Tomographic Image Segmentation for Automated Anomaly Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current tomographic analysis methods for mechanical parts require significant operator intervention to detect anomalies, leading to delays and potential oversight due to the reliance on human assessment of three-dimensional images.

Innovation Solution

A method involving tomographic analysis that automatically subdivides the part into sub-parts, analyzes grey level distributions, compares parameters with standard ranges, and identifies risk regions to detect anomalies without operator intervention.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual assessment of three-dimensional images is used to detect anomalies, then detection reliability is maintained, but analysis time and operator workload increase significantly

Engineering Contradiction:
Improveanomaly detection reliabilityVSAvoidanalysis time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent introduces an automated analysis system that acts as an intermediary between the tomographic imaging device and the human operator. This system processes three-dimensional images automatically, extracting features and detecting anomalies without requiring manual assessment, thereby reducing analysis time while maintaining detection reliability through algorithmic precision

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the manual mechanical assessment process with an automated computational system. Instead of human operators manually examining three-dimensional images, the system uses automated image processing algorithms to detect anomalies, substituting human labor with machine-based analysis that is both faster and equally reliable

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

2Productivity

If automated analysis is implemented to reduce operator time, then productivity increases, but risk of missing subtle anomalies may increase

Engineering Contradiction:
Improveanalysis throughputVSAvoidanomaly detection accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements feedback mechanisms where the automated analysis system continuously refines its detection algorithms based on identified anomalies and comparison with known defect patterns. This feedback loop ensures that the system maintains high detection accuracy by learning from each analysis and adjusting its parameters to minimize false negatives

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent performs preliminary actions by pre-processing three-dimensional images and pre-identifying potential anomaly regions before final detection. This preliminary analysis prepares the data in advance, allowing the automated system to efficiently scan and detect anomalies with high accuracy without compromising thoroughness

Inventive Principle:
Principle #10Preliminary action

3Reliability

If comprehensive inspection of entire part volume is performed, then detection completeness is ensured, but analysis complexity and time increase

Engineering Contradiction:
Improvedetection completenessVSAvoidanalysis process complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the three-dimensional image into multiple sections or regions of interest, analyzing each segment separately rather than processing the entire volume as a single unit. This segmentation reduces analysis complexity by breaking down the large dataset into manageable portions while maintaining comprehensive coverage through systematic processing of all segments

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts and isolates specific features or regions that are most likely to contain anomalies, removing unnecessary data from the analysis process. By taking out only the relevant portions for detailed examination, the system reduces overall analysis complexity while ensuring that potential anomalies are not missed

Inventive Principle:
Principle #2Taking out (Extraction)

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 rapid and reliable detection of anomalies with reduced operator time and minimized risk of missing defects, optimizing the analysis process.

Implementation Method 1

at least one emitting device 3, configured to emit an incident beam 5 of wave pulses towards the part 20, for example radio frequency waves, X-rays or acoustic waves

Methodology Applied
Scientific EffectElectromagnetic radiation: Electromagnetic Induction

Data Source

PatentEP4441493B1Tomographic analysis method
Publication Date: 2026.02.04 SAFRAN AIRCRAFT ENGINES SAS
  • EP4441493B1 patent drawingFigure 1~2
  • EP4441493B1 patent drawingFigure 3~4
  • EP4441493B1 patent drawingFigure 5~6

AI summary

A method for the tomographic analysis a part in order to detect anomalies (30), the method comprising the following steps: - acquiring at least one three-dimensional image (25) of the part by means of a tomography device; - subdividing the image (25) into elementary subparts (27); - analyzing a grayscale distribution in each subpart (27) and obtaining at least one parameter representative of this grayscale distribution for each subpart (27); - comparing the one or more parameters obtained for each subpart (27) with standard values characteristic of a defect-free region and detecting abnormal subparts (31) for which the one or more parameters differ from the standard values; - determining risk regions, which comprise each abnormal subpart (31) and each subpart (33) adjacent to at least one abnormal subpart; and - analyzing the risk regions to detect the anomalies (30) in the part.