Open Access Open Access  Restricted Access Subscription or Fee Access

A Navigation Assistant for the Visually Impaired by Means of Robust and Streamlined Object Recognition

V. Kosaraju, R. Jain

Abstract


The issue of mobility is considered among the foremost of concerns for visually impaired individuals. Scale-Invariant Feature Transform (SIFT) and Content-Based Image Retrieval (CBIR) are techniques often used for object recognition and image retrieval in the field of computer vision. This research develops a Navigation Assistant that aids the visually impaired with mobility in a known environment such as a household, by combining the principles behind SIFT and CBIR. Furthermore, this research tests several techniques developed by the authors to improve the Navigation Assistant. Using a video stream that was obtained from a small camera that the user carries, key objects in the household were identified using the software developed. The control Navigation Assistant, developed without the authors’ enhancements, had an object identification accuracy of 45.555% and an average runtime of 3.427 seconds. The experimental Navigation Assistant, which included the techniques of Color Segmentation, Difference of Images, and Contrast Manipulation created by the authors, had an accuracy of 75.555% and an average runtime of 1.820 seconds. Thus, the Navigation Assistant created using the enhancements was markedly improved in both accuracy and speed from the control Assistant, and is therefore applicable to real-life usage by visually impaired individuals.

Full Text: PDF


Disclaimer/Regarding indexing issue:

We have provided the online access of all issues and papers to the indexing agencies (as given on journal web site). It’s depend on indexing agencies when, how and what manner they can index or not. Hence, we like to inform that on the basis of earlier indexing, we can’t predict the today or future indexing policy of third party (i.e. indexing agencies) as they have right to discontinue any journal at any time without prior information to the journal. So, please neither sends any question nor expects any answer from us on the behalf of third party i.e. indexing agencies.Hence, we will not issue any certificate or letter for indexing issue. Our role is just to provide the online access to them. So we do properly this and one can visit indexing agencies website to get the authentic information.