3D Object Image Generation for Luggage Bin Scanning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Radiation systems face performance degradation when examining objects within larger secondary objects, such as luggage bins, due to increased 3D volume reconstruction and processing times, requiring more computational resources and longer analysis times.
Innovation Solution
A method and system for generating a three-dimensional object image by projecting the image along a first axis to define a two-dimensional boundary, reprojecting it to extract voxels within the boundary, and identifying perpendicular boundaries to separate the object from the secondary object, thereby reducing computational costs and processing time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the 3D image includes both the object and the container (secondary object), then the examination coverage is improved, but the processing time and computational resources increase
Solution Approach 1:
The patent segments the 3D image data by separating voxels belonging to the container from those belonging to the object. The processing system divides the volumetric data into distinct regions, allowing independent processing of the object of interest while excluding the container, thereby reducing computational load while maintaining examination coverage.
Solution Approach 2:
The patent extracts and removes voxels representing the container (secondary object) from the 3D image data before processing. By taking out the unnecessary container data, the system reduces the volume of data requiring reconstruction and analysis, directly decreasing processing time and computational resource requirements while preserving complete object examination capability.
2Reliability
If the 3D image includes both the object and the container (secondary object), then the examination coverage is improved, but the computational resources required increase
Solution Approach 1:
The processing system segments the 3D volumetric data by identifying and separating voxels belonging to the container from those belonging to the object. This segmentation allows the system to process only the necessary object data, reducing computational resource requirements for reconstruction, storage, and analysis while maintaining complete examination coverage.
Solution Approach 2:
The patent extracts and removes container voxels from the 3D image data before processing. By eliminating unnecessary container data, the system reduces the computational burden on reconstruction algorithms, decreases storage requirements, and lowers processing power demands, directly addressing the computational resource complexity issue.
3Strength
If rigid containers are used to ensure carriage of largest bags, then the container strength is improved, but the 3D volume presented for processing increases
Solution Approach 1:
The patent extracts and removes voxels representing the rigid container from the 3D image data. By taking out the container volume after the fact, the system maintains the structural integrity and strength benefits of rigid containers during transport, while eliminating their volumetric impact on processing, reconstruction, and storage operations.
Solution Approach 2:
The patent transforms the problem from a 3D processing challenge to a 2D projection analysis approach. By projecting the 3D volumetric data onto 2D planes and analyzing cross-sections, the system can identify and separate container boundaries without being constrained by the full 3D volume, effectively reducing the computational dimensionality while maintaining accurate separation.
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
This approach effectively separates the object image from the secondary object image, reducing computational resources needed and shortening processing times, allowing for more efficient examination and analysis.
Implementation Method 1
A degree to which the radiation photons are attenuated by the object (e.g., absorbed, scattered, etc.) is measured to determine one or more properties (e.g., density, z-effective, shape, etc.) of the object
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A volume of an object is extracted from a three-dimensional image to generate a three-dimensional object image, where the three-dimensional object image represents the object but little to no other aspects of the three-dimensional image. The three-dimensional image is yielded from an examination in which the object, such as a suitcase, is situated within a volume, such as a luggage bin, that may contain other aspects or objects that are not of interest, such as sidewalls of the luggage bin. The three-dimensional image is projected to generate a two-dimensional image, and a two-dimensional boundary of the object is defined, where the two-dimensional boundary excludes or cuts off at least some of the uninteresting aspects. In some embodiments, the two-dimensional boundary is reprojected over the three-dimensional image to generate a three-dimensional boundary, and voxels comprised within the three-dimensional boundary are extracted to generate the three-dimensional object image.