Back to results
US 2026/0283891 A1Published application
System and methods to improve orientation and mobility of the visually impaired
OwnerArnav CHAPHALKARinventor
Life of this patent
Estimating the term…
Abstract
The present disclosure relates to systems and methods that integrate a cane and mobile phone technology with multi-tier artificial intelligence processing of image data to generate audible guidance and instructions for the visually impaired in a variety of practical environments.
Claims20
A system, comprising:
- a cane configured to transmit an electromagnetic signal in response to a tactile input from a user;
- a downloadable app embodied in a non-transitory computer-readable storage medium, the downloadable app comprising computer-readable program code, the computer-readable program code executable on a mobile computing device to perform operations comprising:
- capturing a digital image in response to the electromagnetic signal;
- transmitting the digital image via a wireless network to at least one server;
- receiving a text description of the digital image via the wireless network from the at least one server;
- and
- playing an audio version of the text description;
- the at least one server configured to execute operations comprising:
- generating a plurality of image captions for the digital image from a first machine learning model;
- generating a data structure from a second machine learning model, the data structure comprising one or more subsets of image captions of the plurality of image captions;
- and
- submitting a prompt comprising the data structure to a third machine learning model to obtain the text description of the digital image.
9 more elements
A method, comprising:
- transmitting an electromagnetic signal from a cane in response to a tactile input from a user;
- capturing a digital image in response to the electromagnetic signal;
- transmitting the digital image via a wireless network;
- obtaining a plurality of image captions for the digital image from a first machine learning model;
- further obtaining a data structure from a second machine learning model, the data structure comprising a first subset of image captions of the plurality of image captions, and a second subset of image captions of the plurality of image captions;
- forming a prompt comprising the data structure;
- submitting the prompt to a third machine learning model to obtain the text description of the digital image;
- receiving a text description of the digital image via the wireless network;
- and
- playing an audio version of the text description.
7 more elements
Dependencies are inferred from the claim text.