Saturday 21/9/2019

Bookmark and Share
Case Studies
  PandCT RSS Case Studies feed
Vision System Enables Fast and Inexpensive Upgrading of Tablet Printing Systems

15 July 2011

In order to meet the Food and Drug Administration (FDA) mandate that each tablet or capsule be identifiable, it’s critical to be able to inspect the identification and eject bad items. In the past, Ackley Machine used cameras to acquire and a personal computer to process images on its tablet printing machines. But when a computer went down, the company had to upgrade to a new version of the Windows operating systems, which required a long and expensive validation process. Another problem with the previous approach was that the single computer responsible for processing all of the images often became bogged down.

Ackley eliminated this problem by switching to Cognex In-Sight vision systems which incorporate both camera and image processing in a single enclosure. “Switching to a new vision system now requires only a short and simple validation procedure that ensures the new vision system provides the same functionality as the old one,” said Mark Ford, Engineering Manager for Ackley Machine. “The In-Sight vision system also is much easier to program because its PatMax pattern matching algorithm easily finds the printed image on the tablet with a single command.” Ackley has used this approach in its latest system, which prints and inspects over 400,000 tablets per hour, much faster than other machines, while providing the unique single tablet rejection capabilities.

Need for drug identification

Since 1975, the FDA has required that all prescription tablets or capsules have a unique identification in terms of size, shape, color and imprint. Drug makers use the imprints both to identify the chemical substance and dosage and for promotional purposes. Acceptable identification has traditionally been achieved by debossing -- pushing the image below the level of the product, embossing -- pushing the image above the level of the paper, or printing. The coating systems used today have a tendency to fill in debossed or embossed areas, sometimes rendering them unidentifiable, especially by those with less than perfect eyesight. Printing is being used more and more because it provides contrast, excellent compatibility with modern coatings, and a unique and attractive multicolor identification that can set the product apart from competition and virtually eliminate counterfeiting.

With the rise in popularity of printing has come the need to inspect printed tablets to ensure the integrity of the identification. Printing is normally accomplished by using a hard roll that is engraved with the image to be printed. The hard roll contacts the ink, then a wiper removes all of the ink except for the engraved area. The ink is then transferred to a soft roll that contacts the tablet or capsule. Inspection is needed because if the wrong ink formulation is used the ink might not adhere to the tablet, or the coating might flake off and contaminate the ink to the point that the printing becomes indistinct. Manual inspection has been used but with hundreds of thousands of tablets or capsules being printed per hour, it’s impossible to inspect more than a small sample of product. This makes it impossible for the manufacturer to be sure that they are meeting the FDA product identification regulations.

Ackley originally used a vision system consisting of a camera, an image processing card and image processing software running on a PC. The company was about to ship a system to a customer when a computer failed and had to be replaced. The computer ran on an older version of Windows that was no longer available so it had to be replaced with a computer that ran a newer version. The software, hardware, and operating system each needed to be validated to meet FDA requirements which was a long and expensive process. “We said, wait a minute, we don’t want to have this problem again,” Ford said. “So we looked for a vision system that was self-contained so that it could easily be validated based on its functionality alone.”

Vision systems offer easy validation

Ackley discovered that Cognex In-Sight vision systems provide a camera and image processing hardware and software within an enclosure that protects them from the factory environment. In-Sight vision systems are well suited to often-crowded pharmaceutical processing stations because they fit into a small 109.1 x 61.4 x 35.5 mm package. This approach makes it possible to write a short and simple validation procedure. So upgrading to a new model of vision system becomes a relatively simple procedure. The new approach also provides substantially higher performance when multiple vision systems are used because each vision system processes its own images.

Ackley Machine’s latest tablet printing machines use Cognex In-Sight vision systems to provide industry-leading speed and the unique ability to avoid even a single reject tablet. After printing the tablets, the machines drop them into the pockets of a carrier bar that holds up to 24 tablets. The carrier bar moves under an array of four Cognex In-Sight vision systems where each camera inspects six tablets. The vision system program first uses a histogram tool to check for the presence of the light-colored tablet against the dark background of the carrier bar. If a tablet is present, then the vision system performs a print quality check.

Inspecting printing quality

The PatMax pattern recognition algorithm looks for the printed image regardless of its location in the image acquired by the camera. This eliminates the need for precision positioning of the camera or tablet. PatMax identifies and isolates the key individual features within an object image and measures characteristics such as shape, dimensions, angle, arcs, and shading. It then correlates the spatial relationships between the key features of the trained image to the run-time image, encompassing both distance and relative angle. By analyzing the geometric information from both the features and spatial relationships, PatMax is able to precisely and repeatably determine the object's position without regard to the object's angle, size, or appearance.

The system is initially trained with an image of a good quality printed tablet. Operators can train the system to recognize new printing without any programming. The PatMax algorithm compares the good image to the most recently acquired image and rates the quality of the recently acquired image based on the match. The quality rating required to meet internal or customer requirements can easily be adjusted. The PatMax algorithm also returns the location of the center of the printing on the product that was inspected. This information is used to determine how well the logo is centered on the tablet. The blob tool is used to find the center of the tablet and the distance tool is used to measure the distance from the center of the printing to the center of the tablet.

Inspecting for tablet damage

The next step is checking to see if the tablet is broken. A broken tablet shows up as several light colored sections against a dark background. It is identified by using the blob tool to look for light colored blobs smaller than the tablet. The final step is to look for a coating defect which can be identified as a smaller white spot, because the core of the tablet is white. The blob tool is used to search for white spots. The camera acquires the image and performs each of these inspections and builds a binary word that contains the inspection results for each of the six tablets in its field of view. The vision system sends the results to the programmable logic controller (PLC) that runs the machine. The vision system also sends each inspection image to the corporate network where it can be called up by the machine operators or engineers.

The PLC then operates a vacuum system that picks up the tablets from the carrier bar and moves them to a discharge chute. Based on the signal from the PLC, specific positions on the vacuum shoe are operated individually to either pick up or leave behind individual tablets. The tablets that have passed the inspection are picked-up and placed in the discharge chute, while those that have failed remain in the carrier bar. In the next step the tablets remaining in the carrier bar are dumped into a reject bin. This approach makes it possible to eliminate in a positive manner individual tablets that fail inspection and ensure that only good tablets are passed along for packaging.

“In-Sight vision systems operate much faster than the modular cameras used in the past, enabling the machine to print and inspect over 400,000 tablets per hour, substantially faster than any other tablet printing and inspection machine,” Ford concluded. “The machine also offers a higher level of quality because it is able to reject individual tablets while other machines on the market can only reject batches of tablets which is wasteful if there is only one bad tablet in a batch.”

Point of contact:

Philip A. Gulotta
Sales Engineer
Ackley Machine Corporation
Ph: 856-234-3626 (office) or 856-905-7484 (cell)
Web site:

For details, visit Cognex on-line at are not responsible for the content of submitted or externally produced articles and images.
Click here to email PandCT about any errors or omissions contained within this article.
Send this page
To send this page to a colleague or friend,
fill in the email addresses below...
Your email address
Their email address:
Page to be sent: shownews.asp?ID=29696

Company gateway pages
for Cognex Germany Inc.:
FREE Information on the products in this article
To request more information from this supplier,
fill in your email address below.
Your email address
Bookmark and Share
Product categories: Accelerometers | Actuators | Agitators | Analysers | Bearings | Compressors | Controllers | Conveyors | Drives | Enclosures | Flowmeters | Heat Exchangers | Motors | Pumps | Relays | Sensors | Transducers | Transmitters | Valves | Weighing
(c) Copyright 1999-2019 Process and Control Today Ltd  |  Reg. no 3733110  |  Email Editor  |  Email Webmaster  |  Sitemap  |  Privacy Policy