Visual Exploration Assistance for Scotoma via Shape Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for assisting visual exploration in individuals with retinal pathologies like age-related macular degeneration (ARMD) and Stargardt's disease are ineffective in optimizing reading performance due to lack of consideration for the interaction between gaze position, scotoma size, and text entities, leading to discomfort and reduced reading efficiency.
Innovation Solution
A method that uses shape recognition software to identify objects in a digital image, determines a blind zone corresponding to the scotoma, and applies image processing to enhance object zones, optimizing their display to compensate for vision limitations, while allowing users to control the enhancement and awareness of their scotoma for improved ocular movements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the system only processes pixels of the image zone to displace information masked by the scotoma, then the image processing is simple, but the interaction between gaze position, scotoma size, and text entities is not optimized, leading to reduced reading performance
Solution Approach 1:
The patent segments the image into multiple zones (object zone, blind zone, vision zone) and processes each zone differently. The object zone is enhanced and reproduced in the vision zone, while the blind zone is identified and compensated. This segmentation allows complex processing to be applied selectively where needed, improving reading performance without unnecessarily increasing overall system complexity.
Solution Approach 2:
The patent applies different processing qualities to different regions of the image. The object zone receives enhanced processing (enlargement, repositioning) while other areas remain unchanged. This local quality approach optimizes reading performance for the critical areas affected by the scotoma without wasting computational resources on areas that don't require enhancement.
2Device complexity
If the system does not propose slaving to the gaze taking account of oculomotor events, then the system is simpler to implement, but the ocular movements of individuals afflicted with retinal pathology are not optimized, reducing comfort and reading performance
Solution Approach 1:
The patent incorporates feedback mechanisms that monitor gaze position and oculomotor events, using this information to dynamically adjust the image processing. The system continuously adapts the enhancement based on real-time gaze data, optimizing reading comfort and performance while maintaining manageable system complexity through targeted feedback loops.
3Loss of information
If the system enhances the image zone masked by the scotoma by increasing size or shifting lines, then the information accessibility improves, but the interaction with text entities and gaze position is not optimized, leading to ineffective reading assistance
Solution Approach 1:
The patent implements dynamic image enhancement that adapts to the reader's gaze position and the specific characteristics of the scotoma. The enhancement parameters (size, position, orientation) are continuously adjusted based on real-time gaze data, ensuring that the most relevant information is always positioned optimally for the reader's current focal point, thereby improving both information accessibility and reading efficiency.
Solution Approach 2:
The system performs preliminary processing of the object zone before it needs to be read, pre-enhancing and repositioning the information so that when the reader's gaze moves to that area, the information is already optimized for immediate processing. This preliminary action reduces the cognitive load and improves reading efficiency.
Data Source
AI summary
A method of assisting visual exploration for individuals suffering from a retinal condition resulting in a scotoma and, more particularly, a method S for assisting visual exploration by such an individual of a digital image on a display device, the method including: —recognizing, using shape recognition software, at least one object contained in an object area of the digital image, —determining a blind area in the image, corresponding to a position of the scotoma in the field of vision of the user looking at the image, —if the blind area and the object area are detected to be in close proximity to each other, applying image processing to the image consisting in highlighting the object area for the user.