X-Ray Object Identification Using Multi-Angle Weight Parameters

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing baggage inspection systems require manual visual confirmation by trained inspectors to identify dangerous objects, which is inefficient and prone to errors, especially when dealing with unfamiliar shapes, and existing automated systems rely on template matching and database comparisons that are computationally intensive.

Innovation Solution

An object identification device that uses weight parameters stored in a memory unit to identify objects based on X-ray images from various angles, determining object presence without needing template comparisons, and provides visual and auditory warnings for dangerous objects.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If template matching is performed to identify objects in X-ray images, then object identification accuracy is improved, but image processing time and computational complexity increase significantly

Engineering Contradiction:
Improveobject identification accuracyVSAvoidimage processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent pre-calculates and stores weight parameters for multiple imaging directions during system initialization. These weight parameters are computed in advance and stored in memory, so that during actual inspection, the system only needs to retrieve and apply the pre-computed parameters rather than performing complex template matching operations in real-time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent transforms the object identification problem from template matching to a parameter-based evaluation approach. Instead of comparing images with templates, the system evaluates objects by computing a value based on pre-stored weight parameters corresponding to different imaging directions, fundamentally changing the identification methodology to reduce computational load.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If manual visual confirmation is performed by trained inspectors to identify dangerous objects, then identification accuracy is improved, but inspector workload and time consumption increase

Engineering Contradiction:
Improveidentification accuracyVSAvoidinspection throughput
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system enables automated object identification by having the determination unit automatically compare computed values against predetermined thresholds. The system serves itself by using pre-stored weight parameters and automated comparison logic, eliminating the need for manual inspector intervention while maintaining identification accuracy.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces the manual mechanical inspection process with an automated computational system. Instead of inspectors visually examining images, the system uses automated determination units that compute identification values based on weight parameters and compare them with thresholds, substituting human visual inspection with automated computational analysis.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If X-ray images are captured from multiple angles to improve object recognition, then identification reliability is improved, but imaging time and system complexity increase

Engineering Contradiction:
Improveobject recognition reliabilityVSAvoidimaging system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent pre-stores weight parameters for multiple imaging directions (first direction, second direction, third direction, etc.) in memory during system initialization. This preliminary preparation allows the system to handle multi-angle imaging data without requiring complex real-time processing, as the evaluation criteria are already prepared for each direction.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The determination unit is designed to universally handle X-ray images from multiple imaging directions using a unified identification approach. The same determination logic and threshold comparison mechanism work across all imaging directions, making the system multi-functional without requiring separate processing paths for each angle.

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

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

Enables efficient and accurate identification of objects irrespective of imaging direction, reducing inspector workload and minimizing the risk of overlooking dangerous items by providing clear visual and auditory alerts.

Implementation Method 1

an X-ray transferred image obtained by imaging the object from an angle of an arbitrary i direction (1 ≤ i ≤ n) by causing the imaging unit to irradiate X-rays to an item to be inspected and to image the item

Methodology Applied
Scientific EffectX-ray transmission: X-Ray

Data Source

PatentEP3748344B1Object identifying device and object identifying method
Publication Date: 2026.04.22 CYBERDYNE INC
  • EP3748344B1 patent drawingFigure 1~2
  • EP3748344B1 patent drawingFigure 3
  • EP3748344B1 patent drawingFigure 4

AI summary

Provided are an object identification device and an object identification method in which objects can be easily identified. The object identification device is provided with a pixel group extraction unit which scans, in units of the image area, an X-ray transferred image obtained from an imaging unit which performs X-ray imaging to an item to be inspected that is supplied, and extracts a plurality of pixel groups including characteristics of a shape of at least a part of the item to be inspected, and a determination unit which determines, with regard to the plurality of pixel groups extracted by the pixel group extraction unit, whether the item to be inspected corresponds to the object by executing all of the series of mappings related to an angle of an n direction by using each of the weight parameters based on the data group read from the memory unit.