3D Data Processing for Security CT Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current security inspection CT machines face challenges in efficiently recognizing prohibited articles due to high passenger flow volumes and low accuracy in three-dimensional data processing and image segmentation, leading to increased working intensity and reduced throughput.
Innovation Solution
A new three-dimensional data processing and recognizing method that involves scanning and reconstructing objects, extracting feature data, merging adjacent points, recognizing cross sections, and cutting objects with perpendicular planes to identify shapes, thereby improving the recognition of prohibited articles such as cones, cylinders, and other dangerous items.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional image segmentation methods are used for three-dimensional data processing, then the system can process prohibited articles, but the accuracy and generality of recognition are insufficient
Solution Approach 1:
The patent applies segmentation by dividing the three-dimensional data processing into distinct stages: initial image segmentation to identify potential targets, extraction of interested targets based on specific features, and subsequent shape recognition. This multi-stage segmentation approach improves both accuracy and generality by handling different aspects of recognition separately with specialized algorithms for each stage.
Solution Approach 2:
The patent transitions from two-dimensional image segmentation to three-dimensional data processing by introducing depth information and volumetric analysis. The system processes three-dimensional CT data, extracts three-dimensional features, and performs shape recognition in 3D space, which significantly improves recognition accuracy and generality compared to traditional 2D methods.
2Productivity
If manual inspection methods are used to recognize prohibited articles, then detailed examination can be performed, but working intensity is high and throughput is reduced
Solution Approach 1:
The system implements self-service by automatically performing image segmentation, target extraction, and shape recognition without requiring manual intervention. The intelligent algorithm autonomously processes three-dimensional data, identifies prohibited articles, and classifies them by shape, thereby reducing working intensity and increasing throughput while maintaining high accuracy.
Solution Approach 2:
The patent replaces manual mechanical inspection with an automated computational system that uses three-dimensional data processing, feature extraction, and shape recognition algorithms. This substitution of mechanical human labor with an automated system significantly increases throughput and reduces working intensity while improving consistency and accuracy.
3Measurement precision
If comprehensive three-dimensional data processing is performed, then recognition accuracy improves, but processing time increases
Solution Approach 1:
The patent applies preliminary action by performing initial image segmentation and extracting key three-dimensional features before conducting comprehensive shape recognition. By pre-processing the data to identify and isolate interested targets with relevant features, the system reduces the complexity of subsequent analysis, thereby maintaining high recognition accuracy while reducing overall processing time.
Solution Approach 2:
The system extracts only the relevant three-dimensional features and interested targets from the complete three-dimensional data, rather than processing all data uniformly. This selective extraction of critical information maintains recognition accuracy by focusing on key characteristics while reducing processing time by eliminating unnecessary data analysis.
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 method enhances the accuracy and efficiency of recognizing prohibited articles, reducing working intensity and increasing throughput by effectively searching, extracting, and recognizing shapes of interested targets in security inspection CT systems.
Implementation Method 1
Known explosive detection systems obtain images of the contents in packages by X-ray (radioscopy) or CT techniques
Data Source
AI summary
A three-dimensional data processing and recognizing method including scanning and re-constructing objects to be detected so as to obtain three-dimensional data for recognition of the objects to be detected; extracting data matching to features from the three-dimensional data, so that the extracted data constitutes an interested target; with respect to the data matching to features, merging and classifying adjacent data points as one group, to form an image of the merged interested target; recognizing a cross section of the interested target; cutting the interested targets by a perpendicular plane which passes through a central point of the cross section and is perpendicular to it, in order to obtain a graph; and recognizing shape of the interested targets based on a property of the graph.