United States of America as Represented by the Secretary of the Air Force

The University of Arizona

The Uab Research Foundation (the University of Alabama at Birmingham)

Kavan Hazeli

Bahram Jalali

Janet Roveda

Joseph Stephen Indeck

Jesus O Mares

A life-assisting system to enhance the quality of life for people with visual impairment. The system integrates state-of-the-art sensing and sensory stimulation with light detection and ranging (LiDAR), machine learning, and advanced haptic navigation to provide the visually impaired population with a real-time haptic map of the three-dimensional environment and auditory descriptions of objects within that environment. By allowing the visually impaired population to sense objects in the surrounding environment through haptic and auditory feedback, the disclosed system has the potential to promote individual independence, reduce anxiety and stress, facilitate access to educational and employment opportunities, and reduce social and economic gaps. To provide those benefits, the system uses multi-modal learning to convert higher dimensional (e.g., four-dimensional) scene data into lower dimensional (e.g., one-dimensional) auditory data and (e.g., two- or three-dimensional) haptic data.

Converting spatial information to haptic and auditory feedback

An image processing system receives a digital image and analyzes the digital image to determine a resolution limit, referred to herein as 'feature resolution,' for measuring a metric for features of the image within an acceptable margin of error. Specifically, the system segments a digital image and calculates the error associated with the segmented data when features within a certain range a measured metric (e.g., size range) are removed from the segmented data. This analysis can be repeatedly performed with different cutoff values for the metric until at least a threshold amount of error is reached, thereby indicating a resolution limit at the boundary of an acceptable amount of error.

Systems and methods for determining feature resolution of image data

Growing community of inventors

Tucson, AZ, United States of America

Kavan Hazeli