All aboard

Share this on social media:

Topic tags: 

Smart cameras are now well established, but continue to find new markets, as Warren Clark discovers

The smart camera continues to divide opinion: some see it as ideal for just about every application, while others consider it expensive and too complex to program for many smallscale applications.

‘Smart cameras are well established and have been around for some time,’ says Fabio Perelli of Matrox Imaging. For some, this means they are not as exciting as newer technologies, such as GigE Vision, but with more and more companies adding smart cameras to their portfolio, it is a market that still has growth ahead of it.

A smart camera differs from any other in that it a certain amount of image processing takes place on board the camera. Once the image has been processed, data can be sent to a thirdparty device (such as a PC) or a decision can be sent elsewhere (e.g. a ‘shut-down’ message to machinery in the event that a faulty product has been found.

‘The smart camera is well suited to many applications,’ says Perelli. ‘It is a small, selfcontained device that can generate “pass/fail” results, for example, without the need for a permanently connected host PC.’

‘The biggest advantage of smart cameras is  obviously that it is a true standalone product,’ says Basler’s Henning Tiarks. ‘There is no need to have a PC nearby, or even to be connected to a PC at all. For example, a customer could install a camera within some machinery on a production line, having programmed it to check for defects on products that pass its field of view. If it spots a defect, the camera can tell the machine immediately and shut down the machine automatically, without any need for a PC or further user intervention. This makes smart cameras particularly useful for large production plants that would otherwise require huge lengths of cable to link units back to a central PC, or for locations where using a PC is unsuitable.

 

A smart camera from Matrix Vision is used in the Ardeche region of France to help with traffic management on the Ardeche and Sorgue rivers. The camera supports the boat rental companies in managing the departure times of boat groups, as well as offering general monitoring of the river.

Programming

The most significant drawback of smart cameras is the level of programming knowledge that is required in order to get the most out of a device. ‘Depending on the camera you choose,’ says Matrox’s Perelli, ‘this can mean anything from using complex C++ programming to detail the functionality required, through to a simplified software interface where a user without any programming knowledge can enter parameters for the camera to check. Such parameters might include measuring tolerances to pattern matching or data matrix code identification. ‘

‘The main drawbacks with smart cameras are complexity and price,’ says Basler’s Tiarks. ‘A customer often needs to have a system with at least 15 to 20 cameras or so, in order to justify the effort that has to go into programming the systems – and of course, individual unit price is much higher than a standard camera.’

Marcus Bleise of Matrix Vision disagrees that the programming of a smart camera is particularly complex. ‘In reality, users of just about any camera type require some level of programming knowledge, so I am not sure this is as significant a downside as some believe it to be, especially because we provide embedded Linux as a standard operating system in the camera.’

Reducing data

With processing taking place on board the camera, one of the biggest advantages of a smart device is that it reduces the amount of data that needs to be transmitted.

‘A smart camera can filter out unnecessary detail onboard before sending the images off elsewhere,’ says Basler’s Tiarks. ‘An example of this is being used in the film-making industry for motion capture purposes, where the camera can be programmed only to follow particular markers [on the actors] and capture that particular data.’

‘Decentralised intelligence has major benefits,’ says Matrix Vision’s Bleise. ‘The camera can be programmed to only send a live stream when an event occurs. In large-scale surveillance applications, where an operator might otherwise have to monitor several feeds at once, this event-triggered activity will catch the operator’s attention. An example of this might be in a hospital or nursing home, where movement detection built into a smart camera could alert the nurses’ station to a sudden fall by a patient, for example. Event-triggered data transmission solves the major bottleneck in vision applications – the interface between the camera and the PC.’

‘The limits of what a smart camera can do are governed by the power of the CPU on board and also by how much data needs to collected,’ says Matrox’s Perelli. ‘Applications such as web inspection, where huge amounts of data need to be processed, are not suited to smart cameras. Some smart camera models might have issues with heat dissipation around the processor, but Intel is working on methods of reducing power consumption and, therefore, the associated heat problem.

There are many advantages to using a smart camera – ‘not least of which is the fact that a smart camera is a self-contained unit,’ says Perelli. ‘This often means there is less chance of a breakdown and, if there is, it will be much easier to pinpoint the location of the fault. Also, if the function of the camera ever needs to change, it can be brought back to the PC and reprogrammed with the new information.’

Firstsight Vision’s Mark Williamson sounds a warning against smart cameras being the answer to everything though. ‘There is a perception that smart cameras are simple units that reduce cost,’ he says. ‘While that can be the case, it is not universally true. The reality is that many so-called smart cameras do not have an I/O output capable of coping with a reject mechanism. In such cases, a separate breakout box is required – and this drives up the cost and wipes out any savings that have been made by having the processing on board.

Smart cameras are also not cost effective for anything that requires more than one view of the product.’

 

The SmartSurv surveillance system, featuring a camera supplied by Matrix Vision, is shown here in a care home. The smart camera recognises and classifies activity – for example, if a person falls in a corridor, it will only alert nursing staff is the person does not then stand up.

Application possibilities

‘An upcoming application for smart cameras is traffic surveillance, including automatic number  plate recognition,’ says Matrix Vision’s Bleise. ‘The applications is particularly suitable as it requires independent units, which can then drive peripheral technologies, such as lighting or a display to warn drivers to slow down, for example. Also, as traffic surveillance often requires copies of the same programmed camera in several different locations, the smart camera provides the perfect answer.

‘Smart cameras also offer vendors the opportunity to offer pre-built solutions that can be incorporated in other machinery and serve a function. For example, in Germany there are automated bottle return machines, where consumers can reclaim a deposit when they return their empty bottles. We have provided a camera that is pre-programmed with all the necessary information, so that it can detect the bottle and read the EAN code, and tell the machine how much deposit to return to the customer. All the intelligence is in the camera, which effectively just slots into the vending machine and hooks up to its “gearbox”.’

While smart cameras do have several advantages and numerous potential applications, it would appear to be vitally important that a customer has their eyes open to the limitations of what a smart camera can achieve, before assuming that it will provide all the answers – and the cost-savings – they hope for.

Other tags: 

Related features and analysis & opinion

15 April 2019

Greg Blackman reports on CSEM's Witness IOT camera, an ultra-low power imager that can be deployed as a sticker. Dr Andrea Dunbar presented the technology at Image Sensors Europe in London in March

29 March 2019

Greg Blackman reports from Embedded World, in Nuremberg, where he finds rapid progress in technology for imaging at the edge

22 February 2019

Ron Low, Framos head of sales Americas and APAC, reports from Framos Tech Days at Photonics West in San Francisco where Sony Japan representatives presented image sensor roadmap updates

20 February 2019

Jeff Bier, founder of the Embedded Vision Alliance, discusses the four key trends driving the proliferation of visual perception in machines

19 February 2019

Greg Blackman reports on CEA Leti's new image sensor, shown at Photonics West, which contains onboard processing and is able to image at 5,500 frames per second

15 November 2018

Greg Blackman reports on the buzz surrounding embedded vision at the Vision Stuttgart trade fair, which took place from 6 to 8 November

31 October 2018

As the worldwide machine vision market continues to expand – with new trends emerging and new elements coming into play that could impact existing business models – companies are searching for those ever-important opportunities to stimulate growth.

One such trend is almost certainly embedded vision, although the technology behind it is not new, as Mark Williamson, managing director at Stemmer Imaging, noted: ‘Embedded vision is a big topic. However, it has been here a long time, because every smart camera that you buy is an embedded vision system.’

31 October 2018

Qualcomm Technologies' Snapdragon board is designed for mobile devices, but can be used to create other embedded vision systems. Credit: Qualcomm Technologies

Embedded computing promises to lower the cost of building vision solutions, making imaging ubiquitous across many areas of society. Whether this turns out to be the case or not remains to be seen, but in the industrial sector the G3 vision group, led by the European Machine Vision Association (EMVA), is preparing for an influx of embedded vision products with a new standard.