Drone Camera Object Alignment Using Circular Roof Markings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Obtaining accurate 'ground truth' data for training neural networks in autonomous vehicles is complex, particularly due to the challenge of determining the precise relative position and rotational shift between a drone and the vehicle, which affects the alignment of camera images and leads to inaccuracies in object recognition.
Innovation Solution
A method involving a regular pattern of circular markings on the vehicle's roof, recognized in drone camera data to determine the spatial alignment, using blob detection and Perspective-n-Point algorithms to convert object positions between drone and vehicle coordinate systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a drone is used to obtain ground truth data from a top-down perspective, then the accuracy of object position recognition is improved, but the complexity of determining relative position and rotational alignment between the drone and vehicle increases
Solution Approach 1:
A regular pattern (checkerboard or circular markings) is introduced as an intermediary object placed on the vehicle. This pattern serves as a mediator that can be easily recognized by the drone's camera, enabling the determination of the drone's position and orientation relative to the vehicle without complex direct measurements. The pattern acts as a reference frame that simplifies the alignment problem.
Solution Approach 2:
The regular pattern on the vehicle creates a known geometric reference that can be detected and used to infer the drone's spatial relationship. By capturing the known pattern from an unknown position and orientation, the system effectively creates a geometric copy problem that can be solved through pattern recognition algorithms, simplifying the overall measurement task.
2Ease of operation
If pixel grid alignment is used to determine spatial alignment, then the alignment process is simplified, but the accuracy decreases at low resolutions and with rotational shifts
Solution Approach 1:
The patent employs circular markings instead of rectangular pixel grid elements. Circles are rotationally symmetric and can be accurately detected regardless of the drone's rotational orientation relative to the vehicle. This curvature-based approach eliminates the directional bias inherent in pixel grids, allowing accurate alignment determination even when the drone is rotated by arbitrary angles.
Solution Approach 2:
The system changes from relying on pixel grid orientation parameters to using circular pattern parameters (center positions, radii). This parameter transformation makes the alignment process invariant to rotational shifts, as circles maintain their geometric properties under rotation. The detection algorithm works with radial distances and angular positions relative to circle centers rather than grid-aligned coordinates.
Data Source
AI summary
A method for recognizing a spatial alignment of an object in camera data. The method includes: a) providing a regular pattern including a plurality of rows of circular markings on the object; b) recognizing the circular markings in the camera data; c) ascertaining the center of at least two of the circular markings in the camera data; and d) ascertaining a spatial alignment and/or position of the object by finding at least one connecting line of at least two centers recognized in step c).