High demands on your products, great time pressure and the competition never sleeps – whether in the food or heavy industry, quality, safety and speed are today more than ever factors that determine the success of your company.
To become not only competitive, but a competitive leader, you, your employees and your systems must do a flawless job, and in record time.
But errors are human and can happen – but they don’t have to. With a sentin VISION system you get the best out of your quality and inventory controls.
With our customers, the system learns to implement requirements in a similar way to a human being, thus supporting employees in doing their job even better and testing products more reliably. With our solutions, our partners can significantly reduce labor costs as well as complaints and pseudo rejects.
The sentin VISION system stands for robust and precise testing by our artificial intelligence (AI). Our recognition software is supplemented by a camera and an evaluation unit and can thus be perfectly embedded in your existing processes.
Unlike the human eye, fatigue and poor eyesight are not an issue with a camera. Likewise, our customers appreciate the reliability of our systems because they do not have to be constantly recalibrated. Our deep-learning recognition software also enables our partners to check much faster and more accurately. Depending on the application, an image acquisition and evaluation can take place in few milliseconds – around the clock, every day of the year.
Even though the human brain is capable of remarkable peak performance, it is beaten by an artificial intelligence, when it comes to durability: Because this is capable of recognizing more complex error patterns than the human brain is capable of. In one of our projects in the aerospace sector, a customer was able to rely on an accuracy rate of over 99.9%.
At the same time, an AI can be calibrated much faster than conventional machine vision systems. Often the system learns from less than 500 examples how something has to look like. For example, in an application with anomaly detection there were less than 30 images – which makes our solution not only more accurate, but also much more cost-effective.
You already have a machine vision system, but the accuracy is not sufficient for a clean inspection? Does the detection rate fluctuate noticeably or does the camera have to be constantly recalibrated? Then our solution offers you the possibility to retrofit your existing system and to supplement it.
By replacing the recognition software, our customers can upgrade their rule-based system and thus avoid slippage, pseudo scrap and recalibration as the system gets more performance through the deep learning approach of sentin VISION.
Image-based inspection is used in a wide range of industries and areas and is an indispensable process in quality assurance, for example in: