Automated Package Dimensioning via 3D Imaging and Shape Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining package dimensions in retail shipping are time-consuming, prone to human error, and expensive due to the need for manual measurements or complex multi-camera systems, lacking an efficient and accurate automated solution.
Innovation Solution
A package-dimensioning system utilizing a low-cost range camera to capture 3D images, analyzing surface features, and categorizing shapes to estimate dimensions, comprising an image-capturing subsystem, features-computation module, classification module, and shape-estimation module, which processes point clouds to provide accurate and efficient dimensioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual measurement with tape measure is used, then equipment cost is low, but productivity is low and measurement precision is poor
Solution Approach 1:
The patent replaces the mechanical tape measure system with an optical imaging system (camera) that captures images and uses computer vision algorithms to automatically calculate package dimensions. This substitution eliminates manual measurement operations, significantly improving both measurement precision through digital image analysis and productivity through automated processing of multiple packages per minute
Solution Approach 2:
The system enables packages to be measured automatically without human intervention. The camera captures images and the processing system automatically identifies package boundaries, calculates dimensions, and outputs results, allowing the measurement process to serve itself without requiring operator time or manual operations
2Productivity
If automated imaging system is used, then productivity is high, but device complexity increases
Solution Approach 1:
The patent extracts only the essential function needed for dimensioning - capturing a single two-dimensional image of the package - and processes it through algorithmic analysis to derive three-dimensional dimensions. This extraction approach avoids the complexity of multi-camera systems or 3D scanning hardware, achieving high productivity through simple imaging combined with sophisticated image processing
Solution Approach 2:
The system creates a two-dimensional digital copy (image) of the three-dimensional package and uses computational methods to extract dimensional information from this copy. This copying approach simplifies the physical system while maintaining measurement capability, as the image serves as a sufficient representation for dimension calculation
3Device complexity
If single two-dimensional image is used, then device complexity is low, but measurement precision deteriorates due to perspective distortion
Solution Approach 1:
The patent changes the parameter space by analyzing multiple parameters from the single image including pixel coordinates, known camera properties (focal length, sensor dimensions), and geometric relationships. By transforming these parameters through mathematical calculations, the system compensates for perspective distortion and recovers accurate three-dimensional dimensions from the two-dimensional projection
Solution Approach 2:
The system performs a dimensional transformation by converting two-dimensional image coordinates into three-dimensional package dimensions through mathematical modeling. By incorporating known camera parameters and geometric constraints, the system solves for depth and spatial relationships, effectively adding the third dimension back through computation rather than through additional physical imaging devices
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method for package dimensioning is provided. The package-dimensioning system includes an image capturing subsystem for acquiring information about an object within the image-capturing subsystem's field of view. A features-computation module analyzes object information and compiles a feature set describing the object's surface features. A classification module analyzes the feature set and categorizes the object's shape. A shape-estimation module estimates the dimensions of the object.