Mobile Application for Blind and Visually Impaired Travelers Improves Success Rate by 75 Percent in Locating Bus Stops in Boston Compared to Standard GPS Tools
Comparative Study of Micro-Navigation Apps Was Conducted to Localize 10 Bus Stops in Downtown and Suburban Boston
Boston, Massachusetts, United States
Field Evaluation of a Mobile App for Assisting Blind and Visually Impaired Travelers to Find Bus Stops
Summary Information
GPS-based navigation apps, while widely used, often fall short in addressing the specific needs of Blind and Visually Impaired (BVI) travelers, particularly for tasks that require precise spatial guidance, such as locating the exact position of bus stops. This study assessed the effectiveness of a signage detection mobile application called All_Aboard, designed to enhance support for BVI individuals. Conducted in Boston, Massachusetts, the evaluation focused on comparing the app's ability to guide BVI travelers accurately to bus stops against the performance of Google Maps, which relies solely on GPS-based navigation. The findings highlight the potential of All_Aboard to provide more precise and reliable support for BVI travelers in urban transit environments.
METHODOLOGY
The All_Aboard app developed in this study involves a deep neural network model where computer vision algorithms are used to learn the appearance of the bus stop signs and detects bus stop signs in real-time by matching the learnt images to the ones taken via smartphone camera and provides distance coded audio feedback to help locate the detected sign. A total of 24 BVI individuals used both the All_Aboard and Google Maps app to localize 10 bus stops each in downtown and suburban Boston, Massachusetts, to navigate as close as possible to the physical bus stop sign, starting from 30 to 50 meters (98 to 164 feet) away. The success rate of the apps was calculated as the percentage of trials where the app successfully guided participants to the vicinity of the bus stop within a measurable distance (≤25 meters [82 feet]). For the All_Aboard app, success occurs if the participant reaches close enough to the bus stop, typically within the app's functional range, as validated by residual vision or tactile feedback. Failures are recorded for instances where participants exceed the 25-meter (82 feet) threshold due to missed detections, inadequate scanning, or technical issues with the app.
FINDINGS
- The results showed that the success rate in locating the bus stop using the All_Aboard app was 75 percent higher than that of Google Maps, comparing 91 percent and 52 percent, respectively.
- The results indicated that the gap distance when using the All_Aboard app was significantly reduced, being 74.3 percent lower (mean = 1.8 meters [5.9 feet]) compared to using Google Maps alone (mean = 7 meters [23 feet]).
