Computer Vision Void Filling for Automated Shipping Packaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional packaging processes for online purchases are labor-intensive and require significant human intervention, leading to inefficiencies and increased costs due to incorrect or excessive use of protective packaging.
Innovation Solution
A system and method utilizing computer vision and machine learning to detect shipping containers, generate 3D point clouds, identify void areas, and autonomously dispense protective packaging to fill these areas, reducing human involvement and optimizing packaging materials.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If human assistance is used throughout the box filling/sealing process, then the process can be completed with simple equipment, but labor intensity and time consumption increase significantly
Solution Approach 1:
The patent replaces manual mechanical operations with an automated system comprising computer vision technology, 3D point cloud generation, and machine learning algorithms. The system automatically detects void areas, calculates required protective packaging quantities, and controls dispensing mechanisms, eliminating the need for human intervention while maintaining packaging quality.
Solution Approach 2:
The system enables the packaging process to serve itself by using computer vision to automatically identify void spaces, calculate appropriate packaging material quantities based on void volume characteristics, and dispense materials without human assistance. The machine learning model autonomously makes decisions about packaging requirements.
2Reliability
If excessive protective packaging is used to ensure product safety, then product protection reliability improves, but material waste and packaging costs increase
Solution Approach 1:
The system performs preliminary detection and calculation of void areas before dispensing protective packaging. By using computer vision to map the 3D space and identify exact void locations and volumes in advance, the system calculates the precise quantity of packaging material needed, preventing both over-packing and under-packing.
Solution Approach 2:
The system dynamically adjusts the quantity of protective packaging based on measured void volume parameters. The machine learning model analyzes void characteristics (volume, shape, location) and determines optimal packaging material quantities, transforming the packaging process from fixed-quantity to variable-quantity based on actual needs.
3Measurement precision
If manual assessment of void areas is used, then the system remains simple, but measurement precision and packaging accuracy deteriorate
Solution Approach 1:
The patent transitions from 2D image analysis to 3D point cloud processing for void area detection. By generating and analyzing three-dimensional spatial representations, the system achieves precise measurement of void volumes, shapes, and locations, enabling accurate calculation of required packaging material quantities.
Solution Approach 2:
The system introduces computer vision technology and 3D point cloud generation as intermediary steps between detecting the package contents and determining packaging requirements. This intermediary processing layer enables precise automated measurement of void spaces without requiring complex manual assessment procedures.
4Productivity
If traditional packaging processes are used, then equipment requirements remain simple, but productivity and packaging speed decrease
Solution Approach 1:
The automated system enables continuous operation of the packaging process without interruptions for manual assessment or material calculation. The computer vision system continuously monitors and identifies void areas, and the dispensing mechanism continuously supplies protective packaging material, maintaining steady workflow and increasing throughput.
Data Source
AI summary
Systems and methods for packing a shipping container. The methods comprise: detecting, by a computing device, a presence of the shipping container; identifying, by the computing device, a void area inside the shipping container; dividing, by the computing device, the void area into a plurality of void area sections; obtaining, by the computing device, a first amount for a first void area section of the plurality of void area sections and a second amount for a second void area of the plurality of void area sections; and causing, by the computing device, the protective packaging to be dispensed into the void area such that the first amount of the first void area section is filled with the protective packaging and the second amount of the second void area section is filled with the protective packaging.