For a PDF copy of this document, please send a request to firstname.lastname@example.org - we will be pleased to send it to you!
Bar Code Verification and ensuring that the quality of printed bar codes meets and exceeds the international standards and retailer requirements can seem complex, however we try to make this complexity as easy to understand as possible. A number of the most asked questions are listed here, however it is often easier just to talk to someone - so please don't hesitate to give us a call or send us an enquiry and we'll be delighted to make the process of ensuring that your bar codes are the best you can expect is easier for you.
If your question is not in this list feel free to Contact us.
Click here for Retail Specifications and other key info: ASDA Guidelines, How bar codes work and "Barcoding - Getting it right"
At Bar Code Systems, we do not believe in selling something to a customer and then leaving them to it - especially in critical Quality Assurance systems and bar code verification - We will supply our products at fair and competitive prices that enable you to count upon our lifetime support promise. Whaen you have bought a bar code verifier from us we will always provide you with support if you have problems with your bar code quality. Send us your verification reports when you have problems that you cannot resolve yourself and we'll be pleased to give you our opinion and suggestions for improvements.
We provide the following:
A scanner reads a bar code using a red laser and it decodes a symbol by measuring the amount of light reflected back into it. As a result of the laser, a scanner cannot "see" the colour red and therefore red cannot be used for the colour of the bars. A
s a general rule, we recommend you use warm colours for the background/spaces and cold colours for the bars. To ensure you are using the correct colours, visit our Services section of this website. You will find our Spectrum Checker software which you can download. All you need are the pantone references for your bars and the background colour. The rest is simple.
The basic principle of scanning is that light is reflected in different measures from dark and light surfaces (bars and spaces of the barcode). The light source of the scanner illuminates the symbol and a light detector converts the level of reflected light into electrical signals which are then decoded.
The electrical signals are then converted into meaningful machine readable data. This is achieved by a microprocessor and software within the scanner.
The data is then sent to a PC or till which then displays the barcode number or uses the number to access a database. This then displays the information relating to the number eg price.
No two bar code readers are the same. The optical arrangements available for scanners vary widely, ranging from light pens or wands to CCD scanners and hand-held or omni-directional laser scanners, and from manually-operated to automatic, unattended devices, any of which might be found at the various points in the distribution chain to the retail store or warehouse through which the product passes. Inevitably, these show quite noticeable differences in their scanning performance. Also, in order to maximise their performance, manufacturers of bar code readers build all kinds of clever features into their decode algorithms to help the equipment decode even poor quality symbols reliably and as rapidly as possible. However, not all of these work in the same way and two different readers might well have different degrees of success with the same symbol.
So test scanning a symbol with, for example, a wand reader will not give any reliable indication of whether it would read with a laser scanner, nor even that any other wand reader could read it successfully. Nor does it help you understand whether the symbol deviates from perfect and if so what is wrong with it. At best, it can be used as a "go/no-go" test of whether a symbol can be read by that scanner (only), and to check the data content; it is risky to extrapolate any wider conclusions. But a verifier bases its assessment on the use of a standardised reference decode algorithm specified as part of the symbology specification, and on calibration of its optical response. Both of these enable consistent and objective quality assessments to be made irrespective of what type of scanner will be used in the application.
The table below summarises the specified aperture sizes for verification of the various symbols used in the EAN/UCC system.
EAN & UPC 5mil 0.15mm
UCC/EAN-128 10mil 0,25mm
ITF 14 (X<0,635mm) 10mil 0.25mm
ITF 14 (X =>0.635mm) 20mil 0,50mm
The scan reflectance profile is analysed in terms of a series of parameters. A number of these are reflectance-related, such as Maximum and Minimum Reflectance, Symbol Contrast, Defects, Edge Contrast and Modulation, while some bear more relationship to dimensional features (Decodability and Bar Width Gain/Loss). These parameters have been shown both to be measurable from the profile and to have an effect on scanning performance.
First, the profile is divided into candidate bar and space regions by setting a Global Threshold midway between the highest light reflectance (Rmax) and the lowest dark reflectance (Rmin) measured anywhere in the scan; areas above the Global Threshold are treated as spaces and those below it as bars. For the detailed parameter analysis, each element edge position is determined, not where the profile crosses the Global Threshold, but where it passes through the reflectance value midway between the peak reflectance of the adjoining space and the lowest reflectance of the adjoining bar. In Figure 1 below, the peak reflectance of the wide space is 74%; the lowest reflectance of the adjoining bar is 10%, giving an Edge Contrast of 64%, half of which is 32%, so the edge between the two is located at reflectance value 10 + 32 = 42%. Similarly, the Edge Contrast between the bar and the space to its right is 52%, so the edge in this case is located at reflectance value 10 + 26 = 32%. The Global Threshold is shown by the solid line at reflectance 40%.
Figure 1 - Illustration of edge location in a profile
The following scan reflectance profile shows an example of the parameters measured.
Figure 2 - Scan reflectance profile with key measurements indicated
The key values shown in figure 2 above have great significance in the evaluation of the profile. In this case, Rmax (80%) is found in the left quiet zone and Rmin (6%) is located in bar 7 (counting from the left), enabling the Global Threshold to be set at 43%. The minimum Edge Contrast value is found on the trailing edge of bar 9 (67% - 12% = 55%). The most significant defect is located in bar 12, with a depth of 9%. The positions of individual element edges are not shown on the above profile, although some verifiers do show these positions, and highlight the critical parameters in various graphical ways.
It is an analogue plot of the light reflected from the symbol as a scanning spot or sampling aperture moves across it. The x-axis of the plot shows linear distance across the symbol, while the y-axis shows the reflectance values. Light areas, of course, show high reflectance values; dark areas show low values. The profile therefore consists of a series of peaks and valleys, the widths of which are proportional to those of the bars and spaces. There is not an instantaneous transition from the low to high reflectance values or vice versa, but the transition slopes steeply, since when the aperture is crossing a bar edge, its area covers both light and dark areas in proportions which change progressively as the spot moves over the edge.
Figure 3 - Scan reflectance profile of EAN 13 symbol
The scan reflectance profile is a picture of the raw material which a scanner uses to derive a digital reconstruction of the bar and space pattern forming the symbol, which the decoder then interprets back into its original data values.
For this reason, verification based on analysis of a scan reflectance profile is able to show very close correlation with scanning performance.
The standards set out a reference optical arrangement, consisting of a source of flood incident light at 45° to the surface and a collector (through an aperture) of the diffusely reflected component of this light, at right angles to the surface. For consistent reliable measurements this distance must be constant. The vertical plane in which the light source is located is parallel with the height of the bars. This set-up is intended to minimise the effect of specular (mirror-like) reflection from glossy surfaces.
The primary advantage of the scan reflectance profile assessment over the traditional element width/PCS measurement is that it provides a far better indication of how well a symbol is likely to perform when read under typical application conditions, at the point of sale, or in back of store and distribution operations. But where it falls down is that it is difficult to deduce from the scan reflectance profile grading what specific corrective action needs to be taken to improve quality grades, in terms of aspects that the symbol producer can easily control. Scan reflectance profile grading on its own is of little help for process control purposes, one of the two reasons for which symbols are verified.
So direct measurement of bar width gain or loss as measured in the traditional verification method, is the most useful process control tool since it provides the symbol producer with an easily understandable and quantifiable measurement.
Each parameter value is graded: some on a pass/fail basis, others on a five-step scale. The table below shows the values of each parameter corresponding to the thresholds between grades. Each scan reflectance profile is then given a profile grade, which is the lowest of any of the individual grades for the parameters in that profile (on the principle of the weakest link in a chain).
Grade thresholds for scan reflectance parameters
The first step in analysing the profile, after identifying the bar and space regions in the profile and determining the position of each element edge, is to apply the reference decode algorithm - the set of rules/steps for decoding a symbol defined in the symbology specification - to the elements "seen" in the scan reflectance profile. If a valid decode results, the decode parameter passes and is given grade 4, otherwise it fails (grade 0). If the wrong number of elements is seen, the decode clearly fails. Note that in the ANSI standards this last case is graded separately as an "edge determination" failure, although the final effect on the profile grade is the same
The Symbol Contrast is the difference between the highest and the lowest reflectance values in the profile. The maximum reflectance (Rmax) may occur anywhere, in a space or a quiet zone. The minimum value (Rmin) will always be in a bar. The importance of this parameter is that the higher the Symbol Contrast, the more easily distinguishable from each other the bars and spaces will be. SC of 70% or higher is graded 4, while SC below 20% is grade 0.
Minimum reflectance (Rmin) Rmin must always be no higher than half of Rmax. This is because, for a given level of Symbol Contrast, many scanners have greater difficulty distinguishing relatively light bars against a high-reflectance background than they do darker bars against a relatively low reflectance background. This will tend only to affect symbols with grade 2 or 1 Symbol Contrast, where the value of Rmax is in the upper part of its range. It is a pass/fail parameter: it is assigned grades 4 or 0. The symbol shown in Figure 4 below, printed in light brown on a white background (which appears to give good visual contrast) yielded a scan reflectance profile (Figure 5) which failed on this criterion. Rmax was 83%, so that Rmin should have been 41.5% or less; the actual Rmin was 43%.
Figure 4 - Symbol with failing minimum reflectance
Figure 5 - Scan reflectance profile showing failure to meet Rmin criterion
This is a measure of local contrast between two adjacent elements. It is defined as the difference of the highest and lowest reflectance values (Rs and Rb respectively) in a pair of adjacent elements (bar + space or space + bar). Quiet zones are considered spaces for this purpose. Every element has its own value of Rs or Rb. The lowest Edge Contrast (ECmin) in a profile determines the grade for the parameter; this is also a pass/fail graded parameter (grades 4 or 0 only) where, if ECmin is less than 15%, it is graded 0. Variations in ink weight in different parts of a symbol, or fluctuations in the background reflectance (e.g. with corrugated brown kraft substrates) are one cause of Edge Contrast problems, but another is that scanners tend to see narrow elements less distinctly than they do wider ones (a narrow space has lower apparent reflectance than a wide one, and a narrow bar appears similarly lighter than a wide one).
Figure 6 - Scan reflectance profile showing edge with lowest Edge Contrast 55% (67%-12%)
This parameter is related to the previous one, and is a measure of Edge Contrast as a proportion of Symbol Contrast. Low Modulation values will be caused by the same factors as low Edge Contrast. The difference is that Modulation relates Edge Contrast to Symbol Contrast (MOD = ECmin/SC). A low Edge Contrast value carries a greater risk of causing poor reading results when Symbol Contrast is high than the same Edge Contrast value has when Symbol Contrast is low. In the profile in Figure 10 above, the minimum Edge Contrast value of 55%, when divided by the Symbol Contrast of 74%, gives a MOD value of 0.74, which would be graded 4.
Spots of ink in the quiet zones or spaces, or light areas in the bars, will cause a ripple in the scan reflectance profile at the point where the scan path crosses them. This is referred to in the profile analysis as Element Reflectance Non-uniformity (ERN). In the profile of a space, they show as a valley; in that of a bar, they show as a peak. If this peak or valley approaches the threshold between light and dark, the risk of the element being seen as more than one, and of the scan failing to decode, increases. As already indicated, the use of the correct measuring aperture ensures that the effect of defects is not exaggerated or underrated. The defect parameter measures the relationship of the depth of the highest peak or deepest valley to Symbol Contrast (Defects = ERNmax/SC), indicative of its relative severity.
Figure 7 - Scan reflectance profile of a bar with a Grade 1 defect
Although a fairly easy concept in principle, Decodability is harder to explain in a few words. It measures how close the dimensions within the symbol are to their ideal values, and as such is a measure of its dimensional accuracy. However, it only applies to those measurements needed by the decode algorithm to determine the width of an element, or the combined widths of elements, in order to decode the symbol character.
Taking a simple case, if it is a question of determining a particular width, the decode algorithm might say something like "If the measurement is between 2.5 modules and 3.5 modules, treat it as equal to three modules." In other words, 2.5 modules is the threshold value between two modules and three modules, and 3.5 modules is the upper threshold between three and four modules; there is a 0.5 module margin on either side of the nominal measurement of 3.0. Decodability measures how much of this margin is left in the worst (most deviating) measurement: assume the measurement is 2.7 modules, then only 0.2 is left between the actual measurement and the 2.5 threshold, out of the total margin of 0.5, so the Decodability value is 0.2/0.5, which is 40%.
The lower the Decodability, the harder it will be for a decoder to decode the symbol. Some symbologies have particular features which require a more complex Decodability calculation, and the relevant ones are briefly discussed in the sections on EAN/UPC and UCC/EAN-128. Generally speaking, Decodability values should not be used for process control purposes, since they only refer to a single measurement; the measurement of average bar width deviation is much more reliable.
Decodability is not helpful for print process control. The reason is that Decodability does not depend on linear or metric deviations so the results may be misleading for print process control. The average bar width gain or loss may also be misleading if used alone. Firstly the extreme bar deviations should be checked to see if there are irregular or regular bar deviations. In case of regular bar deviations you use the average value to correct bar width adjusment. In case of irregular bar deviations the user needs to check if there is a problem in prepress (roundings due to zooming, resolution changes or wrong bar code sizes in relation to resolution) or if the problem is caused by machine adjustment.