Optical Marker Tracking With Camera Pose for Real-Time Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current object tracking systems using optical markers face challenges in accurately detecting and classifying optical markers in real-time, especially in dynamic environments, leading to issues with false positives and negatives.
Innovation Solution
The system employs an imager and computer vision to detect and classify optical markers affixed to objects, using a processor to determine camera position and relative position information, and projects this information to detected markers, utilizing optical markers with unique characteristics and machine-readable formats like bar codes or QR codes, and integrating with cameras that can detect various spectra.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If computer vision is applied to detect optical markers in real-time, then tracking accuracy is improved, but processing time increases
Solution Approach 1:
The system pre-generates an optical marker dictionary containing multiple distinct optical markers before tracking begins. This preliminary preparation allows the processor to efficiently match detected markers against known patterns during real-time operation, reducing processing time while maintaining high detection accuracy through pre-computed reference data
Solution Approach 2:
The system varies multiple parameters of optical markers including shape, size, color, and machine-readable code patterns to create a diverse dictionary. This parameter variation enables the system to maintain high detection accuracy across different viewing conditions and distances while the processor can quickly eliminate non-matching markers through efficient parameter comparison
2Reliability
If optical markers with unique characteristics are used, then detection reliability is improved, but false positives increase
Solution Approach 1:
The optical marker dictionary is segmented into multiple distinct marker types with unique characteristics. Each marker in the dictionary has distinct visual features that can be independently identified. This segmentation allows the system to reliably distinguish between different markers and reduce false positives by matching detected patterns against specific known marker designs rather than generic patterns
Solution Approach 2:
The system uses feedback from the detection process to verify marker identification. The processor compares detected optical markers against the pre-generated dictionary, and the system can adjust detection parameters based on verification results. This feedback mechanism reduces false positives by confirming detections against known marker patterns and allowing for correction of misidentifications
3Measurement precision
If camera position information is determined and projected to markers, then tracking precision is improved, but system complexity increases
Solution Approach 1:
The system uses an optical marker dictionary as an intermediary between the camera system and the tracking objectives. The pre-generated dictionary serves as a reference medium that simplifies the projection calculations by providing known marker positions and characteristics. This intermediary structure reduces system complexity by decoupling the complex projection mathematics from real-time processing, allowing tracking precision to be maintained through reference matching rather than complex real-time calculations
Data Source
AI summary
Embodiments include system and processes for tracking objects using a camera. An optical marker dictionary including one or more optical markers is generated, the optical markers being optically distinct indicators. An optical marker within the optical marker dictionary is associated with and affixed to an object. A processor is in communication with the camera, receiving image data from the camera and applying computer vision to the image data in order to detect the presence of one or more optical markers within the optical marker dictionary within the image data. The processor determines camera position information and applies computer vision to the image data in order to determine relative position information for the detected optical markers and projects a position from the camera to a detected optical marker.