DocumentCode
3774480
Title
Detection of multioriented texts in natural scene images
Author
Jiffy Jacob;Abraham Thomas
Author_Institution
Dept. of Applied Electronics and Instrumentation Engineering, Rajagiri School of Engineering and Technology, Kochi, India
fYear
2015
Firstpage
625
Lastpage
628
Abstract
Detecting texts of multiorientation has been studied for long, yet it remains a challenging task because of the complex background and diversity of texts. For the results to be accurate, it will make use of a classification scheme using random forest classifier. Text detection becomes more challenging if the input image is a blurred one. So we make use of a blind image deblurring method for image deblurring and thus text can be detected more effectively in the deblurred image.
Keywords
"Image edge detection","Feature extraction","Deconvolution","Image restoration","Filtering","Detectors","Histograms"
Publisher
ieee
Conference_Titel
Control, Instrumentation, Communication and Computational Technologies (ICCICCT), 2015 International Conference on
Type
conf
DOI
10.1109/ICCICCT.2015.7475353
Filename
7475353
Link To Document