Initial Situation

In the semiconductor industry, reliable weld seams on pipes are extremely important. If a point is leaky, the factory must be stopped, leading to very significant financial consequences. Georg Fischer offers products for welding pipes. During welding by heating and pressing together, molten material emerges at the pressing point. An inspection method was sought that objectively records geometric quality features of the weld seam, such as thickness and pipe offset, and from this assesses the quality of the connection.

Bild2 e1576161713166

Project Overview

Customer
Georg Fischer

Area
Industrial & Finance

Solution
Image-based inspection of weld seams.

System Environment
Backlight setup, inexpensive camera modules and lighting, classic vision algorithm

Approach
Innovation workshop, feasibility study in the SCS Vision Lab with rapid iteration cycles

Award
1st Place IVS Innovation Award 2019, Product Launch February 2020

SCS’s Contribution

In an innovation workshop, SCS and Georg Fischer developed solution ideas, which SCS subsequently developed and tested in a feasibility study in its own Vision Lab.

Rapid Iteration and Discarding

Experimental setups enabled a quick evaluation of various approaches. An initially promising laser setup was discarded after initial tests because the different optical properties of the pipe materials did not allow for sufficiently consistent evaluation.

Solved with Transmitted Light

The chosen solution works with transmitted light, thereby creating comparable image conditions even with different pipe materials. This allowed for the use of simple, inexpensive camera modules and lighting.

Classic Image Processing

A classic vision algorithm analyzes the captured images, measures relevant geometries, and from this determines the weld seam quality, without the need for Machine Learning.

Objectively Assessed Instead of Visually

The method reproducibly captures geometric features such as weld seam thickness and pipe offset from the image. This creates an objective basis for quality assessment.

Laser with results 1536x506 1

Result

The feasibility study showed a simple and cost-effective image-based method with very good inspection results. Based on this, Georg Fischer further developed the approach into the Weld Bead Inspection Tool. The product achieved 2nd place in the IVS Innovation Award in 2019 and was launched in February 2020.

Transmitted light instead of laser

simpler setup for different pipe materials.

2nd Place

in 2019 at the IVS Innovation Award.

February 2020

Launch of the resulting tool.

Services Used

Computer Vision

Algorithms, ML & AI

Industrial & Finance

The bald man in the white shirt, smiling at the camera, represents the friendly SCS team.

Your Contact Person

Pascal Iselin

Algorithms, Analytics & AI
+41 43 456 16 00

For questions about computer vision, feasibility studies, and image-based quality inspection.

More Case Studies

  • Cable drawing

    Banedanmark (Danish State Railways)

    Digitization of Cable Diagrams

    Automated digitization of thousands of cable diagrams.

  • White lab device labeled "zenCELL owl" utilizing edge computing; blurred lab utensils in the background.

    Manufacturer of zenCELL owl

    AI at the Edge – Counting Cells

    Deep-learning cell counting directly on the lab PC.

  • A person in a lab coat monitoring schematics of industrial systems via a touchscreen.

    Schubert

    Line Management System

    Central control of an entire packaging line.