CT Threat Detection via 3D Shape Histogram Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional CT baggage scanners are inefficient in detecting threat objects due to slow scanning speeds and high false alarm rates, failing to provide adequate resolution and thoroughness in identifying potential threats like plastic explosives or sheet items, which limits their effectiveness in high-throughput airport environments.
Innovation Solution
A method and system that processes CT data by identifying voxels with density values, computing eigen-values and eigen-vectors, generating a 2D eigen-projection, extracting the object contour, and calculating a shape histogram to classify objects as threats or non-threats, using pre-computed threat shape histograms and a threshold to determine the presence of threat objects like guns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional CT scanning methods are used to detect threat objects, then the scanning process is simple and fast, but the detection accuracy and resolution are insufficient leading to high false alarm rates
Solution Approach 1:
The patent segments the detection process into multiple stages: initial CT scanning to identify candidate objects, extraction of shape features from candidate objects, comparison with stored threat profiles, and final classification. This multi-stage segmentation allows thorough analysis of potential threats while maintaining operational efficiency by only applying complex processing to candidates of interest.
Solution Approach 2:
The patent transitions from conventional 2D CT scan images to 3D object modeling and shape analysis. By reconstructing three-dimensional representations of detected objects and analyzing their geometric properties in multiple dimensions, the system achieves superior detection accuracy and resolution compared to traditional two-dimensional imaging methods.
2Measurement precision
If conventional CT scanning is used, then the scanning speed is fast, but the resolution and thoroughness in identifying threat objects like plastic explosives or sheet items are inadequate
Solution Approach 1:
The system performs preliminary CT scanning at standard speed to quickly identify candidate threat objects, then applies enhanced 3D shape analysis only to these candidates. This preliminary action approach maintains fast scanning speeds for the overall process while achieving high resolution and thoroughness for potential threats through targeted detailed analysis.
Solution Approach 2:
Rather than applying full high-resolution 3D reconstruction to all scanned objects (which would be time-consuming), the system applies partial action by performing detailed shape analysis only on objects that meet certain criteria or show potential threat characteristics, thus balancing resolution requirements with scanning time constraints.
3Reliability
If conventional object classification methods are used, then the processing is simple and fast, but the false alarm rate is high
Solution Approach 1:
The patent employs multi-parameter characterization of objects including density, shape descriptors, and geometric features - analogous to using multiple spectral channels or 'colors' of information. By analyzing objects through multiple descriptive dimensions and comparing against comprehensive threat profiles, the system achieves high reliability in distinguishing threats from non-threats while managing classification complexity through systematic methodology.
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 significantly improves the detection rate and reduces false alarms by incorporating shape information into the classification process, enabling more accurate identification and discrimination of threat objects within CT scanning systems.
Implementation Method 1
expose the material to X-rays and to measure the amount of radiation absorbed by the material, the absorption being indicative of the density
Data Source
AI summary
A method of and system for detecting threat objects represented in 3D CT data uses knowledge of one or more predefined shapes of the threat objects. An object represented by CT data for a region is identified. A two-dimensional projection of the object along a principal axis of the object is generated. A contour of the object boundary in the projection image is computed. A shape histogram is computed from the extracted contour. A difference measure between the extracted shape histogram and a set of pre-computed threat shape histograms is computed. A declaration of a threat object is made if the difference measure is less than a pre-defined threshold.