CT Explosive Detection Vessel Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computed tomography (CT) based explosive detection systems (EDS) face challenges in accurately identifying contraband, particularly low-density explosives and those concealed within dense containment vessels, leading to false alarms and misidentification due to inaccuracies in density measurements.
Innovation Solution
A method and system that differentiate between containment vessels and non-vessels by applying distinct sets of rules for classification, adjusting density thresholds based on vessel type, and compensating for inaccuracies caused by dense materials to reduce false alarms and improve detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If standard density thresholds are applied to all objects, then detection sensitivity for standard density explosives is maintained, but false alarm rate increases due to benign objects with similar density
Solution Approach 1:
The system applies different density thresholds and detection rules to different object types (containment vessels vs. non-vessels) based on local characteristics. Containment vessels use adjusted thresholds accounting for their dense material composition, while non-vessels use standard thresholds, thereby reducing false alarms without compromising detection accuracy.
Solution Approach 2:
The system dynamically changes detection parameters (density thresholds) based on the identified object type. When a containment vessel is detected, the system adjusts the CT number threshold and applies specialized rules to compensate for the dense material's effect on internal density measurements, optimizing detection performance for each object category.
2Measurement precision
If dense containment vessels are scanned with standard CT protocols, then vessel structure is visualized, but density measurements of contained items become inaccurate
Solution Approach 1:
The system introduces an intermediary correction model that accounts for the dense containment vessel's effect on CT number measurements. By detecting the vessel and applying compensatory rules, the system mediates between the distorted measurements and the true density values, recovering accurate density information despite the vessel's presence.
Solution Approach 2:
The system uses feedback from the detected containment vessel characteristics (material density, thickness) to adjust the CT number threshold and measurement interpretation. This feedback loop compensates for the vessel's distorting effect on internal item measurements, maintaining measurement precision despite the harmful interference.
3Reliability
If low-density explosives are detected without containment vessel consideration, then detection sensitivity is maintained, but false negatives increase due to shielding effects
Solution Approach 1:
The system applies specialized detection rules specifically for items within containment vessels, adjusting sensitivity thresholds and detection algorithms for these localized regions. This targeted approach ensures low-density explosives within vessels are detected with appropriate sensitivity while avoiding false negatives from shielding effects.
Solution Approach 2:
The system uses feedback from containment vessel detection to adjust the search criteria and detection sensitivity for internal items. By knowing a vessel is present, the system modifies its detection parameters to compensate for shielding effects, ensuring accurate detection of low-density contraband without increasing false positives.
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 effectively reduces false alarm rates and improves the detection of low-density contraband and dense vessel-contained items by applying specialized rules for containment vessels, enhancing the accuracy of contraband identification without increasing false positives or negatives.
Implementation Method 1
The numeric value of each pixel represents a CT number, which is an estimate of density. As used herein, a CT number is used as an estimate of density of a material, although the CT number is an indication of an attenuation coefficient of the material
Data Source
AI summary
A method and system for identifying an object in an acquired image. The method includes detecting an object within the acquired image, and determining whether the detected object is a containment vessel. If the object is not a containment vessel, the method includes applying a first set of rules for classifying the object. If the object is a containment vessel, the method includes applying a second set of rules for classifying the object.