carrier Pulse Width Modulation strategies. For the same nine level inverter output this particular...

 On Gr-Separation Axioms

In the present paper , we introduce and study the concept of gr- Ti- space (for i =0,1,2) and obtain the characterization of gr –regular space , gr- normal space by using the notion of gr-open sets. Further, some...

 A Total Ergonomics Model for integration of health, safety and work to improve productivity of thermal Power Plants

 This paper presents a outline for development of integrated health, safety and work in complex critical systems of thermal power plant. In this paper operation and maintenance department of a thermal power plant wa...

 Wavelet Analysis of Vibration Signature of a Bevel Gear Box in a Stand of Steel Rolling Mill of a Steel Plant

 In manufacturing industry Gear boxes are one of the important components in power transmission. Consequently any damage in the gearbox can lead to unprecedented downtime. In machines the gear trains vibrate due...

Download PDF file
  • EP ID EP130564
  • DOI -
  • Views 115
  • Downloads 0

How To Cite

Mohanabharathi. R (2013).  Detecting and Localizing Color Text in Natural Scene Images Using Region Based & Connected Component Method. International Journal of Modern Engineering Research (IJMER), 3(1), 331-335. https://europub.co.uk/articles/-A-130564