Underwater Spatial Property Determination via Interactive 3D Bounding Volume
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for determining spatial properties of submerged offshore objects require prior knowledge of the object's markings or CAD models, making them unsuitable for ad-hoc measurements, especially when limited or no information is available.
Innovation Solution
A method using an underwater movable platform with a camera and position/orientation measurement system, where image data and camera positions are captured, and a 3D bounding volume is projected onto the images for user-adjusted fit, allowing determination of spatial properties like heading, position, verticality, and inclination without prior object information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sonar is used to determine spatial properties of submerged objects, then measurement can be performed from a distance without physical contact, but spatial resolution is low making accurate determination difficult
Solution Approach 1:
The patent replaces the mechanical/acoustic sonar system with an optical imaging system (camera). Instead of using sound waves to detect spatial properties, the system uses light to capture images of the submerged object, which are then processed to determine spatial properties such as position, orientation, and dimensions with higher precision.
Solution Approach 2:
The patent creates a visual copy of the submerged object through camera imaging. By capturing optical images of the object and processing them to generate a digital representation, the system obtains spatial property information without physical contact, achieving both distance measurement capability and high spatial resolution.
2Measurement precision
If markings are attached to the object or detailed CAD models are used, then spatial properties can be determined accurately, but the system becomes unsuitable for ad-hoc measurements where no prior information is available
Solution Approach 1:
The patent enables the system to determine spatial properties using only the captured images themselves, without requiring external markings or pre-existing CAD models. The image processing algorithms extract spatial information directly from the visual data, allowing the system to serve itself and adapt to any submerged object regardless of whether prior information is available.
Solution Approach 2:
The patent performs preliminary image capture and processing to generate a digital representation of the object before any measurement is made. By pre-processing the images to extract spatial features and create a measurable model, the system prepares the data in advance, enabling accurate measurement without requiring pre-attached markings or prior knowledge of the object.
3Measurement precision
If complex processing algorithms are used to determine spatial properties from images, then measurement precision improves, but processing requirements and computational complexity increase
Solution Approach 1:
The patent divides the complex task of spatial property determination into separate processing stages: image capture, feature detection, model generation, and measurement extraction. By segmenting the processing pipeline, the system reduces the computational complexity of each individual stage while maintaining overall measurement precision through the coordinated execution of multiple simpler tasks.
Data Source
AI summary
Method for determining a spatial property of a submerged object in a 3D-space, using an underwater moveable platform, that is provided with a camera and with a position and orientation measurement system configured to output a position and orientation of the camera. The method includes receiving user input for visually fitting a bounding volume to the object, and determining the spatial property of the object based on a corresponding spatial property of a bounding volume once the user is satisfied that the bounding volume sufficiently fits the object. The method provides an interactive manner of visually fitting the bounding volume to the object, based on images of the object taken at different positions and/or orientations of the camera relative to the object, as well as on a representation of a position and orientation of the camera in the 3D space at the time each image was captured.