X-ray Diffraction Imaging Voxel Aggregation for Signal Enhancement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing X-ray diffraction imaging (XDI) systems face challenges in accurately identifying materials due to low signal levels from voxels, leading to unreliable spectra and difficulty in distinguishing objects of interest, especially when they are proximate or adjacent to each other, resulting in high false alarm rates and reduced detection efficiency.
Innovation Solution
A computer-implemented method and system that reconstructs a four-dimensional voxelized representation of scatter cross-sections within a container, segments voxels into contiguous groups, computes an aggregated momentum transfer spectrum, and classifies objects as threat or non-threat based on the spectrum, enhancing material identification and reducing false alarms by improving signal processing and segmentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If small voxels are used to improve visual resolution of smaller object features, then visual resolution is improved, but signal level per voxel becomes very low (0-25 photons per voxel)
Solution Approach 1:
The patent combines signals from multiple contiguous voxels to form an aggregated signal. By merging the photon counts from several adjacent voxels that likely contain the same object or material, the system achieves sufficient signal levels (100+ photons) for reliable material characterization while maintaining the ability to resolve smaller features through the segmentation process.
2Measurement precision
If signals from adjacent voxels are aggregated to provide sufficient material spectra, then material identification accuracy is improved, but visual discrimination of proximate objects becomes difficult
Solution Approach 1:
The patent divides the scanned volume into multiple small, contiguous voxels before aggregation. This segmentation allows the system to maintain spatial resolution by keeping voxels small enough to represent distinct objects, while the subsequent aggregation process combines signals only from voxels that likely contain the same material, achieving both visual discrimination and material identification accuracy.
3Measurement precision
If polychromatic x-ray beams are used to irradiate each object voxel from several different directions, then material discrimination is improved, but signal level per voxel remains very low
Solution Approach 1:
The patent combines multiple approaches: using polychromatic x-ray beams from multiple directions for improved material discrimination, and simultaneously aggregating signals from multiple contiguous voxels to achieve sufficient signal levels. This dual strategy resolves the contradiction by maintaining the benefits of multi-directional polychromatic irradiation while overcoming the low signal level problem through signal aggregation.
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
The approach enables cost-effective, high-probability detection of materials of interest with low false alarms, improving the resolution of objects and substances, particularly in cabin and hold baggage screening, by aggregating voxel spectra and effectively distinguishing between threat and non-threat substances.
Implementation Method 1
an x-ray source transmits x-rays through an object or a container, towards a detector
Implementation Method 2
x-ray diffraction imaging (XDI) systems feature an x-ray multisource emitting a multiplicity of polychromatic x-ray beams
Implementation Method 3
each object voxel is irradiated from several different directions. These systems measure spatially-resolved x-ray diffraction profiles of the constituent voxels
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method of performing a security inspection of a container including a plurality of objects includes irradiating the container with polychromatic x-rays. The method also includes reconstructing a 4-D voxelized representation defining a momentum transfer spectrum and generating a 3-D image by determining a single value at each voxel as a function of the momentum transfer spectrum. The method further includes segmenting the voxels into segments of contiguous voxels. Each segment includes a plurality of contiguous voxels that at least partially map onto at least one object. The method also includes computing an aggregated momentum transfer spectrum over at least a portion of the plurality of contiguous voxels. The method further includes classifying the aggregated momentum transfer spectrum as one of threat and non-threat and distinguishing the at least one object as one of a threat segment and a non-threat segment based on the aggregated momentum transfer spectrum.