Multi-Level Image Annotation System for Visually Impaired Users
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for presenting electronic images to the visually impaired are limited, as they provide static audio descriptions that do not allow users to access information at their desired level of detail or in their preferred manner, failing to offer a dynamic and customizable experience.
Innovation Solution
A system and method for annotating electronic images by storing them in a data repository with multiple levels of descriptive characteristics, allowing users to select and retrieve annotations through a user interface, and rendering these annotations in a dynamic and customizable way, enabling detailed and sequential descriptions based on user preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If static audio descriptions are used to describe electronic images, then the system is simple to implement, but the user cannot access information at their desired level of detail or in their preferred manner
Solution Approach 1:
The annotation system is segmented into multiple hierarchical levels (first level, second level, third level, etc.), where each level provides increasingly detailed descriptive characteristics of image pixels. This allows users to select their desired level of detail without requiring a completely different system structure.
Solution Approach 2:
The patent adds a new dimension to the annotation system by introducing hierarchical levels of description depth. Instead of providing a single flat description, the system organizes annotations in multiple layers (from basic pixel characteristics to complex object relationships), enabling users to navigate through different depths of information.
2Loss of information
If detailed multi-level annotations are provided for each pixel, then users can access information at their desired level of detail, but the data storage and processing complexity increases
Solution Approach 1:
The system performs preliminary organization of annotation data into hierarchical levels during the annotation creation phase. By pre-structuring the data repository with clearly defined levels and relationships, the system eliminates the need for complex real-time processing when users request annotations, as the data is already organized for efficient retrieval.
Solution Approach 2:
Different regions of the image can have annotations at different levels of detail based on their importance. The system allows selective annotation where critical areas receive more detailed multi-level descriptions while less important areas have simpler annotations, optimizing both information completeness and data management.
3Ease of operation
If annotations are provided in a fixed sequence, then the system is simple to implement, but users cannot receive information in the manner and sequence they desire
Solution Approach 1:
The annotation retrieval mechanism is made dynamic, allowing users to interactively select which pixels to annotate and at which hierarchical level. The system adapts to user preferences by enabling selection of specific annotation levels (first level for basic info, second level for detailed characteristics, third level for complex relationships) rather than following a fixed predetermined sequence.
Data Source
AI summary
A method and system for construction and rendering of annotations associated with an electronic image is disclosed. The system comprises a first data repository for storing the electronic image, which has a plurality of pixels, with one or more pixels annotated at a plurality of levels, which contain descriptive characteristics of the pixel, in ascending magnitude, such that the descriptive characteristics at a subsequent level are with reference to descriptive characteristics of one or more pixels surrounding the pixel. The system comprises a second data repository for storing the annotations. An image display module is configured to display the electronic image. A pixel and level identification module is configured to receive pixel and level selection details from a user-interface. An annotation retrieval module is configured to retrieve annotations corresponding to the pixel and level selection from the second repository and renders the retrieved annotations for the electronic image.