Security CT Image Analysis via 3D Volume Rendering and Slicing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing CT-based security inspection systems fail to provide fast and accurate information about objects, especially under time constraints, as screeners have limited time to analyze images and distinguishing hazardous objects is challenging due to overlapping and disordered luggage.

Innovation Solution

A method for operating a CT-based three-dimensional image that allows for single-click or frame selection of objects, providing information such as image contour, size, density, atomic number, CT number, and substance state, with alerts for hazardous objects, utilizing three-dimensional volume rendering and surface depth analysis to enhance image judgment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If CT-based three-dimensional imaging is used for security inspection, then the completeness of object information is improved, but the time required for analysis increases

Engineering Contradiction:
Improveobject information completenessVSAvoidimage analysis time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The patent divides the three-dimensional image into multiple two-dimensional sectional slices that can be independently selected and analyzed. Screeners can focus on specific sections containing objects of interest rather than analyzing the entire three-dimensional volume, thus reducing analysis time while maintaining information completeness for relevant areas.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent enables switching between three-dimensional volume rendering and two-dimensional sectional views. This dimensional transformation allows screeners to navigate through the image data in different ways, quickly locating objects in 3D space and then examining detailed 2D cross-sections, thereby balancing comprehensive information access with efficient analysis.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If detailed object information is provided for accurate identification, then the accuracy of hazard detection is improved, but the complexity of the system increases

Engineering Contradiction:
Improveobject identification accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent integrates multiple functions into a single system: three-dimensional volume rendering, two-dimensional sectional slicing, object selection, and information display. This multi-functional approach provides comprehensive object analysis capabilities without requiring separate dedicated systems, thus improving accuracy while controlling overall system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces an intermediary interface that mediates between the complex three-dimensional image data and the screener's needs. The interface provides tools for selecting objects and displaying relevant information in an organized manner, simplifying the interaction and making the complex system easier to use effectively.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Ease of operation

If objects are displayed with color coding for category identification, then the ease of object classification is improved, but the usefulness for comprehensive recognition is reduced

Engineering Contradiction:
Improveobject classification easeVSAvoidcomprehensive object recognition
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The patent applies color coding selectively to specific regions or objects within the three-dimensional image rather than uniformly to the entire image. This allows screeners to quickly identify categories of interest while maintaining the ability to examine detailed structural information of individual objects, thus balancing ease of classification with comprehensive recognition.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3130915B1Operation method and device relating to security inspection three-dimensional CT image
Publication Date: 2020.07.22 NUCTECH CO LTD
  • EP3130915B1 patent drawingFigure 1
  • EP3130915B1 patent drawingFigure 2
  • EP3130915B1 patent drawingFigure 3~4

AI summary

Provided are an operation method and device relating to a security inspection three-dimensional CT image, the method comprising: providing a security inspection three-dimensional CT image; selecting an object image from the three-dimensional image; and providing a response based on the selection. The present invention can acquire related information of the object image from the security inspection CT image by the selection and response performed with respect to the security inspection CT image, is practical, and provides effective reference information for image determination in the field of security inspection CT.