Vision-Assist Device Confidence Feedback for Object Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Blind or visually impaired individuals face difficulties navigating their environment due to inaccuracies in computer-based vision systems, leading to a lack of trust in the information provided, as these systems often incorrectly identify objects.
Innovation Solution
A vision-assist device equipped with image sensors, a processor, and an auditory system that generates image data, performs object recognition, and provides an auditory message indicating the detected object's classification along with a confidence value, allowing users to assess the reliability of the information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If computer-based vision systems detect objects to assist blind or visually impaired persons, then navigation assistance is provided, but object classification accuracy deteriorates leading to incorrect identification
Solution Approach 1:
The system provides feedback to the user about the confidence level of object detection by announcing confidence values (e.g., 'highly confident,' 'moderately confident,' 'slightly confident') along with object classifications. This feedback mechanism allows users to assess the reliability of detected objects and make informed decisions about navigation, resolving the contradiction between providing navigation assistance and maintaining classification accuracy.
2Loss of information
If object detection is performed to provide navigation information, then useful information is provided, but trust in the system deteriorates due to incorrect object identification
Solution Approach 1:
The system maintains reliability and user trust by providing feedback about detection confidence levels. By announcing whether the system is highly, moderately, or slightly confident about object classifications, users can evaluate the trustworthiness of the information and understand the system's limitations, preventing complete loss of trust even when detections are uncertain.
3Reliability
If confidence values are provided with object classifications, then user trust is improved, but device complexity increases
Solution Approach 1:
The system manages device complexity by changing parameters in the information presentation layer rather than the detection layer. It processes confidence values through predefined thresholds and converts them into simple verbal descriptors (e.g., 'highly confident,' 'moderately confident,' 'slightly confident'), allowing trust improvement without significantly increasing overall system complexity.
Data Source
AI summary
Vision-assist devices and methods are disclosed. In one embodiment, a vision-assist device includes an image sensor for generating image data corresponding to a scene, a processor, and an auditory device. The processor is programmed to receive the image data from the image sensor, perform object recognition on the image data to determine a classification of a detected object that is present within the scene, and determine a confidence value with respect to the classification of the detected object. The confidence value is based on a confidence that the classification of the detected object matches an actual classification of the detected object. The processor further generates an auditory signal based on the confidence value. The audio device receives the auditory signal from the processor and produces an auditory message from the auditory signal. The auditory message is indicative of the classification of the detected object and the confidence value.