Machine vision and image processing

Food packaging system uses multiple cameras

2011-06-03T14:37:00.000-07:00

Based in Taastrup, Denmark, Superfos a/s is Europe’s largest manufacturer of injection-molded plastic packaging, supplying packaging for food, non-food and healthcare markets. To ensure quality assurance of their products, Superfos partnered with TriVision a/s,, to implement a machine vision inspection system on their Randers, Denmark production lines for butter and dairy products. TriVision’s Packaging Inspector can detect defects such as overmolding in injection-molded plastic containers and check printed labels. For the inspection of square butter boxes, TriVision selected Stingray F-046B cameras from Allied Vision to control material defects. Two cameras are positioned below a conveyor belt and inspect the inside surface of the box as it passes above them. Simultaneously, an AVT Stingray F-201B inspects the printed artwork from above to check the correct position of the label and barcode, etc. Round cream cups are checked using a combination of Stingray F-146B and Marlin F-145C cameras. The Stingray F-146B camera inspects the inside surface of the cup as it passes below it on a conveyor belt, while a lateral Marlin F-145C camera checks the presence of the label on the outside surface. For more information, click HERE.

Don Braggins remembered

2011-06-03T13:58:00.000-07:00

- by Conard Holton, Editor in Chief, Vision Systems Design magazine. Reprinted with permission.

Don Braggins, a long-standing and highly respected figure in the machine vision industry, has passed away at age 70. Founder of Machine Vision Systems Consultancy in Royston, England, in 1983, Don specialized in image processing and analysis and was a frequent contributor to and participant in organizations such as the European Machine Vision Association and the UK Industrial Vision Association (UKIVA). First as a founding member of the UKIVA in 1992, he became its director in 1995, and helped guide its development for many years. He remained a consultant to the association until diagnosed with an inoperable brain tumor in 2010.

Traveling frequently with his wife Anne, Don was welcomed by companies, universities, and trade organizations around the world for his experience, insights, and good humor. Before establishing his own company, he was product marketing manager for image analysis products at Cambridge Instruments. A graduate of Clare College, Cambridge University, he was a Chartered Engineer and a Fellow of SPIE.

Machine Vision Systems Consultancy was known for its independence as a source of information about machine vision products and services. Clients varied from multi-nationals, to startup companies, venture capitalists, and OEMs.

As editor of technical journals and frequent contributor to trade press magazines, Don regularly researched the European market for industrial vision systems for individual clients and associations. Between 2000 and 2002 he served as a non-executive board member of Fastcom Technology, a Swiss spinout from EPFL Lausanne. He was also a board member of Falcon Vision in Hungary, providing international marketing advice and technology sourcing, and introduced Falcon to the French company Edixia, which subsequently bought a controlling stake.

“Don knew the machine vision industry like the back of his hand,” remembers Andy Wilson, Editor of Vision Systems Design. “You could always rely on him to direct you towards the latest developments and innovations shown at a trade show. He was not only knowledgeable but would freely share his valuable opinions and thoughts with anyone who cared to ask. I will miss him.”

In addition to his wife Anne, Don is survived by two children and five grandchildren.

The staff of Vision Systems Design extends our sincerest condolences to the Braggins family.

Microscopic lens enables 3D imaging

2011-06-02T14:29:00.000-07:00

Engineers at Ohio State University have invented a lens that enables microscopic objects to be seen from nine different angles at once to create a 3D image. Other 3D microscopes use multiple lenses or cameras that move around an object; the new lens is the first single, stationary lens to create microscopic 3D images by itself. Allen Yi, associate professor of integrated systems engineering at Ohio State, and postdoctoral researcher Lei Li described the lens in a recent issue of the Journal of the Optical Society of America.Yi called the lens a proof of concept for manufacturers of microelectronics and medical devices, who currently use very complex machinery to view the tiny components that they assemble. Though the engineers milled their prototype thermoplastic lens on a precision cutting machine, the same lens could be manufactured less expensively through traditional molding techniques, Yi said. For more information, go HERE.

Tea time for Justin

2011-05-27T11:00:00.000-07:00

Making a cup of tea or coffee is surprisingly complex. Yet this seemingly unexceptional challenge has been exercising the mighty brains of robotics experts for the last twenty years. One such expert is Dr. Michael Suppa, head of Department Perception and Cognition at the Institute of Robotics and Mechatronics at the German Aerospace Centre (DLR). His team has been working on Rollin’ Justin, a robot that can interpret objects with a stereo camera system and understand their significance to the task. Dr. Suppa is working with researcher Dr. Andrew Davison at Imperial College, London in robot navigation. “Once the robot starts moving, the scene is updated - a very graphics intensive task”, says Davison. Even when Justin knows what objects are, it still has to decide where to grip the object and how tightly. As research partner, Dr. Jeremy Wyatt at University of Birmingham explains, “Justin uses learning as a predictor to both mass and the forces needed to be applied.” For more information, click HERE.

Vision for service robots

2011-05-26T13:50:00.000-07:00

Vision for Service Robots is a new report from Vision Systems Design that provides a comprehensive review of the technologies, markets, and opportunities for suppliers of vision components, the researchers and integrators who are building service robots, and companies and organizations interested in entering the market. This groundbreaking 120-page report is based on extensive interviews and original research compiled over eight months by Adil Shafi, president of Advenovation, and Conard Holton, Editor of Vision Systems Design. It identifies multiple near-term market opportunities for vision components and systems totaling several billion dollars, and opportunities in the longer term that promise to be even greater. For more information, click HERE.

Win an Apple TV digital media receiver

2011-05-25T08:00:00.000-07:00

Cognex’s latest DataMan 500 fixed-mount barcode reader is the subject of a 30 min webcast scheduled for Thursday, May 26, 2011 at 2:00 PM (Eastern Daylight Time - US) EDT. Presented by Matt Engle, Manager of ID Products at Cognex, the webcast will give attendees the chance to win an Apple TV digital media receiver (approx. value $99 USD) by registering for the event. The Contest is only open to legal residents of the USA ages 18 and over currently living in the 50 United States and the District of Columbia (void in U.S. overseas territories and wherever else prohibited and/or restricted by law). Those who would like to register for this event, click HERE. It’s worth a look. I saw the product in action and it is rather impressive.

Thermal imaging analyzes automotive welds

2011-05-24T10:35:00.000-07:00

Surface welding instead of penetration welding, allows a laser to produce a weld that is only visible on one side. But how do you control the laser power to prevent it burning a hole through the sheets of metal? A new camera system analyzes thermal images in real time – and ensures a perfect weld. As if controlled by an invisible hand, the welding head on the robot’s arm races along the sheet metal parts. Where the laser hits, sparks fly and the metal glows red hot. The process lasts just a few seconds. The outer door panel and the door frame are then welded together perfectly. A thin weld seam extends along the join, but it can only be seen on one side. From the other side of the welded car door the join is invisible. For more information, click HERE.

Steve Montellese (1951-2011)

2011-05-23T13:37:00.001-07:00

It is with great sadness that we must report the passing of Steven Montellese who at 59, died in Pittsburgh, PA on May 20, 2011, surrounded by his family and friends. Steve was born August 10, 1951, in New York City. He attended college at the City College of New York and graduate school at Brown University. He was a brilliant man with many great achievements, and he filled many roles. He was a pioneer in computer vision, a legendary host, a food critic, an author, an inventor, a philosopher, a world traveler, a brother, a son, and a father. From 1986, Steve was president of Millennia 3, a company that developed and marketed low cost, custom machine vision solutions for the industrial manufacturing industry. M3 was the first company to market personal computer based vision systems, as well as expanding the technology to three dimensional gauging and color machine vision. Perhaps many of our readers will remember him most as the Manager of Applications at Automatix, Inc (Billerica, MA) where he was responsible for product planning, marketing support, and corporate strategic planning. Steve is survived by his father, Pat Montellese; brother, Mark Montellese; son, Eric Montellese; daughter, Julia Montellese; and two grandchildren. He was preceded in death by his mother, Libby Montellese. A visitation will be held at H.P. Brandt Funeral Home from 2 pm to 4 pm, followed by a memorial service at 4 pm. In lieu of flowers, donations should be sent to the Pittsburgh Cultural Trust, 803 Liberty Avenue, Pittsburgh, PA 15222. Phone: (412) 471-6070.

Vision system checks food packaging

2011-05-20T11:40:00.001-07:00

Based in Taastrup, Denmark, Superfos a/s is Europe’s largest manufacturer of injection-molded plastic packaging, supplying packaging for food, non-food and healthcare markets. To ensure a 100% quality assurance of their products, Superfos partnered with TriVision a/s, a leading Danish machine vision solution provider, to implement a machine vision inspection system on their Randers, Denmark production lines dedicated to containers for butter and other dairy products. For the inspection of square butter boxes, TriVision selected Stingray F-046B cameras to control material defects: Two cameras are positioned below a conveyor belt and inspect the inside surface of the box as it passes above them. Simultaneously, an AVT Stingray F-201B inspects the printed artwork from above to check the correct position of the label and barcode. The lids used to seal the containers are controlled on separate lines using Stingray F-201C color cameras. For more information click HERE.

Haddon replaces Braggins at the UKIVA

2011-05-19T11:10:00.000-07:00

Dr. John Haddon has been appointed as Technical Consultant for the UK Industrial Vision Association (UKIVA), following the retirement of Don Braggins. John is the founding director of Panther Vision Ltd, which joined the Association towards the end of 2009. John will be the first point of contact for anybody requesting technical advice from the Association, and will contribute articles to the trade and technical press as part of the Association’s continuing objective to raise and maintain awareness of the capabilities of vision technology in manufacturing industry. He has been a visiting Professor at the Centre for Vision, Speech and Signal Processing at the University of Surrey and a Visiting Fellow at both Surrey University and the University of Exeter. For more information, click HERE.

Unmanned submarines locate air crash scene

2011-04-14T09:56:00.000-07:00

Nearly two years after the dramatic crash of Air France flight 447 between Rio de Janeiro and Paris, France’s air accident investigators finally discovered the wreckage of the Airbus A330-200 aircraft 600 miles off the Brazilian coast on the floor of the Atlantic ocean at 3,900 m (12,800 ft) below sea level. Three REMUS 6000 autonomous underwater vehicles (AUVs) were used to search the sea floor for the wreckage. The REMUS 6000 is an underwater drone manufactured by Hydroid Inc. for deep water exploration, where depth and pressure are too high for human divers. The autonomous submarines searched the sea floor with sonar detectors for the wreckage and finally found it at nearly 4,000 m depth. Using their on-board Pike F-421B digital cameras from Allied Vision Technologies, they could transmit first images of the aircraft to investigators. For more information, click HERE.

Looking to hire?

2011-04-11T17:18:00.000-07:00

Last week, I was contacted by someone (who will remain nameless) about obtaining a position in the image processing/machine vision business. Here is the brief resume I received:

Strong C/C++, lots of embedded experience assembly on TIC64x DSPs, ARM7s (wrote a mini OS), designed and wrote a simulator for his own architecture, has VHDL/Verilog experience and wrote a peak detector. His thesis is on 3D face tracking using the Kinect system from Microsoft. He’s looking (perhaps) to move to the USA.

Does anyone know company that could use this sort of experience? If so, please leave a message on this blog with your contact information.

Also, if you are in the same position, why not post your resume as a message?

Inspection of high-volume baking processes

2011-04-07T12:30:00.000-07:00

The baking industry is the third largest segment of Georgia’s food processing industry (13% by sales volume) with operations located throughout the state. One of the largest segments of the market is buns and rolls for food service and fast food customers. A growing number of these customers (Arby’s, Wendy’s, Burger King, McDonald’s) are placing increasing demands on quality control for bun size, shape, color, and topping coverage (sesame seeds, etc.). Accurate control of the quality is challenging considering the highvolume production, with rates of up to 1,000 buns per minute on a line. The high volume also indicates the need for automated control (correcting a drifting color before it goes out of specification would save thousands of buns). The standard inspection process is for workers to remove a few samples of the product each hour and to inspect both the top and bottom manually against customer specifications. Customers are pushing for a more accurate and uniform assessment process, and a means to actively control this product. This article describes an imaging system to inspect the bottom color and characteristics of baked buns for automated feedback to an oven control system. For more information, go HERE.

Machine vision in medical applications

2011-04-06T12:22:00.001-07:00

The health service is under enormous pressure. Firstly, individual preliminary medical services and health care should be available with high quality standards while secondly, costs should be reduced. “Machine vision can be a key technology to combine these contradictory demands”, said Alexander Temme of Basler. That’s because camera-based medical technology can help to improve early detection, diagnosis, treatment, archiving and training methods. VISION 2011 will pay closer attention to this future-oriented topic according to its organizers. “Feedback from VISION visitors in the last two years revealed that the topic of medical technology is becoming increasingly more important. This industry is now regarded as the fifth most important visitor industry”, said Florian Niethammer, VISION Project Manager. Devicemed will support the key topic of medical technology as a media partner. The VDMA Machine Vision Group is also planning to hold talks relating to medical technology applications during Industrial VISION Days. VISION 2011 will be staged for the 24th time at the Stuttgart Trade Fair Centre from 8 to 10 November 2011. For more information, click HERE.

Object tracking algorithm learns patterns

2011-04-05T11:27:00.000-07:00

Popular Science reports that a Ph.D student at The University of Surrey named Zdenek Kalal has released a video demonstrating an object tracking algorithm he has developed, which he calls Predator, for video image streams. Kalal, can be seen on a video rolling around on his desk chair in a demonstration of the camera’s tracking abilities. After selecting something for Predator to focus on with a bounding box, the system begins recognizing patterns, learning how that object looks at different distances and angles, and even finding it amongst a sea of similar objects. When Kalal tells Predator to track his face, it can pick him out of a page full of small photos of other people. For more information, click HERE.

Kinect sensor targets palletization problems

2011-04-03T09:07:00.000-07:00

Eye Vision Technology (EVT) has used the Kinect sensor from Microsoft - originally developed for the Xbox 360 - with its EyeScan 3D system for robotic applications such as depalletization and sorting components on the assembly line. The EyeVision system combines the depth sensor and a camera for applications such as bin picking using a robot to find an object’s spatial position. The robot-gripper is then navigated to a grip position with EyeVision software to execute tasks. For more information, go HERE.

Vision inspects power train systems

2011-04-01T12:22:00.000-07:00

ThyssenKrupp System Engineering (Auburn Hills, MI, USA) is a designer and manufacturer of turn-key assembly and test systems for automotive power train components, including engines, transmissions, and axles. Recently, a manufacturer of construction and heavy equipment recently commissioned the company to build an engine assembly line for a new plant in the U.S. Southwest. In addition to its role as a line builder, ThyssenKrupp had the mandate to provide this factory with a machine vision solution for the entire engine assembly line. Acting as the system integrator, ThyssenKrupp supplied the vision, mechanical and electrical systems, PLC connectivity, and training of plant staff. The first steps of the vision solution were developed using six Matrox Iris GT smart cameras. As an engine moves down the line, it goes through multiple smart camera-based vision inspection stations. These applications vary from absence/presence to gauging/measurement. The addition of machine vision to the line came about when the customer expressed concern with the final test machines failing engines due to operator errors during assembly. ThyssenKrupp presented a camera-based solution similar to what they had developed in the past for other engine builders. This solution would use vision inspection at the station where the parts were being installed. Any errors would then be fixed at the assembly station. Fore more information, click HERE.

Machine vision maps marine wildlife

2011-03-31T12:07:00.000-07:00

Researchers at the National Oceanic and Atmospheric Administration's (NOAA) - National Marine Fisheries Service (NMFS) working off the coast of Alaska, are using machine vision to map underwater marine wildlife populations. Traditional assessment is performed with a net and cod end where the fish are caught and brought aboard for species, length, and sex identification. This new "non-lethal fish sampling device" uses stereo optical sampling to obtain these characteristics, preventing the accidental destructive sampling of thousands of fish per year. The "Cam-Trawl" camera system sits at the end of a trawl net that has the cod end removed. The net simply herds the fish past the camera where they are imaged and then free to swim about their way through the other end. For more information, go HERE.

Image analysis of agricultural crops

2011-01-14T12:41:00.001-08:00

A team of researchers from the Universidad Complutense de Madrid (UCM) and the Center for Automation and Robotics (CAR) of the Consejo Superior de Investigaciones Cientificas (CSIC) has developed a method for identifying relevant texture images of agricultural crops using techniques based on the combination of color information of images. The system distinguishes between three main types of textures (plants, soil and sky), avoiding unnecessary image analysis where not required. For more information, click HERE.

Vision system moves to view its target

2011-01-13T12:56:00.001-08:00

Traditionally, vision systems used for quality inspection have been installed at fixed points on assembly lines, with the inspected items moving past on a conveyor belt. Festo claims to have turned this process on its head with a vision system that is fixed to an axis gantry that moves to view the items. In one of the first applications of the technology, a German vision system integrator called JAM Automation has developed an assembly and inspection cell for an automotive subcontractor. The machine first presses nine soft plastic sockets into a flange plate made of a harder plastic. The vision system then moves automatically to each socket position to carry out a 100% inspection. For more information, click HERE.

Wind curtain keeps camera lenses clear of debris

2011-01-12T08:50:00.000-08:00

Dust, drizzle and spray water are common obstacles for the installation of camera systems in rugged industrial environments, especially if the application requires the camera to face upwards. For such tasks, autoVimation has developed an air curtain which reliably protects the front panel of the camera enclosure from dirt, thus allowing for use even in very demanding environments. The ultra-compact air curtain, which can be easily attached to all autoVimation enclosures, merely adds 10 mm to their length. Pressurized air (5 bar, 50-100 l/min) enters through a control valve at the side of the air curtain unit, and is then expelled through nozzles directly above the enclosure's front panel. The nozzles are arranged to create a “tornado effect”, which efficiently deflects any foreign bodies flying towards the front panel while using very little air, thus providing a clear view for the camera. In most cases, an air flow of 50 l/min is sufficient. At diagonal image angles below 30°, an additional tube can be attached to increase efficiency. The air curtain can be operated continuously or shortly before image acquisition in order to clean the front panel. Practical tests have shown that the system keeps even oil mists and aluminum chips away from the front panel or removes them efficiently. For more information, click HERE.

Cameras catch traffic offenders

2010-09-21T10:27:00.001-07:00

Traditional traffic enforcement devices require laser, radar or inductive loop sensors and a camera to provide documentation of the traffic violation. These systems are limited to vehicle speed measurement, one vehicle at a time. Because of these limitations, Italian traffic control company Kria developed T-EXSPEED, an automated system that relies on cameras and stereo software algorithms to track and detect traffic violations. T-EXSPEED operates without the use of external sensors and can detect multiple traffic violations such as excessive speed, red light violations, forbidden left or right turns directly from images taken by industrial cameras that repeat the real-time stereo process 24/7. The system can measure vehicle speed up to 300 km/h and track vehicles in parallel (up to a three-lane footprint), in both directions. The original T-EXSPEED system is based on three Prosilica GC cameras while the newer and T-EXSPEED 2, relies on three Prosilica GB cameras. For more information, click HERE.

Evaluating facial recognition techniques

2010-09-20T08:13:00.000-07:00

Rapid improvements in facial-recognition software mean airport security workers might one day know with near certainty whether they’re looking at a stressed-out tourist or staring a terrorist in the eye. A research team led by Dr. Alice O’Toole, a professor in The University of Texas at Dallas’ School of Behavioral and Brain Sciences, is evaluating how well these rapidly evolving recognition programs work. The researchers are comparing the rates of success for the software to the rates for non-technological, but presumably “expert” human evaluation. “The government is interested in spotting people who might pose a danger,” O’Toole said. “But they also don’t want to have too many false alarms and detain people who are not real risks.” The studies in the Face Perception and Research Laboratories are funded by the U.S. Department of Defense. The agency is seeking the most accurate and cost-effective way to recognize individuals who might pose a security risk to the nation. For more information, click HERE.

Computer vision systems decipher outdoor scenes

2010-09-10T10:36:00.000-07:00

In much the same way that a child might use a set of toy building blocks to assemble something that looks like a building depicted on the cover of the toy set, the computer would analyze an outdoor scene by using virtual blocks to build a three-dimensional approximation of the image that makes sense based on volume and mass. “When people look at a photo, they understand that the scene is geometrically constrained,” says Abhinav Gupta, a post-doctoral fellow in CMU’s Robotics Institute. “We know that buildings are not infinitely thin, that most towers do not lean, and that heavy objects require support. It might not be possible to know the three-dimensional size and shape of all the objects in the photo, but we can narrow the possibilities. In the same way, if a computer can replicate an image, block by block, it can better understand the scene.” Gupta’s approach to automated scene analysis could eventually be used to understand the objects in a scene and the spaces in between them and what might lie behind areas obscured by objects in the foreground. That level of detail would be important, for instance, if a robot needed to plan a route where it might walk. For more information, click HERE.

Open source vision sensor reads data matrix codes

2010-09-09T14:11:00.000-07:00

The most important difference between software created by the open source communities and commercial software sold by vendors is that open source software is published under licenses that ensure that source code is available to everyone to inspect, change, download and explore as they wish. In this article, Anthony Oliver shows how to use open source software to build your own open source vision system to read a data matrix code. For more information, click HERE.