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.

Machine vision applications

01 October 2007

Originally used for inspection, modern digital cameras have combined with high speed communications, increased processing power, and highly articulate robots to extend vision applications into areas that were previously thought impossible. Here are five examples of new technology that would have been impossible five years ago.

On track with machine vision
Nagle Research in Austin, Texas, USA integrated two of Sick IVP's 3D Ranger machine vision cameras onto a special truck platform for Georgetown Rail's railroad inspection system. Said to be the first system of its kind anywhere, it looks for defects in the railroad crossties and fastening hardware and can travel at speeds up to 50 km/hr, inspecting 70,000 crossties per hour. Detailed reports are automatically generated within hours. Manual inspection, done by a person walking along the track, is much slower and more subjective.

The challenge for Nagle Research was the condition of the railroad crossties: the chemical treatments used on them caused discolouration, making standard two dimensional analysis impossible. The Ranger camera allows measurement in all three dimensions to a resolution of 2 mm. (useful links:,

Textile manufacturing
Textile manufacturing is renowned for product variability, and small defects created at one stage of manufacturing carry forward to the next. Textile industry products range widely, from traditional woven or knitted fabrics for clothing to fibreglass and technical textiles for automobiles and body armour.

Traditionally, inspections are performed in two areas. First, machine operators keep an eye on manufacturing processes and make adjustments as required to keep production within acceptable bounds.
At production speeds sometimes exceeding 150 metres per minute, human quality assurance is limited to
identifying gross defects. Second, detailed off-line inspection occurs some time after manufacturing. The delay results in multiple manufacture-induced faults. The process often requires multiple inspectors to keep
pace with production, and abilities to distinguish flaws and apply defect standards vary by inspector.

Shelton Vision Systems' WebSpector surface inspection system provides quality feedback in real time and at production speeds. It can identify and accommodate points of variation, including product size, width, colour, production speed, environment, and textile construction complexity.

Products inspected by WebSpector include fibreglass sheet, cellulose and glass paper, aluminium coil,
lithographic coating, and textiles. The system runs up to 200 metres per minute on a fibreglass production line, checking for manufacturing and structural defects. It operates to a level of accuracy and consistency impossible for human operators even at a slower 20 metres per minute rate.

Robot knife sharpener
You might say it's the cutting edge of robotic vision. SIR, an Italian machine builder based in Moderna, created a vision-guided robotic cell for re-working worn out knives. Since the original shape of the knife is unknown, a machine vision system is used to examine it and make the best guess as to what it was.

To begin the process, the Kuka robot positions the knife under the vision system, which provides a shape
estimate based on wear. The vision system first recognises the knife handle and notes its position; then the blade is scanned to calculate points necessary to reconstruct the original shape.

After vision analysis, the system selects a standard profile to restore the knife's original shape. A second analysis verifies the degree of blade use to correct working parameters, such as speeds and incidence angles. The next step is to decide what point to start the work from, to avoid ruining the handle. The robotic arm moves the knife to grind each side separately, and then scours it to make the blade edge smoother. At the end, cold trimming eliminates tailings.

The Kuka robot includes a Cognex vision system with VisionPro software and PatMax technology, which uses geometric information in place of pixelgrid based correlation to locate objects without regard to their angle, size, or appearance. (

Counting free-falling parts
In industries that make ball bearings, chemical pellets, and pharmaceuticals, there is a strong demand for systems that accurately measure counts, times, and positions of objects falling in high speed and high rate. Such systems can improve manufacturing processes and quality control. Previously developed techniques have fallen short:

....Grease belt systems do not take realtime measurements and require¡¤ extensive post-measurement processing;

....The LED / photodetector grid provides real-time, high-speed measurement, but has poor spatial resolution, limiting the capability to measure small objects (<4 mm), and is unable to resolve multiple objects forming a clump;

....A machine vision based system with one line scan camera demonstrated better results than the grease belt and LED / photodetector grid methods, but the one camera design did not distinguish object clumps, or
multiple objects that are too close and appear to be one object.

The goal for V I Engineering was to design and develop a system for farm implement manufacturer John Deere to measure time spacing and XY position of fast-falling objects, distinguishing objects in a clump, while delivering counting accuracy higher than 99%, at 200 parts per second.

V I Engineering devised a machine vision system with IEEE 1394 line scan cameras and backlighting units that more than doubles that rate. Special image acquisition algorithms were developed that exceed the goal. The minimum detectable object size is under 1 mm, the maximum is more than 25 mm, and the rate of falling objects can exceed 450 per second. National Instruments LabView, the NI Vision Development Module, and NI-Imaq for IEEE 1394 were used in system development.

With the backlighting, each falling object appears as a black particle in a white background regardless of
surface condition, brightness, and colour of the falling object, so the vision algorithm does not need to be
adjusted according to an object’s appearance. The vision algorithm matches and identifies all objects in
the two camera images and separates ‘clumped’ objects.

Quality inspection, accurate rejection
Mold Rite, a manufacturer of containers and closures for the pharmaceutical industry, sought to improve the quality of automated production inspection of tightly controlled pharmaceutical container closures. Different colours and sizes of caps and closures were to be inspected for more than 10 failure criteria, at 1,200 caps per minute. A qualification test procedure was designed to ensure failure capture rates in the 95% to 99% range, well beyond prior inspection methods.

Dave Cross, Mold Rite automation manager who installed the system, said caps range from 25 mm to 100 mm in diameter, and come in many colours from white to black. Also, the cap liner ranges from white to black to foil.

Siemens worked with Mr. Cross and promised the requested 95% performance. Human eyes see 20 caps
per second as a blur, but the lights are strobed simultaneously as each part is presented under the camera by a pocket wheel or conveyor belt.

A custom user interface enables an operator to quickly set up the vision system to inspect different cap and
liner colour combinations. Siemens Simatic 1744 Visionscape Accelerated Frame grabber is mounted in a fast PC connected to two progressive scan double speed cameras. A Simatic Opto IO board provides an interface to the PLC, receiving triggers and control information and sending pass / fail and system status information. Mold Rite now has one system throughout the plant that is capable of working on all cap types, allowing operators and technicians to become ‘experts.’

‘The entire cost of system was about two-thirds of what we expected,’ says Mr. Cross. Return on investment,
though, originally figured at about a year, was reduced to nine months, he says.

Contact Details and Archive...

Related Articles...

Most Viewed Articles...

Print this page | E-mail this page