マシンビジョンピープルの未来ブログ

Sharing

VITRONIC took a closer look at the topic of artificial intelligence (AI) in machine vision in the blog article "Machine vision gets intelligent." In this article, I discuss the integration of AI in machine vision systems and look at it using a concrete example from weld seam inspection — the detection and classification of weld spatter.

Weld spatters reduce the quality of visible surfaces or surfaces relevant for assembly. Therefore, it is essential to have an extremely reliable detection.

What are the benefits of integrating AI?

Classical machine vision (MV) has its limitations when dealing with very heterogeneous backgrounds and a high variation of a defect feature. Therefore, the result of classical edge detection is not always conclusive in the case of spatters. AI eliminates this disadvantage because AI autonomously detects patterns that humans themselves do not perceive or cannot quantify. In this way, AI makes MV more flexible, and the system learns from new conditions.

Spatter detection with neural network

But how can the assessment of "non-acceptable" weld spatter be expressed in terms of the tolerance of an inspection system? To do this, experts evaluate which irregularities the AI should classify as weld spatter. Convolutional neural networks (CNNs) are trained with this pre-assessed data. Such a trained network can classify other spatters better after only a very short training phase compared to the classical approach.

When the customer buys our VIRO WSI system, it already knows 50,000 variations of weld spatter. So, the customer doesn't have to train it first, VITRONIC has already done the work for him.
Oliver Hopp Key Account Manager Automotive

It is important to have a large number of different data sets for the defect type so that later a high bandwidth of different splashes can be recognized reliably. The more input data is available, the better the network can be trained.

The training data set, or more precisely, each abnormality is assigned a label. These labels form the target values for the training.

By matching the result with the target value, the CNN can gradually train its neurons to  give the expected response.
By matching the result with the target value, the CNN can gradually train its neurons to give the expected response.

It should be considered that neural networks are not deterministic. There is a high probability in error detection, but no uniqueness. For these probabilities, you can then set a threshold value at which classification as a spatter takes place. The lower the threshold value is set, the greater the uncertainty (100% minus threshold value). This in turn means that there can be pseudo errors in the classification even with neural networks.

Detected weld spatter with error probability
Detected weld spatter with error probability

Use of AI must be weighed up

Inline inspection requires compliance with the given cycle times of the production line. Therefore, in practice, the time for required for inspection is also essential.

For evaluation purposes, we have compared total inspection times of classical and AI-supported MV. Certain AI algorithms (such as SSD, Single Shot Detection) achieved almost the same speeds as classical machine vision.

Taking into account the higher accuracy of the SSD method, it is thus the better alternative for inspecting weld seams.

Comparison of evaluation times of conventional and AI-supported MV.
Comparison of evaluation times of conventional and AI-supported MV.
We are constantly working on optimizing our solutions, including the SSD for weld spatter. My focus is on further reducing pseudo defects while maintaining at least the same processing speed.
Daniel Heinze Developer VIRO WSI

Outlook

Of course, AI can be employed for more than just classifying weld spatter. In the future, VITRONIC will also use AI for detecting other defects such as pores and indentations. There is great potential to train AI on other criteria.

Conclusion:

In brief

  • A well-trained network delivers good results
  • High evaluation speed is suitable for inline testing
  • Higher customer satisfaction due to increasing inspection and product quality

Summary

The use of artificial intelligence offers possibilities where classical machine vision reaches its limits.

Oliver Hopp

Oliver Hopp

Key Account Manager Automotive
E-Mail
oliver.hopp@vitronic.com
Telefon
+49 151 689 622 73
Oliver Hopp has been supporting the sales team at VITRONIC as Key Account Manager Automotive since 2023. Thanks to his experience in the field of industrial machine vision, he is familiar with the requirements of various industries and a wide range of applications. From rule-based to AI-based algorithms, he advises and helps customers from industry and mechanical engineering to implement digital measures in the field of automated optical inspection. The quality requirements of the customers are always decisive.

フォローする

詳しく読む

アプリケーション

車軸部品の溶接ビード検査

溶接ビードの自動検査によるエラーのない車軸生産
溶接ビード検査による最高品質のリチウムイオンバッテリーの生産
レーザービード溶接による欠陥のない車体へ

関連記事

Contrary to conventional image processing, Artificial Intelligence (AI) does not need programmable rules. This often facilitates its use.
Are self-optimizing welding lines possible?
Subscribe to Newsletter
the machine vision people
Whether in automation or traffic engineering: With our systems our customers see more. The combination of hardware, data and intelligent evaluation makes applications and analyses possible that were unthinkable until recently. A previously hidden world becomes visible and usable.