June 2013
Volume 54, Issue 15
Free
ARVO Annual Meeting Abstract  |   June 2013
An Object Localization and Feedback System
Author Affiliations & Notes
  • Nii Tete Mante
    Department of Biomedical Engineering, University of Southern California, Los Angeles, CA
    Department of Computer Science, University of Southern California, Los Angeles, CA
  • Gerard Medioni
    Department of Computer Science, University of Southern California, Los Angeles, CA
  • Armand Tanguay
    Department of Biomedical Engineering, University of Southern California, Los Angeles, CA
    Department of Electrical Engineering, University of Southern California, Los Angeles, CA
  • Thang Dinh
    Department of Computer Science, University of Southern California, Los Angeles, CA
  • Furkan Sahin
    Department of Electrical Engineering, University of Southern California, Los Angeles, CA
  • Patrick Nasiatka
    Department of Electrical Engineering, University of Southern California, Los Angeles, CA
  • James Weiland
    Department of Biomedical Engineering, University of Southern California, Los Angeles, CA
    Department of Opthamology, University of Southern California, Los Angeles, CA
  • Footnotes
    Commercial Relationships Nii Tete Mante, None; Gerard Medioni, None; Armand Tanguay, University of Southern California (P); Thang Dinh, None; Furkan Sahin, None; Patrick Nasiatka, None; James Weiland, Second Sight Medical Products, Inc. (F)
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science June 2013, Vol.54, 2775. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Nii Tete Mante, Gerard Medioni, Armand Tanguay, Thang Dinh, Furkan Sahin, Patrick Nasiatka, James Weiland; An Object Localization and Feedback System. Invest. Ophthalmol. Vis. Sci. 2013;54(15):2775.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: The goal of the project was to create a system which assists blind/low vision subjects with reaching and grasping for objects. This system is referred to as the 'Object Localization and Feedback System' (OLFS). The OLFS aims to enhance the autonomy of blind patients, by utilizing advanced algorithms and feedback mechanisms for daily tasks.

Methods: The OLFS utilized a 100° wide-angle camera (Tanguay, Sahin, Nasiatka), central processing unit (CPU) and bone conduction headphones. Computer Vision algorithms (Context Tracker, Medioni and Dinh) on the CPU parsed, and processed the camera input to determine object(s) position. Bone conduction headphones produced auditory feedback for the blind user based on object position. For example, if the object is positioned left to the camera's center of vision (COV), the computer would say “Left” to the test subject. Once the object was centralized within the users/cameras center of vision, test subjects were asked to reach out and touch the object. Two blind test subjects evaluated the device. Subjects were trained to use the system initially. Afterwards, autonomous testing was carried out. In both stages, subjects were asked to reach and grasp for an object with the aid of the OLF system. If test subjects were not able to grasp the object within 45 seconds, the trial was stopped and considered unsuccessful.

Results: In ten object localization trials, patient RS-2 was able to grasp the desired object 6/10 times. The average time to grasp the object for the 6 successful trials was 15.43 seconds. Patient NM was able to grasp the object 10/10 times, and average the time to grasp the object was 13.95 seconds. Both patients were given a System Usability Scale (SUS) test. Patients RS-2 and NM gave the OLFS system a score of 87.5% and 67.5%, respectively. A 100% score symbolizes a perfectly usable system (as judged by the subject). Object tracking path data within the FOV was also recorded for each test and each test subject.

Conclusions: An initial system for assisting the blind in reaching and grasping for objects was successfully designed, implemented and tested. Test subjects were able to successfully grasp the object. Future improvements to the system will include enhanced robustness of vision algorithms, and new implementations of physical feedback. These two improvements will act to make the use of the system more intuitive for the patient.

Keywords: 584 low vision  
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×