Intelligent Warehouse Automation for reliable identification of 2,500 Tray Variants
Matrium GmbH is a full-range service provider in the fields of logistics, purchasing and customs. With the help of STEMMER IMAGING, the company has developed an automated logistics system to sort JEDEC trays for a customer in the electronics industry. Up to 2,500 different tray variants can be reliably distinguished using machine vision before being placed in their allotted compartments by a robot.
The task
Develop an inspection system to reliably distinguish between different variants of JEDEC trays serving as transport containers for electronic components.
The challenge
The transport containers for electronic components have up to 2,500 different geometries.
Furthermore, the trays have very small markings with various fonts and cause different reflections depending on the material composition.
The solution
The machine vision system developed by STEMMER IMAGING is equipped with three high-performance cameras to ensure reliable detection of the trays and uses the first images to verify that the brightness regulation for final inspection is optimally matched to the colour and material composition of the trays.
In addition, STEMMER IMAGING's CVB Polimago imaging tool enables the reliable training of patterns within a minimum of time.
Reliable distinction between over 2,500 tray variants of the same size
For some years now, Matrium has been doing the tray sorting as a service provider to one of the world's largest distributors of electronic components. After shipping the components, the trays have to be returned to the manufacturer sorted by type.
This sorting was previously done manually, which was extremely time-consuming considering the more than 2,500 tray variants. However, the requirements for automated machine vision were equally high as all trays have about the same footprint and differ only very slightly in terms of shape, material and thickness. In addition, the markings used to identify the trays are only slightly raised and up to 1.8 mm small. What makes it even more difficult is the non-uniform material composition, colour and mattness of the individual trays, which cause different reflections when the image is captured.
"The manufacturers wish to get these trays back sorted according to type – and this task is not easy to solve when one considers that there are about 2,500 existing variants, some of which differ only very slightly in terms of shape, material and thickness. The trays have a raised marking at the edge of the top side and optionally on one side on which the exact type is encoded. This is a bit like the markings on car tyres, but the markings on our trays can be as small as 1.8 mm."
Martin Eikel, IT Manager at Matrium
Customised solution: Automated intralogistics solution using state-of-the-art machine vision
Together with the experts from Matrium, STEMMER IMAGING developed a semi-automated sorting system based on a gantry robot that uses machine vision as a sensor system to run more economically and reliably - with a recognition rate above 95%.
The so-called Matrium Tray Sorting Machine (MTS-1) has five stack shafts that are manually filled with the different trays during operation. The MTS-1 sorts the trays before placing them in the corresponding storage compartments - fully automatically and for up to 3 hours without refilling.
Machine vision ensures stable and fully automated sorting
A specially designed gripper takes the top tray from a stack and positions it in front of the three cameras of the integrated machine vision system. The cameras capture images from above and from the side and send the image data to the system's industrial PC. Since the non-uniform material composition, colouring and mattness of the trays cause different reflections, the first images of each tray are evaluated as a histogram, whose results were then used to regulate the brightness for the following captures.
In this way, images with a medium grey-scale value are achieved for each object creating opimum conditions for the subsequent image evaluation. In total, up to six images are captured per camera - which is not critical in terms of time, as up to five electrical measurements of the tray surface resistance have to be carried out in parallel to the image capture and recognition.
Clever programming with CVB Polimago
On account of the large variety of trays, the programming of patterns for the Matrium project was just as important as the optimum image capture. The CVB Polimago imaging tool from the Common Vision Blox (CVB) software library from STEMMER IMAGING proved here to be the perfect tool for fast and reliable pattern recognition and for reading the markings. CVB Polimago has one decisive advantage: The automatic creation of training images greatly reduces the amount of programming required by the user.
The solution’s key success factors
Minimum programming work in the training phase
CVB Polimago’s algorithm generates artificial views of the model during programming in order to simulate various positions of a component in reality. In this way, stable results can be achieved with between 20 and 50 training images.
Seamless integration of software library
CVB Polimago’s well documented software library enables users to seamlessly integrate the image functions into both the Windows service for the machine and our own database-based management software.
Reliable pattern recognition
Geometrical transformations such as rotations, size changes, tilting, occlusions or changes in the lighting situation are already programmed during the training phase, which significantly increased the detection rate in the subsequent process.
„With around 2,500 tray variants it was important to us to keep the programming work to a minimum. CVB Polimago saves us a lot of time in the training phase. All in all, the support from STEMMER IMAGING was an enormous help to us in implementing the system quickly and achieving reliable sorting results."
Martin Eikel, IT Manager at Matrium
More successful projects in transport & logistics
Improve track monitoring efficiency for London Underground
STEMMER IMAGING has worked closely with London Underground to develop and deliver industrial camera assemblies that are mounted directly onto passenger trains to provide image data on the condition of the wheel/rail interface and the track itself. By acquiring this data during normal timetable periods, more time is available during the 4-hour night closure period to maintain the network’s 1,000 km of track, minimising any disruption to services and helping to make the new weekend extended operating times possible.
Benefit from our expertise
Turn our industry expertise and comprehensive services into your competitive advantage. Our experts will be happy to show you the potential for your company and work with you to realise tailor-made solutions to suit your requirements. We look forward to hearing from you and will get back to you promptly.
