AR City Light Recognition via Flexible Image Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current augmented reality technologies face limitations in recognizing and interacting with real-world objects, particularly 2D and 3D printed items, as they often require specific markers or restricted areas of recognition, which can restrict flexibility and adaptability in capturing and providing digital overlays.
Innovation Solution
The system employs image recognition techniques that allow for flexible definition of recognition areas, excluding security features if necessary, and uses cloud-based databases to provide dynamic digital overlays, enabling recognition and interaction with a wide range of printed objects without the need for QR codes or hidden signatures, allowing for real-time updates and integration with various devices like smartphones and AR glasses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If specific markers or restricted areas of recognition are required for augmented reality recognition, then recognition accuracy is improved, but flexibility and adaptability in capturing printed objects deteriorate
Solution Approach 1:
The patent extracts and removes the requirement for specific markers or restricted areas of recognition from the augmented reality system. By using general image recognition techniques that can identify any printed object without requiring predefined markers, the system eliminates the constraint that limited flexibility while maintaining recognition accuracy through advanced image processing algorithms.
Solution Approach 2:
The patent creates a universal recognition system that can handle multiple types of printed objects (advertisements, product packaging, printed materials) without requiring different marker types or restricted recognition areas. The system uses cloud-based databases and machine learning to universally recognize various printed formats, enabling the same device to interact with diverse printed content across different contexts.
2Adaptability or versatility
If cloud-based databases are used to provide dynamic digital overlays, then adaptability and real-time updates are improved, but device complexity and data processing requirements worsen
Solution Approach 1:
The patent introduces a cloud-based database as an intermediary between the printed object recognition and the digital overlay delivery. The cloud database stores extensive information about various printed objects and serves as a mediator that processes recognition results, retrieves appropriate digital overlays, and delivers them to user devices. This intermediary approach enables real-time updates and adaptability while distributing computational complexity to the cloud infrastructure rather than requiring complex processing in every user device.
3Reliability
If image recognition techniques exclude security features from recognition areas, then recognition reliability is improved, but information loss from excluded areas worsens
Solution Approach 1:
The patent segments the printed object image into different functional areas: recognition areas (excluding security features for reliable identification) and information areas (including excluded regions for data extraction). The system performs segmented processing where image recognition focuses on reliable identification zones while separate information extraction processes capture data from previously excluded areas such as security features, watermarks, or restricted zones, ensuring both reliability and comprehensive information retrieval.
Data Source
AI summary
A city light information system recognizes a real world object relating to a city light and uses augmented, mixed and/or virtual reality to display information relating to a property and/or the transaction on a smart device such as a smartphone or AR glasses.