Smarter than the average camera

Mar 5, 2012 | Machine Vision & Systems, News

By incorporating the latest technology from smartphones, it is now possible for smart cameras to offer image processing capabilities that were previously only available on PC-based systems. This new generation of smart cameras offers greater flexibility to the machine builder or systems integrator who want to use vision as an integral part of a process or machine.

Smart cameras have been important to the industrial vision industry for many years. All-in one units that combine image capture and processing in a single housing, often with an integrated illumination source, smart cameras process the image within the camera itself and output the results from the analysis over industry standard connections. Smart cameras can be used in all of the traditional industrial vision applications such as high volume component inspection, robot guidance, 1D and 2D (DataMatrix) code reading and verification, optical character recognition etc.

Smart camera performance has increased significantly in recent years to accommodate increased image resolution and more complex and faster processing and analysis requirements. Perhaps the most striking evidence of development in smart camera technology has been the emergence of 3D smart cameras with integrated laser source and optics for 3D triangulation.

Given that true 3D measurement methods have only fairly recently been used in industrial vision applications due to the computational loads involved and the difficulties associated with calibration and set-up, it is an excellent indication of the current state of smart camera technology that 3D capabilities are now available within a smart camera framework.

Advanced processing capabilities

The BOA Pro smart camera from Teledyne DALSA is the latest addition to the BOA range and features a digital signal processor as well as an ARM cpu, which allows it to run the full Sherlock inspection software suite, previously only available on PC-based systems.

Sherlock is an advanced machine vision software interface that can be applied to a wide variety of automated inspection applications for simple or very demanding tasks. It provides a rich suite of proven tools and capabilities that have been deployed in thousands of installations worldwide. This offers machine vision system professionals the comprehensive tools and capabilities needed for design, development, debugging, and user administration. Users can customise algorithms, construct scripts and develop operator interfaces that best suit their factory environment.

A comprehensive range of measurement and analysis tools is complemented by advanced pattern recognition tools for object alignment and robot guidance, together with calibration correction facilities to convert pixels into real world co-ordinates. Other facilities include 1D and 2D code reading and verification and optical character recognition.

The embedded software for the camera is setup via a local PC connection to the camera using Ethernet. Once configured for runtime, the Ethernet link can be used to communicate with other devices on the factory floor, such as PLCs, Robots and HMIs. With a choice of sensor resolutions from VGA to 1,600 x 1,200 pixels, a size of just 44 x 44 x 44mm and an industrial IP 67-rated housing, this is a compact and robust camera.

3D Smart Cameras

There are high computational demands placed on 3D smart cameras due to the need to process 3D cloud datasets. However, in the same way that the capabilities of 2D smart cameras continue to develop, regular enhancements to 3D cameras are being announced. The latest version of the Gocator series of 3D smart cameras from LMI Technologies now offers whole part measurement including volumetric measurements, 2D surface inspection and Modbus TCP support, all on cameras with megapixel resolution that can provide 3D profile data down to micron level detail. The ability to process whole parts makes factory automation easier and less expensive by eliminating the multiple components and software engineering required for automated part scanning and detection.

By automatically detecting the start and end point of a single or multiple components travelling along a conveyor, the whole part can be scanned and measured. With discrete parts segmented into 3D cloud datasets, it is

possible to perform volumetric measurements such as volume, centroid, orientation, etc to provide information on dimensions, location, and orientation. 3D smart cameras have applications in a wide range of industries including automotive, electronics, metals, transportation and timber. Not surprisingly, the 3D technique is popular in the automotive industry with applications such as: gap & flush; doors, hoods, decklids; door header correction; body in white; body sides; underbody; frames; motor compartment width variation; and rear axle inspection.

Support for record/playback is also provided and it is possible access whole part data using a software development kit. This new mode offers comprehensive 3D displays including pseudo colour visualisation of each part as well as data export for off-line analysis. With Ethernet communication over Modbus TCP protocol, PLC connected cameras integrate more easily and cost effectively into existing factory automation systems. Facilities are also provided for 2D imaging by extracting laser intensity information to build a 2D image for use with standard 2D processing tools to easily identify surface defects and patterns including barcodes, enabling the camera to inspect surface features in the same way as a traditional 2D camera system.

Stemmer Imaging

01252 780000