Web Inspection System for Real-Time Full Web Dirt Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current web inspection systems are inadequate for real-time detection and analysis of defects and formation irregularities in web manufacturing, particularly failing to provide full web coverage and high-density dirt analysis, and are not capable of simultaneous detection of strong and weak defects and formation irregularities.

Innovation Solution

A method utilizing a line scan camera to acquire and process digital images of the web in real-time, identifying regions of interest through decision rules based on thresholding, spatial, feedback, or statistical analysis, and employing morphometric methods for classification and reporting, enabling 100% full web coverage and consistent detection results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If snapshot images or limited cross direction band or scanning imaging methods are used, then device complexity is reduced, but measurement precision and productivity are insufficient for real-time full web coverage and high-density dirt analysis

Engineering Contradiction:
Improvedirt analysis precisionVSAvoiddirt counting speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The web inspection system divides the web into multiple cross-directional bands or zones that are processed independently and in parallel. Each band can be analyzed separately for dirt particles, allowing the system to handle high-density dirt analysis across the full web width while maintaining real-time processing capability. This segmentation enables the system to process over 1000 dirt particles per second by distributing the computational load across multiple processing channels.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If full web coverage and high-density dirt analysis are implemented, then measurement precision is improved, but processing speed and real-time capability deteriorate

Engineering Contradiction:
Improvefull web measurement accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSSpeed

Solution Approach 1:

The system performs preliminary processing of the captured web image by pre-segmenting it into manageable cross-directional bands before detailed dirt particle analysis. Pre-processing steps such as noise filtering, contrast enhancement, and initial defect detection are applied to the entire web image first, preparing the data for faster subsequent analysis. This preliminary action enables real-time processing of full web coverage by reducing the computational complexity of the main analysis stage.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If simultaneous detection of strong and weak defects and formation irregularities is performed, then adaptability is improved, but device complexity increases

Engineering Contradiction:
Improvedefect detection versatilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The web inspection system employs a universal image processing platform that can detect and classify multiple types of defects including strong defects, weak defects, and formation irregularities using the same hardware and software infrastructure. The system uses multi-threshold analysis and adaptive processing algorithms that automatically adjust detection parameters based on the type of defect being searched for, enabling one system to perform multiple inspection functions without requiring separate dedicated systems for each defect type.

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

4Productivity

If online real-time classification is performed, then productivity is improved, but measurement precision and reliability worsen due to high dirt densities over 1000 particles per second

Engineering Contradiction:
Improveonline classification capabilityVSAvoidclassification accuracy at high density
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

At very high dirt densities exceeding 1000 particles per second, the system applies partial classification where not all dirt particles are individually classified in real-time. Instead, the system uses statistical sampling and density-based estimation methods that provide reliable overall contamination level assessment without requiring complete individual particle classification. This approach maintains online real-time capability while achieving sufficient measurement precision for quality control purposes at high production speeds.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3459045B1Real-time, full web image processing method and system for web manufacturing supervision
Publication Date: 2020.03.11 ABB (SCHWEIZ) AG
  • EP3459045B1 patent drawingFigure 1
  • EP3459045B1 patent drawingFigure 1a
  • EP3459045B1 patent drawingFigure 2

AI summary

A method for detection of distinctive features in a web being transported in a moving direction during a web manufacturing process is presented, the method comprising the steps of: a) acquiring an image of the web, said image being representable as a digital image comprising a plurality of pixels P i with i ∈ {1;...; p}, b) identifying a plurality of regions of interest each corresponding to a distinctive feature by processing the plurality of pixels P i by: c) selecting a local pixel unit comprising a subset P j with j ∈ S ⊂ {1;...; p} of the plurality of pixels, said subset i) being representative of a subregion of the digital image, and ii) different from previously selected local pixel units, d) deciding whether the local pixel unit is of interest or not, i) if the local pixel unit is of interest, 1) identifying whether the local pixel unit is located within an impact area A k of a previously identified region of interest R k with k ∈ A ⊆ {1;...; n}, 2) if the local pixel unit is not located within any impact area A k of any previously identified region of interest R k with k ∈ A ⊆ {1;...; n), or no regions of interest have previously been identified, (a) identifying the local pixel unit as a new region R n +1 of interest; (b) initializing an impact area A n +1 for said new region R n +1 of interest; (c) incrementing a counter n representative of the number of previously identified regions of interest; (3) if the local pixel unit is located within an impact area A k 0 of a previously identified region of interest R k 0, (a) merging, depending on a merging condition, the local pixel unit with said previously identified region of interest R k 0, (b) if the merging condition is fulfilled, updating the impact area A k 0 of said region of interest R k 0; ii) preferably, if the local pixel unit is not of interest, (1) identifying whether the local pixel unit is located within an impact area A k of a previously identified region of interest R k with k ∈ {1;...; n}, (2) if the local pixel unit is located within an impact area A k 0 of a previously identified region of interest R k 0, updating said impact area A k 0; e) repeating steps b) through d) until at least essentially all pixels of the image have been processed.