Markerless Augmented Reality Overlay for Subsurface Infrastructure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality technologies require the placement of markers on surfaces to achieve precise overlaying of subsurface infrastructure, which is time-consuming and not feasible in many applications, especially when working with subsurface structures that are partially hidden or in environments with limited localization accuracy.
Innovation Solution
A method that uses camera images and spatial reference information to derive a two-dimensional projection of subsurface infrastructure, comparing anchor elements' positions between the image and projection to align and overlay the augmented view without the need for markers, utilizing motion tracking technologies and anchor elements visible in the real world scenery to compensate for inaccuracies in camera position and orientation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If markers are placed on surfaces to achieve precise overlaying of subsurface infrastructure, then overlaying precision is improved, but time consumption and operational complexity increase
Solution Approach 1:
The invention extracts and eliminates the marker placement requirement from the AR system. Instead of requiring external markers on surfaces, the system uses naturally visible anchor elements (manholes, utility covers, building features) that already exist in the environment, thereby achieving precise overlaying without the time-consuming marker placement process
Solution Approach 2:
The system enables self-service by using pre-existing environmental features as anchor points. The AR device automatically detects and utilizes visible anchor elements in the scene without requiring manual placement of markers, allowing the system to self-calibrate and provide accurate subsurface infrastructure visualization
2Measurement precision
If markers are placed on surfaces to achieve precise overlaying, then overlaying precision is improved, but ease of operation deteriorates
Solution Approach 1:
The invention removes the operational burden of marker placement from the user. The system automatically identifies and uses visible anchor elements in the environment, eliminating the need for users to manually place markers while maintaining high overlaying precision
Solution Approach 2:
The system achieves universality by working with any visible environmental feature as an anchor element - manholes, utility covers, building structures, or other recognizable features. This multi-functional approach allows the same system to operate across different locations and environments without requiring specialized marker placement procedures
3Speed
If camera position and orientation data is used directly for projection, then processing speed is improved, but manufacturing precision (alignment accuracy) deteriorates due to localization inaccuracies
Solution Approach 1:
The system implements feedback by comparing the projected position of anchor elements based on camera data with their actual detected positions in the image. The detected position offset is fed back to correct the camera position and orientation data, creating a closed-loop system that maintains high alignment accuracy while preserving real-time processing speed
Solution Approach 2:
The invention replaces the direct mechanical reliance on camera localization data with an optical-computational correction mechanism. Instead of trusting raw camera position data, the system uses image-based anchor element detection and projection geometry to computationally correct alignment, substituting mechanical precision requirements with computational accuracy
Data Source
Figure 1~2
Figure 3~4
Figure 5a~5b
AI summary
The invention relates to providing an augmented view (11) of a real world scenery and of an occluded subsurface infrastructure (6). An image (8) is taken by a camera (2) with an image sensor and image reference information comprising a camera position and a camera orientation is provided. From three dimensional information (9) of a subsurface infrastructure (6) a two dimensional projection on the image sensor is made by using the reference information. A projection position (7b') of an anchor element (7) of the subsurface infrastructure (6) being visible on the at least one image (8) is compared with an image position (7a') of the anchor element (7). A difference (10) between the image position (7a') and the projection position (7b') is compensated for matching and overlaying the two dimensional projection derived from the three dimensional information (9) of the subsurface infrastructure (6) with the at least one image (8) and thereby providing an improved augmented view (11) .