Handheld 3D-Vision Device for Flexible Object Dimension Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current measurement systems for object dimensions are limited to recognizing cuboidal objects and are not flexible due to their stationary position, making them unsuitable for measuring objects with different shapes.
Innovation Solution
A method and system using a handheld 3D-vision device to acquire depth data, convert it into a point cloud, identify and remove the ground plane, and extract point clusters to facilitate the measurement of objects with varying shapes, allowing for flexible and efficient dimension calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a stationary scanner is used for measurement, then the measurement position is fixed and stable, but the system lacks flexibility and cannot measure objects with different shapes
Solution Approach 1:
The patent transitions from a stationary scanner to a handheld mobile measurement device that can be dynamically positioned and oriented. The measurement device includes a controller that processes data from multiple positions and orientations to reconstruct three-dimensional object dimensions, enabling both flexibility in positioning and reliability in measurements.
2Device complexity
If a stationary scanner limited to cuboidal objects is used, then the system is simple and reliable, but it cannot measure objects with different shapes
Solution Approach 1:
The patent segments the measurement process into multiple discrete steps: acquiring two-dimensional image data from different positions and orientations, converting images to point clouds, stitching point clouds together, extracting planes, identifying the ground plane, removing the ground plane, extracting point clusters, and calculating three-dimensional dimensions. This segmented approach enables measurement of complex shapes while maintaining systematic processing.
Solution Approach 2:
The patent transitions from two-dimensional image capture to three-dimensional measurement by acquiring images from multiple positions and orientations, converting them to point clouds, and reconstructing three-dimensional object dimensions. This dimensional transformation enables measurement of objects with various shapes beyond cuboidal forms.
3Loss of information
If ground plane points are not removed from the point cloud, then all points are available for analysis, but computational effort and memory usage increase
Solution Approach 1:
The patent extracts and removes ground plane points from the point cloud after identifying the ground plane. This separation eliminates unnecessary points that would otherwise increase computational burden and memory usage, while preserving all relevant object points for accurate dimension measurement.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables the accurate and efficient measurement of objects with different shapes by simplifying the identification of target objects and reducing computational effort, allowing for the calculation of volume and weight, and supporting the measurement of non-cuboidal shapes.
Implementation Method 1
A 3D-camera system comprising a camera and a structured illumination device is used to acquire the depth data
Implementation Method 2
The structured illumination device can be arranged distantly to the camera of the 3D-camera system
Data Source
Figure 1
Figure 2~3A
Figure 3B~3C
AI summary
The invention relates to a method for measuring dimensions of a target object. The method comprises - acquiring depth data representative of the physical space, the depth data comprising data of the target object, - converting the depth data into a point cloud, - extracting at least one plane from the point cloud, - identifying a ground plane, - eliminating the ground plane from the point cloud, - extracting at least one point cluster from the remaining point cloud, - identifying a point cluster of the target object, - estimating dimensions of the target object based on the point cluster of the target object.