Automatic Text Detection and Classification in Natural Images
C.P. Chaithanya1, N. Manohar2, Ajay Bazil Issac3
1C.P. Chaithanya, Department of Computer Science, Amrita School of Arts and Sciences, Amrita Vishwa Vidhyapeetham, Mysuru Campus, (Karnataka), India.
2N. Manohar, Department of Computer Science, Amrita School of Arts and Sciences, Amrita Vishwa Vidhyapeetham, Mysuru Campus, (Karnataka), India.
3Ajay Bazil Issac, Department of Computer Science, Amrita School of Arts and Sciences, Amrita Vishwa Vidhyapeetham, Mysuru Campus, (Karnataka), India.
Manuscript received on 23 April 2019 | Revised Manuscript received on 02 May 2019 | Manuscript Published on 08 May 2019 | PP: 176-180 | Volume-7 Issue-5S3 February 2019 | Retrieval Number: E11330275S19/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Text detection is the method of locating areas in a picture wherever, text is present. Text detection and classification in natural pictures is very important for several computer vision applications like optical character recognition, distinguish between human and machine inputs and spam removal. Currently the challenge in text identifying is to detect the text in natural pictures due to many factors like, low-quality image, unclear words, typical font, image having a lot of color stroke than the background color, blurred pictures due to some natural problems like rain, sunny, snow, etc. The main aim of this work is to identify and classify the text in natural pictures. Here system detects the text and finds the connected regions, chainthem together in their relative position. Uses a text classification engine to filter chains with low classification confidence scores.
Keywords: CNN, OCR, Extraction.
Scope of the Article: Image Security