Texture-based Real-Time Character Extraction and Recognition in Natural Images
Arun Kumar Singh

Arun Kumar Singh, Asst. Professor, College of Computing and Informatics, Saudi Electronic University, Kingdom of Saudi Arabia-KSA
Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 3302-3306 | Volume-8 Issue-8, June 2019 | Retrieval Number: H6514068819/19©BEIESP
Open Access | Ethics and 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: Extraction and Recognition of text from any natural unstructured image is challenging. In this paper, an algorithm has been proposed which provides an efficient extraction of the text from the images for translation purposes. Optical character recognition (OCR) has been used to recognize the extracted information from images with the help of Prewitt operator. Pre-processing of images is carried out to enhance the desired information. Whereas noise and artefacts are removed from image using various filters to get optimized and quick results. From the experimental results, it can be concluded that the developed OCR system shows 100% accurate results in 5 scenarios out of 9 whilein the other cases the obtained results are near to desired results
Keyword: OCR, Detection, Translation, Prewitt operator, Noise and Artefact removal, Recognition.
Scope of the Article: Natural Language Processing and Machine Translation.