A Novel Approach to Detect and localize the Text inNatural scene imag
| Author(s) | : | Prof. B. K. Sarojini, Mr. Mahesh Mahadar |
| Institution | : | Department of Electronics and Communication Engineering, Basaveshwara Engineering College, Bagalkot India |
| Published In | : | Vol. 2, Issue 5 — May 2015 |
| Page No. | : | 1386-1392 |
| Domain | : | Engineering |
| Type | : | Research Paper |
| ISSN (Online) | : | 2348-4470 |
| ISSN (Print) | : | 2348-6406 |
The informationavailablein the naturalscene provides very importantcluesformanyimagebased application. So thedetectionandlocalization of textfromnaturalscene imageisimportant taskforcontent basedimageanalysis. Inthispaper,wedevelop anovelapproach todetectand localizetextsin naturalsceneimages. Thisproblem ischallengingduetothefact that text has different in size, alignment, orientation, style,complexbackground of imagesaswellasimage having thelowcontrast. Initially,theRGBimageisconvertedintograyscaleimage andapplied to thelocal binarizationalgorithmfor thesegmentation. Afterthat,HaarWaveletTransform(DWT)isused which isfastestas compared withallwaveletsbecauseitsfiltercoefficientsareeither 1 or -1. It decomposeimageintofoursub band, oneisaverage andotherthreearedetailedwhichhelps tofind outtheapproximatelyconfidentregion.To filter out the non-text component, aconditional randomfield(CRF) is usedwhich appliedonconfidentregionimage. Finally,by usingsome predefinedcondition,thetextis obtainedin boundingbox.
Prof. B. K. Sarojini, Mr. Mahesh Mahadar, “A Novel Approach to Detect and localize the Text inNatural scene imag”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 2, Issue 5, pp. 1386-1392, May 2015.
