Added Features Give Image-Based Code Readers a Boost Over Laser Scanners
| By: Winn Hardin, Contributing Editor
But that is one of the primary challenges faced by image-based auto identification (auto ID) code readers as they try to supplant laser scanners in lucrative manufacturing, warehouse, and logistics markets. Other challenges include cost and depth of field, or ability to read codes randomly placed in a large volume of space, such as the wide conveyors used in courier distribution centers and retail-fulfillment warehouses.
According to Darrell Owen, vice president of marketing for Datalogic’s (Minneapolis, MN) Identification Business — which manufactures laser scanners, image-based readers, and machine vision systems — image-based code readers have achieved parity with laser scanners in manufacturing industries, driven in large part by the compatibility between 2D codes and direct part marking. In the warehouse and courier industries, laser scanners still have an edge in terms of sales and units deployed. However, image-based code readers are growing faster than their laser-scanning competitors.
“In automation 1.0, it was all about reading the code,” Owen says. “Today, in automation 2.0, accurately reading the code more than 99.9% of the time is a given. In the past, every new generation of laser reader gave a huge increase in efficiency to the customer, but that’s not the case today, where improvements yield points of a percent in efficiency gains. So now, customers are being driven to image-based readers where they can see what’s happening instead of just seeing the results as you do with a laser. Instead of just ‘read’ or ‘no read,’ the customer can see why the read failed. Is it a bad printer, was the box upside down, etc.
At the same time, Owen adds, the cost for 2D imagers has been decreasing as performance increases. “It’s pretty much following Moore’s law where every 18 months, performance doubles while costs stay the same,” he notes.
Manufacturers, particularly those in the pharmaceutical, medical device, and growing numbers of consumer packaged goods (CPG) markets, are being driven toward using 2D codes (data matrix, QR, maxicode, etc.) instead of 1D barcodes for a combination of regulatory and business reasons. While governments are requiring pharmaceutical and medical device manufacturers to mark each product down to the single-unit level for safety reasons, many consumer goods manufacturers need proof that what they shipped is what the customer ordered, again necessitating single-unit traceability. In these cases, 2D codes better accommodate single-unit serialization. The digital nature of the 2D markings (the mark is there or it’s not) is also more robust for dense symbologies than the analog nature of the barcode, which encodes information based on the analog nature or width of each line, according to John Agapakis, director of American sales at Microscan Systems Inc. (Renton, Washington), a supplier of laser scanners, image-based auto ID readers and smart cameras, and the original developer of the data matrix 2D code.
Microscan’s auto ID systems are used widely in manufacturing industries, which Agapakis points to as the entry point to the logistics supply chain. “In addition to the amount of information that a 2D code can contain, 2D codes can be applied directly to the product using a number of methods, such as laser etching and dot peen direct-part marking,” Agapakis says. “The growth of image-based auto ID systems in manufacturing markets is connected to the compatibility of direct-part marking and 2D codes, which laser scanners cannot read. But that doesn’t mean laser scanners are done.”
Agapakis adds that laser scanners dominate in another key industry for Microscan, namely high-speed document scanning as well as low-cost code readers integrated into larger OEM machines that need 1D code reading functionality. Outside of manufacturing lines, where part location is controlled and product variability on a single line is limited, image-based systems still lag behind the use of laser scanners.
“In warehouse and courier facilities, image-based systems with fast autofocus can struggle to read codes on boxes that may be located anywhere across a large conveyor and moving at high speed,” says Datalogic’s Owen. “That’s where lighting and optics really come into play. In these environments, laser scanners are still king of the hill versus 2D readers.”
Despite laser scanners’ greater depth-of-field advantage for warehouse and courier facilities, the growing use of 2D codes has translated to faster growth for image-based readers. While Owen says few customers fully utilize the error correction and statistic process control (SPC) capabilities that machine vision systems afford, they do appreciate having an actual image to look at. To overcome lower read rates from image-based systems while still using automation to push throughput, warehouse and courier facilities use a layered auto ID approach. Each box or carton will pass two to three automated code readers. If the first system fails to read, an image-based system will acquire the code and try to read type on the label using optical character recognition (OCR). If both systems fail to read, the box can be removed from the main sorting line, with the images sent to a central workstation or manual rework area where a human can look at the image and see what’s wrong. In most cases, the information can be keyed in remotely and the package relabeled and sent back through the automated sorting lines, maintaining the high throughputs that courier applications require.
“Lasers still have a place if you’re just talking about cost differentials, but the trade-off is the advanced capability that image-based readers give you,” says Jay Stone, vice president of sales and marketing at Vitronic Machine Vision Ltd. (Louisville, Kentucky). “If a package has to go to a rework station, that adds a lot of cost to the operation.” Vitronic provides line-scan-based tunnel readers for high-speed image-based auto ID applications and hybrid systems, such as the VICAMsnap!, which can operate as a fully automated stand-alone or as part of manual-scan system.
“We’re seeing a lot of customers wanting compliance image archiving as proof that the right products were shipped to the right customer,” Stone explains. “We’re also seeing growing demand for OCR as a stop-gap solution when the image wasn’t processed automatically. It’s a way for high-volume operations to squeeze out a few extra fractions of a percentage in throughput without having to resort to manual sorting and rework. We’ve also recently added automated dimensioning to our logistics product mix for customers that need certified weights and measures with their shipments.”
Realizing that code reading is becoming a commodity as 1D and 2D code readers close in on 100% read rates, image-based code suppliers are leveraging the computer processing and software capabilities of their machine vision solutions to offer application-specific solutions and gain a leg on the competition. For example, Datalogic recently developed a pattern-sorting tool that not only can read a code but also compares the code data against labels and packaging on the actual product. This has been well-received in the liquor distribution market, where shrinkage is a problem due to incorrect liquor shipping in the wrong container.
Microscan has added its verification monitoring interface, which grades the quality of each code in real time and notifies customers when code quality is degrading. The interface also offers information on the root cause of the error, whether it be a bad print head or related to the label stock, and so on. Microscan also has developed PanelScan, a solution developed originally for electronics manufacturers that uses a line-scan camera to capture images of panels or cases on conveyors containing multiple printed circuit boards (PCBs) or multiple product packages for automatic unit-level identification and potentially other inspection tasks.