Maintain precise dimensions with 3D vision sensors from SICK
By Andreas Wiegelmesser, National Product Management Motion Control Sensors & Vision Solutions, SICK, Düsseldorf, Germany
Monday, 25 September, 2017
The 3D vision sensor designed for industrial use — the TriSpector1000 from SICK. This standalone solution, with its intelligent inspection tools, is suitable for a diverse array of applications relating to the quality control of consumer goods and packaging. The image-processing sensor is deigned to have an exceptionally high resolution for the detection of the most subtle features, easy configuration and operation and the ability to output directly usable measured values in millimetres for all inspection dimensions. Last but not least, thanks to its autonomous operating principle and its Gigabit Ethernet interface, it can be integrated into smart processes via Industry 4.0.
With the configurable TriSpector1000, integrators and end users can carry out a wide variety of inspection activities quickly and easily, including: height and completeness monitoring; volume, thickness and dimension measurement; integrity checks of containers; and counting and positioning objects. This is vital for fast processes and under adverse ambient conditions, as well as when there is a high variance of parts, demanding object properties or extremely high accuracy requirements. The image-processing capability is already integrated into the TriSpector1000, delivering up to 2000 3D profiles per second. These high-resolution measurement results are converted directly into millimetre values by the vision sensor without the need for another PC and output in real time over the Gigabit Ethernet interface. The TriSpector1000, therefore, offers all the capabilities for intelligent quality controls used to ensure, document and track flawless products and processes, for example, in the consumer goods and packaging industries.
3D inspection of consumer goods and packaging
Optical inspection and monitoring of goods and packaging serve to reliably detect faulty product features, incomplete packaging units or insufficient labelling. The key driver in this regard is in many cases the safety of the product, and consequently the safety of consumers. The TriSpector1000 makes it possible to reliably manage numerous 3D inspections of dimensions, quality and completeness in the food, pharmaceutical and packaging industries. The vision sensor features a rugged, anodised aluminium housing resistant to a variety of media, offers the choice between IP 65 or IP 67 enclosure rating, and is available with a front screen made of either glass or break-proof PMMA plastic. Thanks to laser triangulation, measurement is independent of product factors such as colour, shape, gloss, brightness, patterns, surface texture or any moisture on the product. Moreover, intensity data can be recorded by the vision sensor, allowing it to check for the presence of labels or imprinted patterns.
TriSpector1000: set up inspections easily and safely
To carry out various inspection tasks and meet a variety of integration requirements, the TriSpector1000 product family offers three different variants for different working ranges: 56–116 mm, 141–514 mm and 321–1121 mm. The single-housing concept ensures a geometrically stable and precise measuring scenario for any application. 3D image detection of moving objects using laser triangulation is not affected by the properties object mentioned above, background influences or even factors related to illumination and ambient light. As a result, both integrators and end users can set up virtually any inspection process step by step and with little effort. The correct version of the TriSpector1000 is selected and mounted according to the mounting situation and the working distance. The configuration interface visualises the settings and alignment of the measuring window and laser sensor. The movement speed of the conveyor is recorded by the encoder and automatically taken into account in the TriSpector1000 when calculating the 3D profiles.
Configurable camera sensor with an “integrated toolkit”
In the next stage, the four integrated image analysis tools of the TriSpector1000 — the Shape Locator, the Area tool, the Find Plane tool and the Blob Finder — are set up in a flexible and user-friendly manner by the SOPAS configuration software from SICK. These tools open up a variety of inspection possibilities. The Shape Locator, for example, uses a reference image to ascertain the 3D shape of an object (eg, a box of cookies) to be detected in the process by the TriSpector1000. The Area tool of the vision sensor ascertains whether there are parts present and performs measurements to determine whether they are in the correct quantity and height. This tool uses the position and alignment detected by the Shape Locator (eg, of a box) to position the measurement windows onto the correct area of the object. The Find Plane tool is suitable for inspection tasks in which the angle of surfaces (eg, a correctly applied screw top) is to be reliably measured and assessed in a matter of milliseconds. With the Blob Finder tool, the TriSpector1000 locates objects within a user-defined size range, regardless of their shape. For example, this allows the sensor to reliably check for the presence of various parts in the sorting tray of folding boxes while ensuring the integrity of the entire packaging unit. If parts of the specified contents are missing or if there are too many parts in the box, this will be registered as NOK. This will allow you to detect errors in upstream input processes and also prevent the causes of faults in downstream processes, eg, sealing the packaging.
Duplicating the configuration improves cost-effectiveness and availability
Integrated image processing, the factory calibration of the 3D data output and the uniform output of measured values in millimetres allow you to duplicate stored data and configurations according to a specific application. In this way, machine manufacturers and integrators can quickly and easily duplicate identical inspection tasks. In the event of a fault, end users can replace a device without losing time thanks to the use of stored data and configurations, thereby minimising downtime and optimising availability and performance in their plant.
TriSpector1000: more than a vision
Industrial image processing has been developed into what is one of the key automation technologies for industrial production and quality control. Image-based code readers, vision sensors and image-processing systems therefore occupy a corresponding amount of space in the SICK portfolio. With the data and measured values generated by vision sensors such as the TriSpector1000, as well as their integration into real-time Fieldbus environments, they provide valuable information which can be used for autonomous process control in the smart factory of today and tomorrow. So image processing within the context of Industry 4.0 is more than just ‘vision’ — it is an ‘enabling technology’ used for independently optimising the application of cyberphysical production systems (CPPS), eg, as intelligent equipment, thereby enabling responsive and process-oriented production control and quality assurance. The TriSpector1000 offers the capability required here to detect, measure, assess and communicate intelligently, thereby enabling futureproof process and quality control within the context of Industry 4.0.
Lobster protection packaging and easy-to-open milk were gold award winners at the 2020 PIDA...
A label with creases is mostly a 'no-no', but for one chocolate liqueur manufacturer this...
Nerida Kelton from the Australian Institute of Packaging takes us on journey from soft plastics...