The basis of the solution is a camera with deep learning functionality. In the lumber industry, the solution can be used to determine where the age rings and the core are in order to ensure good processing of the lumber.
SICK Automation has developed a sensor solution that operates on the basis of deep learning algorithms. Following the SICK AppSpace eco-system that permits realisation of flexible solutions for automation applications, the deep learning sensor solution permits previously unimagined applications and contributes to driving Industry 4.0 forward.
Presented to delegates at the 2019 Hannover Fair, this deep learning technology is used by SICK in the industrial environment to specialise the functionality of its sensors. In this environment, the sensor learns to process information and therefore receives new functions. In addition, new processes are possible on the basis of adapted sensors. The sensor supplies, processes and analyses data using self-learning algorithms.
As an example, sensors are trained with a large number of images to give an answer to a specific question. On the basis of this training, the sensor can independently assign new unknown images to a result. “We are currently working with deep learning on a pilot project in the lumber industry. The basis of our solution is a camera with deep learning functionality,” says Grant Joyce, Sales and Marketing Manager, SICK Automation Southern Africa.
To ensure optimum use of the raw lumber material, sawmills must know about the conditions in the logs, such as knowing where the age rings and the core are in order to ensure good processing of the lumber. “To find out how the lumber can best be used, we taught the camera to identify these using deep learning, a task that could previously be performed only by humans,” says Joyce.
This technology makes it possible to realise new, previously inconceivable applications that make processes more efficient and more productive. In the pilot project, SICK was able to increase the material use, improve the quality of the products and avoid wasting resources. Additionally the sustainable use is possible not only with materials, but employees too. They no longer have to perform monotonous activities, freeing them up for more complex tasks.
“These developments in sensor technology provide the flexibility for production locations to be adapted for individual tasks,” Joyce concludes.