High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Web-Cam Based Assistive Text and Product Label Read from Hand-Held Objects for Visual Impaired


Janardhana K L , AMC EC Bangalore; Bhuvana Suganthi. D, AMC EC Bangalore


Assistive Gadgets, Visual Deficiency, Dispersion of Edge Pixels, Hand-Held Articles, Optical Character Acknowledgment (OCR), Stroke Introduction, Content Perusing, Content District Restriction


we propose a camera-based assistive content perusing structure to visually impaired persons read content marks and item bundling from hand-held articles in their day by day lives. To disconnect the article from jumbled foundations or other encompassing items in the camera view, we first propose a proficient and successful movement based technique to characterize an area of interest (ROI) in the video by requesting that the client shake the item. This strategy separates moving item district by a blend of-Gaussians-based foundation sub-footing technique. In the extricated ROI, content confinement and acknowledgment are led to gain content data. To consequently confine the content districts from the item ROI, we propose a novel content limitation calculation by learning slope elements of stroke introductions and disseminations of edge pixels in an Adaboost model. Content characters in the limited content locales are then binarized and perceived by off-the-rack optical character acknowledgment delicate product. The perceived content codes are yield to visually impaired clients in discourse. Execution of the proposed content limitation calculation is quantitatively assessed on ICDAR-2003 and ICDAR-2011 Robust Reading Datasets. Trial results show that our calculation accomplishes the condition of expressions of the human experience. The Proof of idea model is additionally assessed on a dataset gathered utilizing ten visually impaired persons to assess the viability of the framework's equipment. We investigate client between face issues and evaluate power of the calculation in extricating and perusing content from various items with complex foundations.

Other Details

Paper ID: IJSRDV4I50372
Published in: Volume : 4, Issue : 5
Publication Date: 01/08/2016
Page(s): 532-538

Article Preview

Download Article