June 2020
Volume 61, Issue 7
Free
ARVO Annual Meeting Abstract  |   June 2020
Effect of peripheral prism configurations on pedestrian collision detection while walking
Author Affiliations & Notes
  • Nish Mohith Kurukuti
    Ophthalmology, Harvard Medical School, Boston, Massachusetts, United States
  • Kenny Tang
    Bioengineering, Northeastern University, Boston, Massachusetts, United States
    Ophthalmology, Harvard Medical School, Boston, Massachusetts, United States
  • Jae-Hyun Jung
    Ophthalmology, Harvard Medical School, Boston, Massachusetts, United States
  • Eli Peli
    Ophthalmology, Harvard Medical School, Boston, Massachusetts, United States
  • Footnotes
    Commercial Relationships   Nish Mohith Kurukuti, None; Kenny Tang, None; Jae-Hyun Jung, None; Eli Peli, Chadwick Opticals (P)
  • Footnotes
    Support  R01EY023385
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 2771. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Nish Mohith Kurukuti, Kenny Tang, Jae-Hyun Jung, Eli Peli; Effect of peripheral prism configurations on pedestrian collision detection while walking. Invest. Ophthalmol. Vis. Sci. 2020;61(7):2771.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Patients with homonymous hemianopia use superior and inferior peripheral (Peli) prisms to expand their visual field. Peripheral prisms are fitted with either horizontal or oblique (designed for driving) prism segments. Combinations of prism segments (configurations) can be used to detect potentially colliding pedestrians walking on their blind side. To find the best prism configuration for detecting such collisions, we used a geometric model to simulate collision scenarios and analyzed four configurations of 57Δ prisms: Horizontal in superior and inferior (HH), Oblique in superior and inferior (OO), Oblique in superior and Horizontal in inferior (OH), Horizontal in superior and Oblique in inferior (HO) (Fig.1).

Methods : The geometric model’s collision scenarios included 0.6m diameter pedestrians of 1-1.8m heights, approaching a collision point at different eccentricities (5-30°) and speeds (0.7-2m/s). A 0.6m diameter patient of 1-1.8m heights, looking at different viewing distances (3m ahead on the ground to straight ahead), approached the collision point at 1.4m/s. The entire collision scenarios take 6 seconds, during which the patient is wearing different peripheral prism configurations. The horizontal and oblique prism segments had a 20° vertical field of view, with horizontal prism starting from 20° in the superior/ inferior periphery and the oblique prism field of view starting from the horizontal midline (Fig. 1). We computed the collision detection rate with each configuration within 6s (total detections) and 3s (safe detections).

Results : The OO configuration resulted in the best collision detections rate (99.8%), followed by OH (95.7%), HO (84.5%) and HH (42.6%) configurations in safe detections (Fig2). There was no large difference in the collision detection rate among configurations in total detections.

Conclusions : The OO configuration is better at detecting colliding pedestrians within a safe time.

This is a 2020 ARVO Annual Meeting abstract.

 

Figure 1: Peripheral prism configurations vertical coverage as a function of distance, from the patient fixating at the bold x. A video illustration can be found in the following link: https://youtu.be/YaWdkDYKJA8.

Figure 1: Peripheral prism configurations vertical coverage as a function of distance, from the patient fixating at the bold x. A video illustration can be found in the following link: https://youtu.be/YaWdkDYKJA8.

 

Figure 2: Pedestrian collision detection rate of peripheral prism configurations.

Figure 2: Pedestrian collision detection rate of peripheral prism configurations.

×
×

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.

×