This website uses cookies primarily for visitor analytics. Certain pages will ask you to fill in contact details to receive additional information. On these pages you have the option of having the site log your details for future visits. Indicating you want the site to remember your details will place a cookie on your device. To view our full cookie policy, please click here. You can also view it at any time by going to our Contact Us page.

Industrial vision: an enabling technology

10 June 2014

Dr John Haddon, technical consultant to UK Industrial Vision Association, explains the many potential roles for vision technology, to offer food and beverage processors a competitive advantage.

Even though food processing conditions are carefully controlled, there is always some degree of variation in shape, size, volume, colour or position in the processed product which must be inspected as part of the quality control procedure. Vision can also be used for sorting raw food materials. It can be integrated into production equipment for grading and portion control for product such as bacon, cheese, and ham in order to maximise the on-weight percentages and minimise giveaway. 

Robot vision is another area of growing interest – particularly in pick-and-place applications. 

Advances in vision technology take place on a continuous basis. Higher resolution cameras with faster frame rates (or line rates in the case of line scan cameras) are regularly announced. Image processing software for both PC-based systems and smart cameras (those which have their own on-board processing capabilities) are capable of more complex measurements in faster times thanks to improved processing power. 

New vision interface standards such as CoaXPress, CameraLink HS and USB3.0 Vision offer improvements in video data transmission speeds and distances, while 3D imaging systems are now routinely affordable and a number of new 3D imaging techniques has been introduced. New, higher power LED illumination sources have become available and the use of near infrared (NIR) imaging allows the viewing of subsurface defects in foods such as fruit, vegetables, nuts and meat. 

This extensive range of cameras, lenses, lighting and software may seem bewildering, but choosing the best system starts with defining and understanding the particular problem that you are trying to solve. Do accept guidance from vision experts who can combine their specialist knowledge with the food processors’ expertise to come up with the best solution. 

Understanding the problem
Good image quality is key to a robust and effective vision system, and this is dependent on camera resolution, lighting, positioning and other factors. Choosing the most suitable vision system requires a good understanding of the requirements of the application, which will include criteria such as:
• Number of inspections required per second.
• Speed of movement of the product to be inspected.
• Size of the area on the product to be inspected.
• Level of accuracy required.
• Whether single or multiple views of the product are required.
• Whether colour inspection is required.
• The actions to be taken as a result of the inspection (pass/fail/reject etc)
• Physical characteristics and location of the inspection positions.
• Interfacing to existing plant and equipment or integration into a new production line or piece of process equipment.

Environmental requirements are also important because food processing areas generally have washdown requirements, meaning that cameras and optics must be housed in food grade enclosures.  

The same approach can be taken to robot vision, which is most commonly used in ‘pick and place’ applications where it is necessary to locate an item, identify the type of object and its orientation and then get accurate co-ordinates to pick the item up. Typical considerations include the shapes and profiles of the objects to be picked; variable heights and whether the arrangement of objects is random or regular.

Interfacing vision systems with robots generally requires expertise in both robotics and image processing so to overcome this imaging interfaces for robot applications have been developed.

Optimising profits
Choosing where a vision system is used in the manufacturing process can have a significant effect on production costs. Traditional end-of-line inspection based on ‘pass/fail’ decisions can prevent defective product reaching the customer, but can result in a lot of waste – in terms of the amount of defective product produced as well as energy consumption and time used in producing the defective product. In principle, the earlier in the process that a vision system can be used, the earlier it is possible to detect if the product is out of specification and allowing preventative action to be taken.

Robot vision
The development of a working prototype of a robot for icing a cake illustrates the combination of robotic guidance with 3D vision. The challenge with icing a cake using a robot arises from the domed or uneven surface of the cake because no two cakes are identical in terms of how they are baked or how they rise during the baking process. If the height of the cake is not allowed for, the icing process will fail because there has to be an optimum distance between the cake surface and the icing tool. The camera is a stereo vision unit that utilises a multi-beam laser. The laser beams illuminate the top of the cake, and the vision software uses the laser lines to create a 3D model of the cake within a second, generating a height map on the fly which can be combined into a string of data which incorporates not only the X and Y coordinates for the icing text but also a height value for every point.  

Confectionery
Confectionery manufacturers need to be able to carry out checks on the geometry of chocolate bars to ensure they conform to specifications. A system has been developed which uses two line scan cameras in an automatic system for random sample testing of the irregularly chocolate bars. The bar is moved with constant speed between two line scan cameras equipped with telecentric lenses to record plan and side views. Measurements are made to an accuracy of 100 µm. Images of the bars are displayed, along with the maximum recorded values for length, breadth and height. These are marked red or green to show clearly whether the test results lie within the required tolerances. Because of the irregular surface structures, several hundred points on the bar are measured, depending on the size of the bar. The test results are passed on to a further program for statistical evaluation for quality assurance. 

Hot cross buns
The visual appearance of hot cross buns is important. A custom-designed vision system was developed using one of the latest image processing software platforms to identify poor crossing quality and centralization, colour and cross variation and poor cross centralisation on hand crossed buns. The system also identified other common ‘defects’ as well as monitoring and reporting on the bake colour. A height measurement resolution of +/- 0.3 mm could be achieved on the selected area of interest.   The introduction of this vision system was used as the key driver for change in quality performance at this bakery which resulted in an increase in orders from 340,000 to 15 million buns in a single year and reduced the incidence of customer returns and return claims.  

2D and 3D imaging
A combination of 2D and 3D imaging has been used to control the size, quality and even weight of sandwich biscuits, produced by combining two biscuits with a fondant filling. Each side of the biscuit needs to be measured to calculate how much filling is required to make the 'sandwich biscuit' the correct overall height, thus minimising wastage of the expensive filling. 2D cameras are used to grab images of the biscuits from above to measure the length and width. An additional camera is used with laser triangulation to reconstruct a 3D image and get accurate height measurements. All this is done on 30 rows of cooked biscuits with 120 biscuits per row (3600 in total) passing the inspection position each minute. Using this procedure, accuracies of ± 0.17 mm can be achieved for length, width and thickness measurement.

This article has given just an idea of what vision can offer the food processing industry and and hasn’t even touched on the applications for vision in packaging – from logo positioning confirmation to checking sell-by dates.

Thanks are due to UKIVA members Multipix Imaging, Scorpion Vision and STEMMER IMAGING for their contributions to this article.


Contact Details and Archive...

Print this page | E-mail this page

RELATED CONTENT...


Article image A new imaging method for food inspection

Hyperspectral imaging is a novel imaging approach that allows us to see things that traditional machine vision cannot show – namely the chemical composition of organic materials. This leads to a host of applications within the food industry. Full Story...

Article image Bin-picking made simple

Suzanne Gill reports on a combined vision and robotic solution that could help to cost-effectively automate bin-picking applications in the food sector. Full Story...

3D Stinger Gives Machines Super-Vision

Spirax Sarco introduces delivery within 48 hours for Spira-trol™ control valves

Bühler takes a giant leap forward in intelligent optical sorting for rice with the all-new SORTEX S UltraVision™

RELATED SPONSORED ARTICLES...


Article image A new imaging method for food inspection

Hyperspectral imaging is a novel imaging approach that allows us to see things that traditional machine vision cannot show – namely the chemical composition of organic materials. This leads to a host of applications within the food industry. Full Story...

Article image The Simple Vision Sensor from SICK that’s as Smart as a Camera

SICK has developed a new vision sensor with dimension verification, position detection and intelligent measurement capability that challenges 2D smart camera performance.Full Story...

Vision inspection supports your food safety programme

MOST VIEWED...


Article image Spray and save on the glazing process

Food glazes are widely used in the bakery sector to improve the look and taste of baked products. Traditionally, this coating process has resulted in substantial waste. Technology advances mean that this is no longer the case. Full Story...

Article image Getting the Industry 4.0 journey started

Suzanne Gill finds out why the UK food industry needs to embrace the idea of Industry 4.0 and why the journey needs to start now. Full Story...

Your flexible friend in the food factory

Don’t get left out in the cold this Christmas

Automating the temperature monitoring task

http://www.appetite4eng.co.uk