Stereopsis method for visually impaired to identify obstacles based on distance

Balakrishnan, G. and Sainarayanan , Gopala and Nagarajan, R. and Sazali Yaacob, (2004) Stereopsis method for visually impaired to identify obstacles based on distance. In: Proceedings - Third International Conference on Image and Graphics, 18-20 Disember 2004, Hong Kong.

Full text not available from this repository.

Official URL: http://dx.doi.org/10.1109/ICIG.2004.136

Abstract

Identification of obstacles in the real time environment is idempotent for human blind to navigate freely without collision. This paper describes a methodology that is required to support the blind to discriminate obstacles based on their distance. The hardware part consists of a sunglass fitted with two mini cameras, laptop computer and stereo earphones. The two cameras capture the visual information infront of blind user. The captured images are then processed using the proposed methodology. The methodology includes image processing module, which isolate objects from the two camera images, and stereo vision module calculates distance through disparity. In the proposed algorithm disparity is obtained with less computation compared to conventional area based stereo correspondence method. Experimentations were conducted in indoor environment, and the results obtained are supportable.

Item Type:Conference Paper (UNSPECIFIED)
Uncontrolled Keywords:Camera images, Electronic travel aids (ETA), Sonification, Visual information, Error analysis, Estimation, Hardware, Image processing, Optimization, Stereo vision, Vision
Subjects:?? TA1501-1820 ??
Divisions:SCHOOL > School of Engineering and Information Technology
ID Code:3562
Deposited By:IR Admin
Deposited On:20 Feb 2012 11:42
Last Modified:10 Sep 2014 12:26

Repository Staff Only: item control page


Browse Repository
Collection
   Articles
   Book
   Speeches
   Thesis
   UMS News
Search
Quick Search

   Latest Repository

Link to other Malaysia University Institutional Repository

Malaysia University Institutional Repository