Dynamic Lighting Surface Inspection for Defect Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing optical inspection methods for object surfaces are limited in flexibility, allowing only predefined image recording and evaluation options, which restricts the ability to capture high-quality images and detect surface defects effectively.

Innovation Solution

A method involving the relative movement of an object and an area camera, with illumination by a lighting device that can produce incident, dark field, or transmitted light, allowing for a sequence of high-quality images to be recorded, which are then processed to generate surface images with specific lighting characteristics, enabling the extraction and combination of image areas to form continuous surface images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a single defined lighting (dark field, grazing light, or incident light) is used for surface scanning, then the inspection method is simple to implement, but the flexibility and ability to detect different surface defects are limited

Engineering Contradiction:
Improveflexibility in image recording and evaluation optionsVSAvoidcomplexity of lighting and image processing system
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent applies dynamics by making the lighting conditions adjustable and changeable during the inspection process. The lighting device can switch between different lighting modes (incident light, dark field lighting, transmitted light) dynamically, allowing the system to adapt to different inspection requirements without changing the physical hardware configuration. This resolves the contradiction by providing versatility through dynamic adjustment rather than through complex multi-device configurations.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements universality by designing a single lighting device that can produce multiple types of lighting conditions (incident light, dark field lighting, transmitted light) and a single image processing system that can generate different surface images with various lighting characteristics. This multi-functional approach allows one device to perform the work of multiple specialized devices, enhancing flexibility without proportionally increasing system complexity.

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

2Measurement precision

If a large number of images are recorded in sequence during relative movement, then high-quality surface inspection is achieved, but the data processing and storage requirements increase

Engineering Contradiction:
Improvequality of surface defect detectionVSAvoidvolume of image data to be processed
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent applies extraction by selectively extracting specific areas from the recorded image sequence based on lighting characteristics. Instead of processing all recorded images equally, the system extracts relevant surface areas that contain defect information under specific lighting conditions (incident light characteristics, dark field characteristics, transmitted light characteristics). This reduces the effective data volume that needs detailed processing while maintaining high detection quality.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent implements preliminary action by pre-recording and storing a large number of images during the relative movement phase before any analysis is performed. This preliminary data acquisition ensures that sufficient high-quality images are available for subsequent defect detection, and the actual processing can then focus on analyzing this pre-prepared dataset rather than acquiring and processing data simultaneously, improving both quality and efficiency.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If multiple lighting characteristics are used to generate different surface images, then comprehensive defect detection is achieved, but the time required for image processing increases

Engineering Contradiction:
Improvecompleteness of defect detectionVSAvoidtime for image processing and analysis
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies continuity by performing image processing operations continuously during and after the image acquisition phase. The system generates surface images with different lighting characteristics (incident light, dark field, transmitted light) in a continuous workflow rather than as separate discrete steps. This continuous processing approach maintains reliability through comprehensive defect detection while minimizing total processing time by eliminating idle transitions between processing stages.

Inventive Principle:
Principle #20Continuity of useful 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

This approach enables the capture of numerous high-quality images in a short time, allowing for the detection of various surface defects, such as scratches and protrusions, and the generation of 3D depth images, providing enhanced flexibility and accuracy in surface inspection.

Implementation Method 1

The object is illuminated with an illumination device (30, 32). By means of incident light and dark field illumination, different surface defects can be shown particularly meaningfully in the surface image

Methodology Applied
Scientific EffectLight reflection and transmission: Reflection

Data Source

PatentEP2799847B1Method for optically testing the surfaces of objects
Publication Date: 2018.10.24 COMPAR
  • EP2799847B1 patent drawingFigure 1~2
  • EP2799847B1 patent drawingFigure 3
  • EP2799847B1 patent drawingFigure 4~5

AI summary

To optically inspect the surface (10) of an object (12), the object (12) is rotated about its longitudinal axis (14). During the rotation, a sequence of a large number of images (52) is recorded using an area scan camera. From the stored images (52), a specific area (56; 58; 60) is extracted, and the extracted areas (56; 58; 60) are combined to form a synthesized surface image.