Industry Insights
The 2D Symbol Reading Scene
POSTED 04/20/2005 | By: Nello Zuech, Contributing Editor
Under the ‘‘hood’‘ of a 2D symbol reader is a machine vision system. More often than not the implementation is in the form of a smart camera – in other words, a 2D symbol reader is essentially an application-specific smart camera.
2D symbols are being embraced by virtually every manufacturing industry. The distinct advantage – the ability to encode more data in smaller space – satisfies a universal requirement in most industries where space constraint is an issue more often than not. There are a number of 2D symbols that have become popular. In some cases the 2D symbol is really a stack of 1D bar codes. Most of the 2D symbol readers have the ability to read the more popular 2D symbols.
Reliable reading of 2D symbols requires the ability to find, enhance and segment the bar code as well as ‘‘read’‘ it based on pattern recognition. In other words, all the functions one finds in a typical machine vision system. The ‘‘smart camera’‘ implementation yields a low cost reader, whether hand held or fixed mounted.
Not only does one find applications in manufacturing today, 2D symbol readers are widely used in logistics applications identifying and routing cartons and other packages in a warehouse or distribution center.
Like the original 1D bar codes which took over 20 years to become widely adopted, while the underlying 2D symbol reading capability has been touted for well over 10 years, it, too, has taken much longer to find widespread deployment. The challenges have been mostly overcome: standards and cost. However, a current ‘‘threat’‘ to 2D bar codes could be RF-based identification technology. In the end, while undoubtedly RFID will impact the applications that might otherwise have been addressed by 2D symbology, it is more likely that the two approaches will complement each other in the market. For certain applications RFID will produce more favorable results while in other 2D symbology will do so.
To gain insights into what is currently taking place in the 2D symbol-reading field, input for this article was canvassed from all companies known to supply 2D symbol readers. The following provided answers to our questions, which follow.
- Jamie Pearce – Business Development Specialist, Cognex ID Products Marketing
- Dennis Kaill – President, Microscan
- John Ashodian – Product Marketing Manager, SICK
- Klaus Michael Bücher – Director Sales and Marketing, VITRONIC
1. Can you provide a brief general description of your 2D symbol reader targeted specifically at machine vision applications? Imager type, fixed mounted, hand-held, generic algorithms, lighting type, performance (throughput, read accuracy, number of characters, network connectivity, etc.), remote diagnostics, etc.?
[Jamie Pearce – Cognex] Cognex provides a complete line of products specifically targeted to address the requirements of Direct Part Mark 2D symbol reading applications. Cognex readers include fixed mount In-Sight® ID Readers and handheld DataManTM ID readers. Cognex fixed mount readers have CCD imagers while the handheld readers have a CMOS sensor. All Cognex ID readers feature Cognex IDMaxTM technology that is based on patented Cognex PatMax TM geometric pattern matching technology for object location.
Cognex In-Sight readers can be used with traditional external lighting sources, but also feature the ability to use integrated lighting (both direct ring light and a diffuse extended ring light). The DataMan 6500 readers feature the ability to illuminate components with bright field light, dark field illumination, and cloudy day lighting for highly reflective/curved surface parts.
All Cognex ID readers can read all standard ECC codes (in addition to a wide range of other symbologies including 1D, RSS/CS and QR Code). The In-Sight fixed mount readers are connected via Ethernet and support a wide range of protocols including TCP/IP, FTP, SMTP, EtherNet/IP, ModBus/TCP and more. The DataMan series of readers are connected via a RS-232 connection or a USB adapter.
[Dennis Kaill – Microscan] Microscan Systems' 2D readers utilize both CCD and CMOS technology and are available as fixed mount and hand held devices. The algorithms are proprietary to Microscan. Lighting is typically visible red LED based. Performance in regards to throughput is dependent upon the application. With Data Matrix ECC200 read accuracy is 100%. Connections include RS-232, 485, and Ethernet. Microscan has also just recently announced the introduction of the LightRay Optics. This is a patented optical lighting accessory that enables quick and easy reading for direct part marks with a hand held imager. This solution is especially valuable in aerospace applications, where operators must use handheld imagers to read direct part marks due to the large size of aero-engine assemblies.
[John Ashodian – SICK] SICK, Inc. manufactures fixed position 2D code readers for use in automotive, aerospace, electronics, packaging, pharmaceutical, document handling, and biomedical OEM applications. Laser-illuminated linear readers and LED-illuminated area array readers are available. They read 2D DataMatrix codes and all common linear bar code types. These devices are suitable for 2D coded items that are either presented in continuous motion or indexed (model and application dependent). Ethernet connectivity is built into each scanner. Additional network connectivity is also provided through a unique connectivity device (CDM- Connectivity Device Modular), which can provide a variety of field bus options, including DeviceNet and ProfiBus. An optional LCD display and cloning module, which stores scanner parameters for easy exchange, are also available for the CDM. A unique feature of these 2D readers, known as the SICK ICR 8xx series, is that they install and function as easily as a common laser based bar code scanner. In fact, they operate on a common software platform, making them compatible with all other SICK CLV laser scanners.
[Klaus Michael Bücher – VITRONIC] VITRONIC provides a large variety of different products in order to meet the specific requirements of each customer. All applications in the field of identification are PC-based systems. Sophisticated software algorithms are identifying barcodes, 2D-codes or plain writings (OCR). These codes and writings can be identified on all kinds of material in both rough industrial environments or clean office environments with high read rates.
Components:
- Cameras: matrix camera, fix focus line scan camera, high speed auto focus line scan camera, video digitizing on board
Special characteristics:
- Fiber optics for image capturing (long distances, between sensor unit and sensor PC, no electromagnetic influences)
- Different CCDs
- Different optics
- Resolution: up to 300 dpi
- Lighting type: LED (line, ring, fields)
- Directly applied on labels, ink-jetted, embossed or laser engraved
- Codes: all standard barcodes, 2D-codes (ECC200, PDF417, Maxicode)
- OCR: signs, numbers, alphanumeric writings, address information
- Special characteristics: database verification, voting of different classifiers,
machine and hand writings - Videocoding (manual completion of code, markings or addresses)
- Interfaces: Digital I/Os, RS232, RS422, RS485, CAN-Bus, Profibus, Ethernet (TCP/IP)
- Interactive Graphical User Interface (statistics, control and surveillance)
- Remote control (via ISDN and/or network)
- Markets: Postal market, industrial automation, quality inspection, traffic technology
2. What specific 2D symbols does your reader recognize?
[Dennis] Generally Microscan 2D readers support all of the symbols that have been released to the public domain. These include Data Matrix, PDF-417, RSS, RSS/Composite, and QRCode. Microcan's main emphasis is on Data Matrix applications due to industry activity.
[John] ICR-8XX readers will recognize and decode DataMatrix codes as well as all common linear codes
[Klaus] The VIPAC 2D code reading system identifies ECC200, PDF417 as well as proprietary 2D codes such as MaxiCode. The same software can read all common 1D barcodes.
[Jamie] The DataMan handheld readers and In-Sight fixed mount readers recognize all standard ECC formats for Data Matrix codes including ECC 200. Both handheld and fixed-mount readers can also read QR Code. In-Sight fixed mount systems also support PDF-417 and Composite Symbology (CS) code reading capability.
3. What manufacturing industries are using 2D symbol readers and for what applications?
[John] Automotive, electronics, packaging, pharmaceutical, document handling, and biomedical equipment manufactures are all using 2D to some extent. Automotive and electronics industry applications are common application types that SICK addresses. In the automotive industry, suppliers and manufacturers are using 2D marking and reading methods for parts traceability. They are implementing processes to permanently mark parts and sub-assemblies so that an item can be identified and tracked throughout its life cycle regardless of the environmental conditions it may encounter. By permanently marking a part with a 2D DataMatrix code, which can contain 10 times the data as a linear code in a smaller space, manufacturers can identify parts any time during manufacture for error proofing the assembly process and for enhancing quality control. Similarly, electronics components manufacturers, biomedical equipment manufacturers, and document handling and packaging firms are implementing 2D reading into their processes to ensure product quality, identify and trace defects, and reduce the overall cost of manufacturing.
[Klaus] Postal industry (letter and parcel distributors), automotive industry, chemical industry, medical industry, traffic technology…and many more
[Jamie] The automotive industry has been the broadest adopter of 2D direct part mark code reading and traceability across a wide range of applications including the engine, transmission and drive train. We see increasing general adoption in the Aerospace industry, especially in conjunction with the US Department of Defense initiative to uniquely identify all components with a value of $5000 or greater with a unique Data Matrix code.
[Dennis] While the acceptance of 2D symbols is on the increase, the predominant industries actively using Microscan 2D products include aerospace, automotive, electronics, semiconductor, consumer goods, banking, pharmaceutical/health care, government. Microscan imagers are commonly used in applications such as parts tracking, manufacturing control, quality control, traceability, document control, ‘‘lot’‘ tracking, and personal identification. For example, in the automotive industry Microscan’s Quadrus EZ fixed imager is used to track 2D codes on anti-lock break components and air bag deployment sensors for work in process identification. The Tier 1 supplier selected Data Matrix over other technologies ‘‘because they wanted smart parts, not smart pallets.’‘ In the semiconductor industry, Quadrus EZ is reading a tiny 2-mil 2D code on flat panel displays. In the aerospace industry, a major aero-engine manufacturer is using Microscan’s MS-Q handheld imagers to read direct part marks on items such as turbine blades to comply with industry directives as well as to improve operation efficiency internally.
4. What are the drivers in those industries for adopting 2D symbol readers?
[Klaus] Customers are requesting to accumulate more information in limited space. The answer is printing embossing or laser engraving 2D Codes onto the relevant surfaces. Additionally 2D codes have a better redundancy, which guarantees a high degree of automation.
[Jamie] The automotive industry has seen the primary benefit in cost reduction and process improvement in the manufacturing of their products. They have also seen the benefits of cost containment to readily identify defective components if an error is made.
The Aerospace industry is primarily looking at 2D traceability for process improvement in addition to error containment. The Department of Defense is adopting 2D symbology (and the readers to read the marks) primarily for general accounting, inventory control, product identification and traceability today.
[Dennis] The main motivation for implementing 2D symbols is symbol size, data capacity and accuracy. Data Matrix has low-resolution requirements, which makes it easier to print with a variety of marking methods such as dot peen.
[John] In the automotive and aerospace industry the TREAD Act mandates that manufacturers and suppliers be able to trace critical parts during manufacture to minimize the introduction of defective or potentially harmful parts. Critical component examples include braking systems, passenger restraint, power train components, and fuel systems. In addition to improving overall product quality and minimizing manufacturing-related costs, one by-product of this effort is a reduction of costs associated with broad based product recalls. It also minimizes the exposure manufacturers face during potential litigation that may come as a result of selling defective products. Implementing a 2D system allows a manufacturer to have better control over its processes and increased accountability.
5. Are there some specific issues associated within the respective industry/applications you have cited? Industry standards? Calibration? Mark verification? Etc.?
[Jamie] One of the most significant issues today is the establishment of universal verification standards that can be used across different competitive systems to accurately determine the potential readability of a particular code with the appropriate 2D symbol reader. Many of the 2D standards today were written from the perspective of grading 1D codes (black print/white label), which is often not appropriate for direct part marks with 2D codes.
A 2D code can rate very low in the contrast metric for verification but the same code could be very readable with the appropriate scanner. The reality is that it is impossible to get high contrast on certain materials with the different marking technologies. It is a matter of grey on grey not black on white. It is more important to look at ‘‘cell separability’‘ rather than overall contrast to determine how easy it is for the software to distinguish between an ‘‘on’‘ cell and an ‘‘off’‘ cell.
Certified System Integrator Program
Set Yourself at the Forefront of the Global Vision Market
Vision system integrators certified by A3 are acknowledged globally throughout the industry as an elite group of accomplished, highly skilled and trusted professionals. You’ll be able to leverage your certification to enhance your competitiveness and expand your opportunities.
Today, there is work underway by standards committees to modify the verification standards to address these issues. Once the standards are revised and implemented, it should allow the adoption of 2D code marking and reading to be even more widespread as the standards can be universally implemented.
[Dennis] Many industries and organizations have established standards for product/part marking and identification. For instance, aerospace, automotive, electronics, semiconductors have all set Data Matrix as a standard for marking applications. Data Matrix is verifiable and can be marked using many different technologies.
[John] Various standards have been published that address the proper usage of 2D codes. The standards are somewhat different for each industry. For example, the AIAG, ISO, ATA, DOD and NASA each publish standards for use in their respective industries. The different standards dictate symbology usage and certain parameters regarding coding methods. Verification is one area where there is some debate. There are few methods or devices available for accurately checking the quality of a 2D code against a specific standard. Lighting conditions, read ranges, and other environmental factors influence the readability and overall quality of a code. An alternative to code grading is performance analysis in which the readability of codes over time is evaluated. The better reading devices are able to recognize when code quality is deteriorating and alert the user to the condition. While not actually providing a specific grade to each code, this method does alert the user to a potential problem with their marking process and provides an opportunity to take corrective action.
[Klaus] The VITRONIC OCR/OCV and barcodes readers are designed in accordance with FDA regulations. The company has been certified according to DIN EN ISO 9001:2000 since 1996.
6. What are the specific application issues that one must be attentive to when deploying machine vision-based 2D symbol readers?
[Dennis] How will the symbols be marked, what will the symbol be marked on, what is the environment that the readers will be operating in.
[John] Symbology selection, code quality and readability are key issues. After that, read distance, read area and lighting requirements are critical. At SICK, the first step in working with a customer is to fully understand their application and what they are trying to achieve. We approach each 2D application much the same way we would a linear code application. To select the best possible solution, we must know how the code will be presented to the reader. How far away from the object should the reader be placed? What type of code is being used? Is the object in constant motion or does it start and stop? Improper placement of the reader, incorrect lighting for the type of code being used, and high variability on code quality will result in poor read rate performance.
[Klaus] Width of field, depth of field, code type, code size, application technique, space requirements (distance to object), clock cycle, object moving or standstill, interfaces, environmental conditions, original code samples, speed, top sided or multisided address label search on parcels.
[Jamie] As with any vision application, one must consider the environment in which the reader will be used. In addition, to the functional and application requirements for a particular job, factors such as lighting, product variability and sensor pixel resolution are also key issues.
Creating an optimal mark is a critical first step in a successful application. Knowing what marking method to use (based on material type), mark location, ensuring you have valid quiet zones, ensuring the cell size is large enough are all important considerations that need to be taken into account.
7. What is it that you require from a prospective buyer of 2D symbol readers for an industrial application to assure you will deliver a 2D symbol reader that will satisfy them?
[John] Typically we will request all the application parameters in order to match the correct reader to the application. This normally includes code type, read distance, speed of object, minimum element size, read range (near point-far point of read area), and background characteristics. We also need to know how the data should be transmitted so we request information about the host communication platform and the data format that is required for output from the reader. Sample codes are always a good way to evaluate a potential application, although they are not required.
[Klaus] VITRONIC is requiring information about the minimum module width of the codes, the Transportation speed, and the depth of field in order to adjust the VIPAC systems to the customers’ requirements. The systems will be supplied as turnkey solutions. Add-ons such as 3D-volume measurement, weighing and even OCR can be integrated in these modular VIPAC systems.
[Jamie] As noted above, first and foremost, we need to determine the specific application requirements. Considerations such as field of view, lighting, code type, reading throughput, and unit connection considerations (Ethernet vs. USB) are all things that we require upfront.
[Dennis] We need to know in specific detail the application requirements such as symbol information (element size), line speed and symbol spacing.
8. Are there specific application issues that must be understood insofar as selecting the correct 2D symbol reader for the application is concerned?
[Klaus] VITRONIC supplies camera based 2D code readers only. The VITRONIC decoding software is based on grey scale reading algorithms, which are much more robust against ambient influences. The systems can even read 2D codes under plastics foil.
[Jamie] Unlike traditional 1D readers that today all provide reliable read rates, the performance of 2D readers can vary dramatically depending on the surface and marking method. Companies with a vision background have leveraged their expertise in both lighting and vision algorithms to provide both fixed mount and hand held products that read both direct part marks as well as printed codes extremely well.
Traditional Auto ID company products typically work very well on high contrast, well-printed applications but have not invested in the lighting nor the algorithmic work needed to reliably read all the various direct part marks being adopted by industry.
In addition to selecting the company with the most reliable read rates for their readers, utilizing advanced location algorithms for instance, great consideration must also be given to the lighting that will be used for a particular product. For example, a low contrast dot peen code will often require a low angle light to correctly illuminate the code. A well-printed standard contrasting code can be read easily in many cases with simply a reader with bright field illumination.
[Dennis] What symbology, will the application involve high contrast labels or low contrast direct marks, will the parts be moving, and will the read stations be fixed mount or hand held?
[John] Basic application parameters must be understood to make the best solution recommendation. Object distance, read area, speed, code type and minimum element size are critical. Depending on the requirements, there may be a variety of solution types that will work. For example, the way a code is presented to the reader is important. An object that is in continuous motion may require a different solution than objects that start and stop in front of the reader. Different marking methods may require different lighting configurations. For example, some readers have integrated laser lighting, while others may offer external LED lighting options. The surface type dictates which may be the best lighting solution. Printed paper type labels are less likely to require optional lighting sources because the contrast of the labels is usually very good. Dot peen marks and some laser-etched metals may require external lighting to optimize the illumination in the read zone.
9. How important are the optics, lighting and imager properties in a specific industrial application of 2D symbol readers?
[Jamie] The optics, lighting and imager properties each play an important role in the ability of a 2D symbology reader to properly illuminate and define a 2D object within the field of view. The reader must also be flexible since different marking types (Dot Peen, Laser, etc.) may require different lighting approaches (direct lighting (90 degree) vs. low angle or dark field lighting).
Another consideration is the part surface itself. A highly reflective, curved surface part for instance can benefit from cloudy day illumination to provide even lighting. Milled parts with a grain running in one direction can benefit from quadrant lighting to provide illumination along the axis of the grain.
Ultimately the goal of optimizing the optics, lighting and imager properties is to provide the reader's object location algorithms the best opportunity to find the 2D code, so that it can apply the decoding algorithm.
[Dennis] All three components are critical. Optics will determine the minimum and maximum symbol size. Lighting will determine the read success based on the technology used to make the mark and the surface that is marked. Imager properties will determine overall performance. Due to very low contrast of some direct part marks (such as dot peen on steel or cast parts), artificial lighting is sometimes needed to enhance mark contrast. However, providing an exterior lighting solution for a handheld imager is not very feasible. For addressing these applications, Microscan has developed the patented optical accessory, the LightRay Optics. The LightRay Optics provides handheld imagers with the optimal lighting configuration needed to read very low contrast marks and marks applied to highly reflective surfaces.
[John] Lighting and optics are critical in 2D code reading applications...particularly some metal etched and dot peen codes. The nature of these code types often require concentrated light on the coded object to effectively detect contrast differences that allow a code to be decoded. Better 2D code reading solutions provide a means to adjust the illumination on the coded object.
[Klaus] The VITRONIC 2D code readers are adaptable to different environments. However, the proper selection of camera and illumination is very important in order to reach optimum read rates, which are resulting in a high degree of automation.
10. Are there some emerging technology changes associated with the underlying technology embodied in 2D symbol readers that will positively impact your 2D symbol reader product line? What will those impacts be?
[Dennis] In general I believe that the availability of larger capacity imagers will allow greater flexibility in mark size and quality. Specific planned product enhancements are naturally confidential.
[Klaus] VIPAC systems consist of the cutting-edge technology and are designed for searching and reading of 2D codes at various speeds of up to 580 feet/min. VIPAC systems identify up to six sides of a package, covering a depth of field of up to 55 inches. The next generation will cover even higher speeds.
[Jamie] One of the more important developments embodied in Cognex 2D symbol readers is the increased processing power and lower cost of processors today. This has allowed Cognex to move its powerful PatMax geometric pattern matching object location algorithms from strictly a PC product environment into our current In-Sight fixed mount sensors and DataMan handheld readers.
IDMax, a Data Matrix decoding algorithm is based on the powerful Cognex PatMax geometric pattern matching technology. The benefit of this development is the ability to read and locate 2D codes on an object that would have formerly required the processing power of a PC.
[John] Developments in illumination technology and imager/processing power will enable more performance flexibility on a wider variety of surface types and sizes.
11. Will RFID eventually eliminate the needs for 2D symbol reading?
[John] Not likely. Laser, 2D and RFID will coexist in the supply chain for some time to come. One technology does not fit all applications or provide the same economic return. So the supplier that can best fit the proper technology to an application will provide the greatest value to its customer.
[Klaus] No, not in the near future due to some technical aspects. The prices for the individual labels are still too high to be profitable for our customers.
[Jamie] More and more we see RFID as a complementary and not a competing technology. The press and media have really hyped this technology without understanding where and how it can be applied. Successful RFID applications are emerging in the supply chain where crates and pallets can be more efficiently tracked. Both 1D and 2D bar code marking will be the predominant method to mark and track individual components and parts for the foreseeable future.
[Dennis] RFID is not practical for many applications, especially where space is limited. Case in point is jet engines. While the main assembly is ideal for RFID, each component, such as individual turbine blades, still must be identified with a permanent mark. I believe that 2D and RFID will coexist.
12. Any advice you can give to someone investigating a 2D symbol reading application?
[Jamie] If a company is investigating a 2D symbol reading application, the first step is to realize that reading a 2D code requires an imaging system and not a traditional laser scanner. The goal is to provide laser scanner-like performance. To do that, the code must be properly illuminated, and the 2D reader must use software to locate the code in the field of view. This is especially important in Direct Part Mark applications because the task of locating the 2D symbol in the field of view becomes much more complex when different potentially low contrast marking technologies and different mark surfaces come into play. We would recommend working with a company that has experience working with these types of object location challenges.
[Dennis] Look to industry guidelines and current successful installations within their specific industry. Work with companies that understand 2D and have a track record. A major consideration should be on support, both pre and post sales. Those planning a 2D project are encouraged to take advantage of Microscan’s free