Vision sensors take image processing onboard

Industrial vision sensors are suited for complex applications where standard photoelectric sensors are not sufficient.

Pepperl+Fuchs has extended its range of applications in image processing with two new configurable vision sensors.

The VOS300 series of vision sensors is designed to give more flexibility in setup and operation without requiring any programming knowledge.

Industrial vision sensors such as the VOS310 and VOS301 are suited for complex applications where standard photoelectric sensors are not sufficient.

These complex applications occur when the position of the object being detected is not defined or the shape and size of the object being detected change frequently.

Applications suitable for Pepperl+Fuchs vision sensors include: checking for presence of objects or patterns, for example label detection and orientation, or date stamps; position recognition of objects (position or angle of rotation), for example pick and place machines and assembly systems; and completeness checks, for example a missing screw in a terminal block.

The VOS310 vision sensor offers up to five different feature tests, covering grey scale, contrast, gradient, pattern and contour.

By adjusting the threshold value the grey scale image is converted in to a binary black and white image ready for processing.

Here the processing software counts the number of black pixels, this is a simple check used for detecting presence on plain backgrounds.

For contrast tests the bright pixels are compared with the dark pixels in terms of quantity and brightness.

The resultant value can be used for example to check for the presence of a date stamp on a plain background.

When testing gradient, the software counts the number of black to white transitions, this is useful when the orientation of the object can vary or when the area to be checked has a more complicated background.

In pattern testing the sensor can be set to recognise a specific pattern, which is ideal for label detection.

The software is tolerant to position changes of the pattern but not to rotation of the pattern.

And in contour tests the contour of a shape is recorded and can be recognised by the software in any orientation.

This method is ideal for detecting the completeness of parts on a conveyor.

For less complex image processing tasks, the VOS301 is the optimum solution as it tests for only one feature.

The parameters that are required can be set directly or using the Teach-In function.

Setup and configuration are easy using the VOS3 configuration software (included).

Once configured, the sensor will operate autonomously without the need for connection to a PC or PLC; alternatively the sensor can be integrated into existing process networks via its Ethernet interface.

Back to top