Automated Surface Inspection

How AI Image Recognition is growing

Wednesday 15th July, 2020

Image recognition with AI is sweeping across the automotive industry, and for good reason. After discussing how BMW is innovating with AI previously, in this blog we’re looking into their image recognition technology. It’s helping their teams become more efficient and ensuring higher quality.

AI (artificial intelligence) image recognition is used primarily in assembly lines to evaluate images of a component/part and compare with hundreds of other images from the same sequence. If the part is different from the standard image, then it is flagged immediately. It’s useful in the automotive industry because faults can be detrimental, and can go easily un-noticed by the human eye.

So, how are these systems helping BMW? We discovered from Metrology that these systems are allowing the production team to identify ‘whether the hazard warning triangle, wider caps and door sills have all been correctly fitted to each car’. Associates have trained the AI systems by taking photos of the component from various perspectives and marking potential deviations on the images. Then, the AI creates an image database for real time evaluation. It all happens automatically in less than seconds!

By having image processing technology to essentially play ‘spot the difference’ between a perfect product and a faulty alternative, you’re saving so much time and money in the long-term. BMW continues to use AI extremely well for quality assurance.

Follow Shelton Vision on Twitter