Deep learning on Meyer sorters

deep learning

Deep learning on Meyer solutions: here's how it works.

Meyer never misses an opportunity to keep up with the times.

Over the past year, we have heard a lot about artificial intelligence, and we are happy to announce that Meyer solutions also have their own deep learning platform!

The deep learning technology used in our solutions attempts to mimic the functioning of the human brain. It assembles data, makes predictions, all with incredible precision.

What is deep learning in detail?

Meyer deep learning, for all intents and purposes, is a sub-category of machine learning. In summary, it is a neural network with three or more layers.
The different neural networks serve to mimic the behaviour of the human brain. It is thanks to this that the platform will be able to ‘learn’!
The more data it analyses, the more data it will learn.

deep learning

What is the difference with solutions that have a single neural network?

A single neural network will make approximate predictions.
A multi-level neural network, on the other hand, will make precise and increasingly optimised predictions.

Detection of impurities and defects with deep learning

Impurities or defects that can be identified by the human eye are difficult to define when using traditional algorithms.
Our sorters’ deep learning platform, on the other hand, solves this problem too!

Meyer’s deep learning is based on a vast database of samples.
It will immediately be possible to calibrate complex details such as:

  • colour;
  • shape;
  • texture;
  • etc.

This will only be the beginning!
Depending on the customer’s needs, our technology will be able to learn by building and iterating (repeating) increasingly precise selection patterns.

Independent learning will result in precise analysis, identification and selection of materials, meeting the customer’s selection needs.

The 7 steps to take to benefit from Meyer deep learning in implants:

  1. Equipment: upgrade and/or equip yourself with our deep learning technology;

  2. Data collection: after the first selections, a large number of high-definition photos of foreign bodies and defects will be collected, thus creating a database;

  3. Calibration: Meyer’s deep learning platform calibrates the machine automatically based on images of foreign objects and defects;

  4. Solution generation: the platform is now able to generate solutions by exploiting the technology embedded in the machines;

  5. Automatic machine learning: the sorter autonomously learns the characteristics of the data, identifying defects;

  6. Verification of results: the solutions developed by deep learning can be verified immediately to analyse the effects on selection;

  7. Optimisation: possibility to support further optimisation for any new requirements.

deep learning esempi

For any information, our team is at your disposal.

Those who look to the future choose Meyer!

Indice articolo