Higher accuracy
In the test setup, the metal hangers were detected 100% of the time by the deep learning system.
Playing it safe: We carried out a feasibility study for our customer to find metal hangers in wire mesh boxes using deep learning.
The metal hangers are placed in mesh boxes to be hardened in an oven. If hangers get caught in the mesh boxes during emptying, they go through the hardening process a second time. This leads to a loss of quality and to additional costs, the aim is to prevent this.
The aim of this feasibility study was to test whether the required inspection characteristics can be reliably recognized using industrial image processing. In the performed tests, the metal hangers were recognized with 100% accuracy, regardless of their position in the mesh box.
It is nearly impossible to recognize all metal hangers using conventional industrial image processing. This is because in certain positions, the hangers are very difficult to distinguish from the background due to the grid. Deep learning is used to solve this problem.
Deep learning is a method of machine learning and, put simply, artificial intelligence. The system is trained using provided images and is ideal for detecting anomalies on complex surfaces.
This illustration shows an empty pallet cage in which no metal hangers were recognized. This box is therefore classified as IO.
The system has recognized stuck metal brackets at the bottom of the mesh box.
The system has recognized stuck metal hangers in the corners of the mesh box.
The system has recognized metal hangers stuck to the side walls of the mesh box.
The deformation of the outer walls is generally not a reason for exclusion in this application, but it does lead to problems when detecting metal hangers that are located exactly under a bulge. Deep learning enables these deformations to be recognized automatically and the affected pallet cages to be excluded.
In the test setup, the metal hangers were detected 100% of the time by the deep learning system.
You can save time and money by automating your optical inspection. Check your wire mesh boxes for anomalies easily and automatically.
If the system requirements change, for example due to different components or boxes, the system can be retrained to meet the new requirements.
The processing time in this application is approx. 1 second.
At AUTFORCE, we specialize in testing systems & industrial software. specialized Get in touch with us. Together we will find the best solution for your project!
Christian Hanbauer
Test systems expert
+43 (664) 88 71 02 50
christian.hanbauer@autforce.com
Get a non-binding initial assessment now: