Sharp Object Detection in CT Scanners via Eigen-Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current detection algorithms for computed tomography (CT) scanners have low detection rates and high false alarm rates when identifying sharp objects, such as knives, due to their distinct characteristics, which are not adequately addressed by existing mass and density-based detection methods.
Innovation Solution
A method and system that generate a 3D CT image, perform eigen-analysis to extract features like axial concavity ratio, pointness measurement, and flat area, and calculate a sharpness score using quadratic penalty functions to specifically detect sharp objects, improving discrimination and classification capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If mass and density-based detection methods are used, then the detection system is simple, but the detection rate for sharp objects is low and false alarm rate is high
Solution Approach 1:
The detection algorithm segments the object detection process into multiple stages: first identifying objects based on mass and density, then performing eigen-analysis to extract shape features, and finally classifying objects based on computed features like axial concavity ratio, pointness measurement, and flat area. This segmentation allows the system to handle complex sharp objects through systematic feature extraction while maintaining overall system manageability.
Solution Approach 2:
The patent transitions from two-dimensional mass-density detection to three-dimensional shape feature analysis by performing eigen-analysis on the CT data. This dimensional expansion enables the system to capture geometric characteristics (concavity, pointness, flatness) that are invisible to traditional mass-based methods, thereby improving detection reliability for sharp objects.
2Reliability
If traditional detection algorithms are used, then the system is easier to operate, but the false alarm rate is high
Solution Approach 1:
The patent extracts and isolates specific geometric features (axial concavity ratio, pointness measurement, flat area) from the CT data through eigen-analysis. By taking out these critical shape characteristics and analyzing them separately, the system can accurately distinguish sharp objects from non-sharp objects, reducing false alarms while maintaining operational simplicity.
Solution Approach 2:
The system changes the detection parameters from mass and density to geometric parameters derived from eigen-analysis. This parameter transformation enables more accurate classification by capturing the essential shape characteristics of sharp objects, thereby reducing false alarms without significantly increasing operational complexity.
3Measurement precision
If mass and density-based methods are used, then the processing time is short, but the detection precision for sharp objects is insufficient
Solution Approach 1:
The patent performs preliminary eigen-analysis and feature extraction on the CT data to pre-compute geometric characteristics before final classification. This preliminary action prepares the data in advance, allowing the system to quickly identify sharp objects based on pre-extracted features like axial concavity ratio and pointness measurement, thereby improving precision without excessive processing time.
Solution Approach 2:
The system applies partial eigen-analysis focused on extracting only the most critical shape features (axial concavity, pointness, flat area) rather than performing complete geometric analysis. This partial action approach maintains high detection precision for sharp objects while minimizing processing time by avoiding unnecessary computational steps.
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 proposed method significantly enhances the detection rate and reduces false alarms by utilizing features specific to sharp objects, improving the overall performance of automatic object identification and classification in CT scanning systems.
Implementation Method 1
a three-dimensional (3D) CT image is generated by scanning an object
Implementation Method 2
Method of and system for sharp object detection using computed tomography images
Data Source
AI summary
A method of and a system for sharp object detection using computed tomography images are provided. The method comprises identifying voxels corresponding to individual objects; performing eigen-analysis and generating eigen-projection of an identified object; computing an axial concavity ratio of the identified object; computing a pointness measurement of the identified object; computing a flat area of the identified object; calculating a sharpness score of the identified object; and declaring the identified object as a threat if the sharpness score is greater than a pre-defined threshold.