AI supported image processing – Impressive Results in Feature Extraction
Artificial intelligence excels in image processing by reliably recognizing defect features, even when the image data contains significant interference factors. For instance, shadows in X-ray images can complicate automatic inspection using traditional methods. However, AI effectively detects these issues. The processed results are then available for established inspection algorithms to further analyze in the inspection plan.
Practical Advantages
Depending on the application, AI can reduce inspection times and increase First Pass Yield (FPY). Artificial intelligence is also capable of replacing complex inspection strategies. Where manual or semi-automatic methods once led to optimal results, AI now offers an alternative approach, making it possible to quickly and successfully tackle previously unsolved tasks with AI supported image processing.
Get in touch with us!
Have we sparked your interest in Viscom’s AI supported image processing? We'd be happy to explain the next steps and guide you to the next level of efficiency.
DOWNLOADS
A Classic Example: Void Detection
High temperatures are expected in electronic components, especially in power electronics. Hidden gas inclusions in planar solder joints can lead to overheating and component failure. Nowadays, with 3D X-ray images and layer images, these voids can be inspected and measured without error. AI is paving the way for new possibilities to achieve equivalent results in 2D, saving cycle time.
The Popular BGA Component
The applications of AI supported image processing are diverse, and the added value is quickly evident when inspecting BGA components. Detecting BGA balls or voids in BGA balls can deliver more stable results with this advanced method. Approaches from pCT and artificial intelligence can be combined here.
Roadmap to the Optimal Application
In Viscom’s operating software, trained personnel incorporate the appropriate AI model as an intermediate step in the inspection plan. Consequently, the input images are then processed, defect features are extracted, and results are forwarded to subsequent process steps. AI technology deepens the scope of the inspection, forming a robust pillar alongside established 2D and 3D methods.
