Image Detection of Concealed Prohibited Objects Behind Concealing Parts

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems face difficulties in accurately detecting prohibited objects concealed within items using concealing parts, such as metallic tubes or combination locks, which attenuate electromagnetic waves and obscure the objects in images.

Innovation Solution

A system utilizing a deep neural network (DNN) for segmentation and object detection algorithms to identify concealing elements and then employ multiple neural networks trained on different types of concealing elements to detect prohibited objects within these areas, enhancing detection accuracy and efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional detection methods are used, then the system is simple to operate, but the detection accuracy of concealed prohibited objects deteriorates due to obscuration by concealing parts

Engineering Contradiction:
Improvedetection accuracyVSAvoiddetection difficulty
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent applies segmentation by dividing the image processing task into distinct stages: first detecting concealing elements (metallic tubes, locks, reinforcement strips), then specifically searching for prohibited objects within and around these detected regions. This segmented approach allows the system to focus computational resources on high-risk areas, improving detection accuracy while managing complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary detection of concealing elements before searching for prohibited objects. By first identifying the location and boundaries of concealing parts, the system prepares targeted search regions in advance, which improves subsequent detection accuracy and reduces false negatives.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If multiple neural networks are used to detect concealed objects, then detection accuracy improves, but computational complexity increases

Engineering Contradiction:
Improvedetection accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the detection task into multiple specialized neural networks, each trained for specific functions: one network detects concealing elements, another detects prohibited objects, and a third verifies relationships between them. This functional segmentation improves accuracy by dedicating resources to specific detection challenges while keeping each individual network relatively simple.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary verification step where a third neural network checks the relationship between detected concealing elements and potential prohibited objects. This intermediary layer reconciles detections from multiple networks, reducing false positives and improving overall reliability without requiring a single overly complex network.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If comprehensive analysis of all image areas is performed, then detection thoroughness improves, but processing time increases

Engineering Contradiction:
Improvedetection thoroughnessVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent divides the image into relevant and irrelevant regions based on detected concealing elements. The system performs comprehensive analysis only within regions containing or adjacent to concealing elements, while skipping analysis of clear regions. This selective segmentation maintains detection thoroughness for high-risk areas while dramatically reducing overall processing time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system applies partial action by focusing detection resources on specific high-risk regions rather than uniformly analyzing the entire image. By concentrating computational effort on areas containing concealing elements and their immediate surroundings, the system achieves sufficient detection thoroughness for security purposes while minimizing processing time.

Inventive Principle:
Principle #16Partial or excessive action

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 system improves the detection of concealed prohibited objects by focusing on specific concealing elements, increasing accuracy and enabling real-time or quasi-real-time identification, thereby enhancing security and safety.

Implementation Method 1

an acquisition device operative to acquire an image of an item

Methodology Applied
Scientific EffectElectromagnetic radiation interaction: X-Ray

Implementation Method 2

concealing parts, such as metallic tubes or combination locks, which attenuate electromagnetic waves and obscure the objects in images

Methodology Applied
Scientific EffectElectromagnetic wave attenuation: Absorption (EM radiation)

Data Source

PatentEP4184449B1Detection of prohibited objects concealed in an item, using image processing
Publication Date: 2026.03.18 SEETRUE SCREENING LTD
  • EP4184449B1 patent drawingFigure 1A
  • EP4184449B1 patent drawingFigure 1B~1C
  • EP4184449B1 patent drawingFigure 1D~1E

AI summary

A system comprising a processor and memory circuitry (PMC) configured to: obtain an image of an item acquired by an acquisition device, perform a first detection using a first software module implementing at least one first deep neural network, to detect at least one given area of the image comprising at least part of a given element of the item enabling concealment of a prohibited object, perform a second detection comprising using a second software module implementing at least one second neural network to detect whether the given area comprises a prohibited object, and perform an action upon detection of a presence of a prohibited object in the image, wherein said detection is based at least on an output of said second detection.