标准编号:	GB/T 23704-2017
中文名称:	二维条码符号印制质量的检验
英文名称:	Two-dimensional bar code symbol print quality test
中标分类:	L71    编码、字符集、字符识别
行业分类:	GB    国家标准
ICS分类:	35.040 35.040    字符集和信息编码 35.040
发布机构:	中华人民共和国国家质量监督检验检疫总局 中国国家标准化管理委员会
发布日期:	2017-12-29
实施日期:	2018-7-1
标准状态:	现行
替代旧标准:	GB/T 23704-2009 信息技术 自动识别与数据采集技术 二维条码符号印制质量的检验
翻译语言:	英文
文件格式:	PDF
中文字符数:	45000 字
翻译价格(元):	2700.00 元
交付时间:	付款后 1 个工作日

Codeofchina.com is in charge of this English translation. In case of any doubt about the English translation, the Chinese original shall be considered authoritative.



This standard is developed in accordance with the rules given in GB/T 1.1-2009.



This standard replaces GB/T 23704-2009 Information technology - Automatic identification and data capture techniques - Bar code print quality test specification - Two-dimensional symbols, and the following main changes have been made with respect to GB/T 23704-2009:



——The standard name is changed to Two-dimensional bar code symbol print quality test;



——ISO/IEC 19762, GB/T 18284, GB/T 21049, GB/T 35402, ISO/IEC 16022, ISO/IEC 16023 and ISO/IEC 24778 are added (see Clause 2);



——The term and definition of "reflectance margin" is added (see 3.12);



——The symbol grade is obtained through one scanning measurement, and does not requires five scans (see 7.4);



——The definitions of "contrast uniformity" and "reflectance margin" parameters are added, and the concepts of "module MOD", "codeword MOD", "symbol MOD", "module RM", "codeword RM" and "symbol RM" are added (see 7.8.4);



——The requirement "the interim codeword grade of each parameter for each codeword is the highest codeword grade for that parameter obtained for all scans of that codeword." is modified to "the interim codeword grade of each parameter for each codeword is the highest codeword grade for that parameter obtained on any scan for that codeword.” (see 6.2.5);



——The proprietary parameters of each symbology for symbol classification related to Data Matrix code and QR Code are not detailed, and only the symbology standards containing these parameters are given (see Annex D);



——The quality parameters of "reflectance margin", "format information" and "version information" are added (see Annex G).



This standard has been redrafted and modified in relation to ISO/IEC 15415:2011 Information technology - Automatic identification and data capture techniques - Bar code symbol print quality test specification - Two-dimensional symbols.



The main technical deviations with respect to ISO/IEC 15415:2011 and the reasons are as follows:



——The adjustments of technical deviations are made for the normative references in this standard so as to adapt to the technical conditions of China. The adjustments are mainly reflected in Clause 2 "Normative references", with the specific adjustments as follows:



 ISO/IEC 15416 is replaced by GB/T 14258 which is modified in relation to the international standard;



 ISO 7724-2:1984 is replaced by GB/T 11186.2 which is modified in relation to the international standard;



 ISO/IEC 19762-1 and ISO/IEC 19762-2 are replaced by the newly revised ISO/IEC 19762;



 GB/T 2828.1, GB/T 6378.1, GB/T 12905, GB/T 18284, GB/T 21049, GB/T 35402, ISO/IEC 16022, ISO/IEC 16023 and ISO/IEC 24778 are added (see Clause 2).



——The measurement process of two-dimensional matrix bar code is described by means of items (see 7.1);



——The representation of the parameter range corresponding to each grade in the grading tables is modified, for example: the parameter range corresponding to grade 3 in Table 2 is changed to 0.64≤CY<0.71, and it is inaccurately represented by ≥64% in Table 2 of ISO/IEC 15415:2011 (see Table 2, Table 3, Figure 2, Table 5, Table 6, Table 8, Table 9, Table 10, Table 11).



——The required "the interim codeword grade of each parameter (Modulation, Defects and Decodability) for each codeword is the highest codeword grade for that parameter obtained on any scan for that codeword" is modified to "the interim codeword grade of each parameter for each codeword is the highest codeword grade for that parameter obtained on any scan for that codeword" (see 6.2.5);



——The concepts of "module MOD", "codeword MOD", "symbol MOD", "module RM", "codeword RM" and "symbol RM" are added (see 7.8.4).



 

For the purposes of this standard, the following editorial changes and structural adjustments have also been made:



——The order of annexes in ISO/IEC 15415:2011 is adjusted herein so that Annex A hereof corresponds to Annex D of ISO/IEC 15415:2011, Annex B hereof corresponds to Annex F of ISO/IEC 15415:2011, Annex C hereof corresponds to Annex B of ISO/IEC 15415:2011, Annex D hereof corresponds to Annex A of ISO/IEC 15415:2011, Annex E hereof corresponds to Annex C of ISO/IEC 15415:2011, and Annex F hereof corresponds to Annex E of ISO/IEC 15415:2011.



——Annex G (Informative) "Examples of two-dimensional bar code symbol test report" is added.



This standard was proposed by and is under the jurisdiction of SAC/TC267 National Technical Committee on Logistics Information Management of Standardization Administration of China.



 

Introduction



The technology of bar coding is based on the recognition of patterns encoded, in bars and spaces or in a matrix of modules of defined dimensions, according to rules defining the translation of characters into such patterns, known as the symbology specification.



Bar codes may be categorised into linear bar codes, on the one hand, and two-dimensional bar codes on the other; the latter may in turn be sub-divided into "two-dimensional multi-row bar codes", sometimes referred to as "two-dimensional stacked bar codes", and "two-dimensional matrix bar codes". In addition, there is a hybrid group of symbologies known as "composite symbologies"; these symbols consist of two components carrying a single message or related data, one of which is usually a linear symbol and the other a two-dimensional symbol positioned in a defined relationship with the linear symbol.



Two-dimensional multi-row bar code symbols are constructed graphically as a series of rows of symbol characters, representing data and overhead components, placed in a defined vertical arrangement to form a (normally) rectangular symbol, which contains a single data message. Each symbol character has the characteristics of a linear bar code symbol character and each row has those of a linear bar code symbol; each row, therefore, may be read by linear symbol scanning techniques, but the data from all the rows in the symbol must be read before the message can be transferred to the application software.



Two-dimensional matrix bar code symbols are normally rectangular arrangements of dark and light modules, the centres of which are placed at the intersections of a grid; the coordinates of each module need to be known in order to determine its significance, and the symbol must therefore be analysed two-dimensionally before it can be decoded. Dot codes are a subset of two-dimensional matrix bar codes in which the individual modules do not directly touch their neighbours but are separated from them by a clear space.



Unless the context requires otherwise, the term “symbol” in this standard may refer to either type of two-dimensional bar code symbology.



The bar code symbol must be produced in such a way as to be reliably decoded at the point of use, if it is to fulfill its basic objective as a machine-readable data carrier.



Manufacturers of bar code equipment and the producers and users of bar code symbols therefore require publicly available standard test specifications for the objective assessment of the quality of bar code symbols, to which they can refer when developing equipment and application standards or determining the quality of the symbols. This standard forms the basis of the process control and quality assessment during bar code equipment manufacturing, bar code symbol production and use.



The performance of measuring equipment for the verification of bar code symbols may be in accordance with GB/T 26228.1 and ISO/IEC 15426-2.



This standard is intended to achieve comparable quality assessment results to the linear bar code symbol print quality standard GB/T 14258, the general principles of which it has followed. It shall be read in conjunction with the symbology specification applicable to the bar code symbol being tested, which provides symbology-specific detail necessary for its application. Two-dimensional multi-row bar code symbols are verified according to the GB/T 14258 methodology, with the modifications described in Clause 6; different parameters and methodologies are applicable to two-dimensional matrix bar code symbols.



There are currently many methods of assessing bar code quality at different stages of symbol production. The methodologies described in this standard are not intended as a replacement for any current process control methods. They provide symbol producers and their trading partners with universally standardized means for communicating about the quality of two-dimensional bar code symbols after they have been printed. The procedures described in this standard shall necessarily be augmented by the reference decode algorithm and other measurement details within the applicable symbology specification, and they may also be altered or overridden as appropriate by governing symbology or application specifications.



Alternative methods of quality assessment may be agreed between parties or as part of an application specification.





Two-dimensional bar code symbol print quality test



1 Scope



This standard specifies methods for testing, grading and overall quality assessment of two-dimensional multi-row and matrix bar code symbols, and gives information on possible causes of deviation from optimum grades and appropriate corrective action.



This standard is applicable to the print quality test of those two-dimensional bar code symbols for which a reference decode algorithm has been defined in the two-dimensional bar code symbology specifications, but its methodologies can be applied partially or wholly to the test of two-dimensional bar code symbols of other similar symbologies.



2 Normative references



The following referenced documents are indispensable for the application of this document. For dated reference, only the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies.



GB/T 2828.1 Sampling procedures for inspection by attributes - Part 1: Sampling schemes indexed by acceptance quality limit (AQL) for lot-by-lot inspection (GB/T 2828.1-2012, ISO 2859-1:1999, IDT)

GB/T 6378.1 Sampling procedures for inspection by variables - Part 1: Specification for single sampling plans indexed by acceptance quality limit (AQL) for lot-by-lot inspection for a single quality characteristic and a single AQL (GB/T 6378.1-2008, ISO 3951-1:2005, IDT)

GB/T 11186.2 Methods for measuring the colour of paint films - Part 2: Colour measurement (GB/T 11186.2-1989, idt ISO 7724-2:1984)

GB/T 12905 Bar code terminology

GB/T 14258 Information technology - Automatic identification and data capture techniques--Verification of print quality of bar code symbols (GB/T 14258-2003, ISO/IEC 15416:2000, MOD)

GB/T 18284 QR Code (GB/T 18284-2000, neq ISO/IEC 18004:2000)

GB/T 21049 Chinese-sensible code

GB/T 35402 Direct part mark (DPM) two dimensional bar code symbol quality test (GB/T 35402-2017, ISO/IEC TR 29158:2011, MOD)

ISO/IEC 16022 Information technology - Automatic identification and data capture techniques - Data Matrix bar code symbology specification

ISO/IEC 16023 Information technology - Automatic identification and data capture techniques - Bar code symbology specification - MaxiCode

ISO/IEC 19762 Information technology - Automatic identification and data capture (AIDC) techniques - Harmonized vocabulary

ISO/IEC 24778 Information technology - Automatic identification and data capture techniques - Aztec Code bar code symbology specification



3 Terms and definitions



For the purposes of this document, the terms and definitions given in GB/T 12905, GB/T 14258 and ISO/IEC 19762 and the following apply.



3.1

pixel

individual light-sensitive element in an array of image capture device [e.g. CCD (charge coupled device) or CMOS (complementary metal oxide semiconductor) device]



3.2

effective resolution

resolution obtained by the measuring device on the surface of the symbol under test, normally expressed in pixels per millimetre or pixels per inch, and calculated as the resolution of the image capture element multiplied by the magnification of the optical system of the measuring device



3.3

error correction capacity

number of codewords in a two-dimensional bar code symbol (or error correction block) assigned for erasure and error correction, minus the number of codewords reserved for error detection



3.4

inspection area

rectangular area which contains the entire symbol to be tested inclusive of its quiet zones



3.5

grade threshold

boundary value separating two grade levels, the value itself being taken as the lower limit of the upper grade



3.6

module error

module of which the apparent dark or light state in the binarised image is inverted from its intended state



 

3.7

raw image

plot of the reflectance values in X and Y coordinates, representing the actual reflectance values from each pixel of the light-sensitive array



3.8

reference grey-scale image

plot in X and Y coordinates, obtained by convolving the raw image with a synthesised circular aperture



3.9

binarised image

binary (black/white) image created by applying the Global Threshold to the reference grey-scale image



3.10

sample area

area of an image contained within a circle 0.8X in diameter, X being the average module width determined by the application of the reference decode algorithm for the symbology in question or, where the application permits a range of X dimensions, the minimum module width permitted by the application specification



3.11

scan grade

result of the assessment of a single scan of a two-dimensional matrix bar code symbol, derived by taking the lowest grade achieved for any measured parameter of the reference grey-scale and binarised images



3.12

reflectance margin

measurement of modulation using error correction and known module colours



4 Symbols and abbreviated terms



For the purpose of this document, the following symbols and abbreviated terms apply.



AN: Axial Nonuniformity



DPM: Direct Part Marking



Ecap: error correction capacity.



e: number of erasures.



FPD: Fixed Pattern Damage



GN: Grid Nonuniformity



GT: Global Threshold



MOD: modulation.



MARGIN: reflectance margin of a module.



RM: Reflectance Margin



Rmax: the highest reflectance in any element or quiet zone in a scan reflectance profile, or the highest reflectance of any sample area in a two-dimensional matrix bar code symbol.



Rmin: the lowest reflectance in any element in a scan reflectance profile, or the lowest reflectance of any sample area in a two-dimensional matrix bar code symbol.



SC: Symbol Contrast. (SC=Rmax-Rmin)



t: number of errors.



UEC: Unused Error Correction



5 Quality grading



5.1 General



The measurement of two-dimensional bar code symbols is designed to yield a quality grade which can be used for symbol quality judgment and process control purposes, and which is broadly predictive of the read performance to be expected of the symbol in various environments.



As a consequence of the use of different types of reading equipment under differing conditions in actual applications, the levels of quality required of two-dimensional bar code symbols to ensure an acceptable level of performance will differ. The required symbol grade shall be determined in the symbol grade form specified in this standard by referring to Annex A, A.4.



Samples shall be taken from the tested sample batch according to the statistically valid number of samples, and the minimum acceptable symbol grade shall be determined. If no sampling plan is specified in the quality control process or the agreement between both parties, an appropriate sampling plan may be selected according to GB/T 2828.1 or GB/T 6378.1.



5.2 Expression of quality grades



Each parameter may have an alphabetic or numeric grade. The numeric grades express different quality grades on a descending scale from 4 to 0. The grade 4 represents the highest quality, while the grade 0 represents failure. The alphabetic grades are expressed on an alphabetic scale from A to F, with a failing grade of F.



Table 1 maps the alphabetic and numeric grades to each other.



Table 1 Equivalence of numeric and alphabetic quality grades

Numeric grade Alphabetic grade

4 A

3 B

2 C

1 D

0 F



5.3 Overall symbol grade values



The overall symbol grade shall be calculated as defined in 6.2.6 or 7.10. Overall symbol grades shall be expressed to one decimal place on a numeric scale ranging in descending order of quality from 4.0 to 0.0.



Where a specification defines overall symbol grades in alphabetic terms, the relative mapping of the alphabetic and numeric grades is as illustrated in Figure 1 below. For example, the range of 1.5 to immediately below 2.5 corresponds to grade C.







Figure 1 Mapping of alphabetic and numeric overall symbol grades



5.4 Reporting of symbol grade



A symbol grade is only meaningful if it is reported in conjunction with the illumination and aperture used. It shall be shown in the format: grade/aperture/measuring light wavelength/angle, where:



——"grade" is the overall symbol grade as defined in 5.3.



——"aperture" is the aperture reference number (from GB/T 14258 for linear scanning techniques, or from 7.3.3 for two-dimensional matrix bar codes).



——"measuring light wavelength" is a numeric value indicates the peak light wavelength in nanometres (for narrow band illumination); the alphabetic character W indicates that the symbol has been measured with broadband illumination ("white light") the spectral response characteristics of which must imperatively be defined or have their source specification clearly referenced.



——"angle" defines the angle of incidence of the illumination. Its absence indicates that the angle of incidence is 45°. It shall be included in the reporting of the overall symbol grade when the angle of incidence is other than 45°.



Note: While illumination from four sides with an angle of incidence of 45° is the default, other angles of incidence, i.e., 30°and 90°, may be specified. In DPM test, the angle of illumination needs to use a combination of numeric and alphabetic angle indicators, see GB/T 35402.



An asterisk following the value for "grade", in the case of a two-dimensional matrix bar code symbol, indicates that the surroundings of the symbol contain extremes of reflectance that may interfere with reading, see 7.7.



Example 1:



3.0/05/660 would indicate that the average of the grades was 3.0 when it is obtained with the use of a 0.125mm aperture (ref. no. 05) and a 660nm light source, incident at 45°.



Example 2:



3.0/10/W/30 would indicate the grade of a symbol intended to be read in broadband light, measured with light incident at 30° and using a 0.250mm aperture (ref. no. 10), but would need to be accompanied either by a reference to the application specification defining the reference spectral characteristics used for measurement or a definition of the spectral characteristics themselves.



Example 3:



3.0*/10/670 would indicate the grade of a symbol measured using a 0.250mm aperture (ref. no. 10), and a 670nm light source, and indicates the presence of a potentially interfering extreme reflectance value in the surroundings of the symbol.



 

6 Measurement methodology for two-dimensional multi-row bar code symbols



6.1 General



The evaluation of two-dimensional multi-row bar code symbols shall be based on the application of the methodology of GB/T 14258, as described in 6.2.2 or 6.3, and if appropriate for the symbology, on the application of the additional provisions described in 6.2.3, 6.2.4 and 6.2.5, to derive an overall symbol grade. Ambient light levels shall be controlled in order not to have any influence on the measurement results. The symbol shall be scanned using the light wavelength(s) and effective aperture size specified in the appropriate application standard. When performing a measurement, the scan lines shall be made perpendicular to the height of the bars in the start and stop characters and shall as far as possible pass through the centres of rows in order to minimise the effect of cross-talk from adjacent rows. In the case of area imaging techniques, a number of scan lines, perpendicular to the height of the bars and sufficient to cover all rows of the symbol, shall be synthesised by convolving the raw image with the appropriate synthetic aperture.



6.2 Symbologies with cross-row scanning ability



6.2.1 Basis of grading



The distinguishing feature of these symbologies is their ability to be read with scan lines that cross row boundaries. Symbologies of this type also share the feature that the start and stop patterns (or equivalent features of the symbol, e.g. the Row Address Patterns of MicroPDF417) are constant from row to row, or the position of only one edge in these patterns varies by no more than 1X in adjacent rows of the symbol. These symbologies shall be graded in respect of:



——Analysis of the scan reflectance profile (see GB/T 14258 and 6.2.2);



——Codeword Yield (see 6.2.3);



——Unused Error Correction (see 6.2.4);



——Codeword print quality (see 6.2.5).



6.2.2 Grade based on analysis of scan reflectance profile



The start and stop or equivalent (e.g. Row Address) patterns of the symbol shall be evaluated according to GB/T 14258. Regions with data content will be evaluated separately as described in 6.2.3, 6.2.4 and 6.2.5. Test scans of the start and stop patterns shall be graded using all parameters specified in GB/T 14258. The effective aperture size is specified in the appropriate application standard or is the default aperture size appropriate for the symbol X dimension given in GB/T 14258.



For the analysis of the scan reflectance profiles, the number of scans shall be 10, or the quotient (integer part) obtained by dividing the height of the symbol by the measuring aperture, whichever is smaller. Scans shall be approximately evenly spaced over the height of the symbol. For example, in a twenty-row symbol the ten scans shall be performed in alternate rows. In a two-row symbol, up to five scans might be performed in each row, at different positions in the height of the bars. The symbology specification may give more specific guidance on the selection of the scans to be used.



To identify bars and spaces, a Global Threshold for each scan has to be determined. Global Threshold shall be equal in reflectance to 1/2 of the sum of the highest and the lowest reflectances in the scan. All regions above the Global Threshold shall be considered spaces (or quiet zones) and all regions below shall be considered bars.



Edge locations shall be determined as the points where the reflectance value is midway between the highest reflectance in the adjoining space (including quiet zones) and the lowest reflectance in the adjoining bar in the scan reflectance profile.



For the evaluation of the parameters "reference decode" and "decodability", the reference decode algorithm for the symbology shall be applied.



Each scan shall be graded as the lowest grade for any individual parameter in that scan. The grade based on scan reflectance profiles shall be the arithmetic mean of the grades for the individual scans.



The measurement of bar width average gain or loss may be used for process control purposes. Note that this method will not be sensitive to printing variations parallel to the height of the start and stop characters. If a full analysis of the printing process is desired, symbols should be printed and tested in both orientations.



6.2.3 Grade based on Codeword Yield



The Codeword Yield (CY) measures the efficiency with which linear scans can recover data from a two-dimensional multi-row symbol. The Codeword Yield is the number of validly decoded codewords expressed as a percentage of the maximum number of codewords that could have been decoded (after adjusting for tilt). A poor Codeword Yield, for a symbol whose other measurements are good, may indicate a Y-axis print quality problem (such as those shown in Annex E, Table E.1).



Obtain a matrix of the correct symbol character values, such as would result from successful completion of the UEC calculations (see 6.2.4). This matrix is used as the "final decode of the symbol" used in subsequent steps to determine validly decoded codewords.



 

An individual scan qualifies for inclusion in the Codeword Yield calculation if it meets either of two conditions:



——The scan did not include recognised portions of either the top or the bottom row of the symbol. At least one of the Start or Stop (or Row Address) patterns shall have been successfully decoded from that scan, together with at least one additional codeword or the corresponding second Start or Stop pattern, or Row Address Pattern.



——The scan included recognised portions of either the top or the bottom row of the symbol. Both the Start and Stop patterns of the symbol shall have been successfully decoded from that scan.



It is important to note that an extension to the symbology’s Reference Decode Algorithm is required, in order to detect and decode a pair of Start and Stop patterns when neither of the adjacent codewords is decodable. As examples, a linear search for a matching pair of PDF417 Start and Stop patterns, or a linear search for a matching pair of MicroPDF417 Row Indicator Patterns, would fulfill this requirement for scans where the Reference Decode Algorithm alone did not decode both patterns; thus this extension can qualify a scan where no codewords (other than the matched end patterns) were decoded. Note however, that a scan that contains only a single decoded Start or Stop pattern found by this linear search does not count as a qualified scan, if no other codewords or corresponding second Start or Stop pattern, or Codeword or Row Address Pattern, were also decoded.



Decode the symbol completely and populate the symbol matrix.



For each qualified scan, compare the codewords actually decoded with the codewords in the symbol matrix and count the number of codewords that match. Accumulate the total number of validly decoded codewords, and update a count of the number of times each codeword of the symbol has been decoded and a count of the number of times each row has been detected. Also record a count of the number of detected row crossings in each scan (a crossing is “detected” when a scan line yields correctly-decoded codewords from adjacent rows).



After processing each scan, calculate the maximum number of codewords that could have been decoded thus far, as the number of qualified scans multiplied by the number of columns in the symbol (excluding the fixed patterns, such as the Start and Stop patterns of PDF417 or the Row Address Indicators of MicroPDF417).



The entire symbol shall be scanned multiple times until three conditions are met:



a) the maximum number of codewords that could have been decoded is at least ten times the number of codewords in the symbol;



b) the highest and lowest decodable rows (which may not necessarily be the first and last rows) of the symbol have each been scanned at least three times; and



c) at least (0.9n) of the codewords (data or error correction) have been successfully decoded two or more times, where n is the number of non-error-correction data codewords in the symbol.



Example:



Taking a PDF417 symbol with 6 rows and 16 columns and error correction level 4, the total number of codewords is 96, of which 64 are data and 32 error correction. To fulfill condition 1, the maximum number of codewords that could have been decoded must be at least 960. To fulfill condition 3, since n is 64, at least 58 of the codewords must have been decoded twice or more (0.9×64=57.6).



If the ratio of the total number of validly decoded codewords to the total number of detected row crossings is less than 10:1, then discard the measurements just obtained, and repeat the measurement process, adjusting the tilt angle of the scan line to reduce the number of row crossings. Otherwise, to compensate for any residual tilt, subtract the number of detected row crossings from the calculated maximum number of codewords that could have been decoded.



Codeword Yield shall be graded as shown in Table 2.



Table 2 Grading of Codeword Yield

Codeword Yield (CY) Grade

CY≥0.71 4

0.64≤CY<0.71 3

0.57≤CY<0.64 2

0.50≤CY<0.57 1

CY<0.50 0



6.2.4 Grade based on unused error correction



Decode the symbol completely and process scans until the number of decoded codewords stabilises. Calculate the unused error correction (UEC) using the following formula.



UEC=1-(e+2t)/Ecap



where,



e——the number of erasures;



t——the number of errors;



Ecap——the error correction capacity of the symbol.



If no error correction has been applied to the symbol, and if the symbol decodes, UEC=1. If (e+2t) is greater than Ecap, UEC=0. In symbols with more than one error correction block, UEC shall be calculated for each block independently and the lowest value shall be used for grading purposes.



Unused Error Correction shall be graded as shown in Table 3.



Table 3 Grading of Unused Error Correction

Unused Error Correction (UEC) Grade

UEC≥0.62 4

0.50≤UEC<0.62 3

0.37≤UEC<0.50 2

0.25≤UEC<0.37 1

UEC<0.25 0



6.2.5 Grade based on codeword print quality



This subclause gives an approach for the assessment of the Decodability, Defects and Modulation parameters. This approach is based on the parameter grading of scan reflectance profile in GB/T 14258, and at the same time takes into account the effect of error correction in symbol quality parameters such as Decodability, Defects and Modulation by error correction. See Annex B for the correction method.



This approach uses the following procedure for the assessment of each of the three parameters. In symbols with more than one error correction block, it shall be applied to each block independently and the lowest value shall be used for grading purposes.



The entire symbol shall be scanned until 0.9n codewords (where n has the same meaning as in 6.2.3) have been decoded more than ten times or until it is certain that each codeword has been scanned at least once without inter-row interference. In each scan, the Decodability, Defects and Modulation parameters shall be measured in each symbol character in accordance with GB/T 14258. The calculation of all three parameters shall be based on the value of Symbol Contrast obtained from Rmax and Rmin in that scan reflectance profile. The interim codeword grade of each parameter for each codeword is the highest codeword grade for that parameter obtained on any scan for that codeword.

Foreword II
Introduction V
1 Scope
2 Normative references
3 Terms and definitions
4 Symbols and abbreviated terms
5 Quality grading
5.1 General
5.2 Expression of quality grades
5.3 Overall symbol grade values
5.4 Reporting of symbol grade
6 Measurement methodology for two-dimensional multi-row bar code symbols
6.1 General
6.2 Symbologies with cross-row scanning ability
6.3 Symbologies requiring row-by-row scanning
7 Measurement methodology for two-dimensional matrix bar code
7.1 General
7.2 Obtaining the test images
7.3 Reference reflectivity measurements
7.4 Requirements of scans
7.5 Scan grading
7.6 Grading procedure
7.7 Additional reflectance check over extended area
7.8 Image assessment parameters and grading
7.9 Scan grading
7.10 Overall symbol grade
7.11 Print growth
8 Measurement methodologies for composite symbologies
9 Substrate characteristics
Annex A (Informative) Guidance on selection of grading parameters in application specifications
Annex B (Informative) Parameter grade overlay applied to two-dimensional bar code symbols
Annex C (Informative) Quality grading flowchart for two-dimensional matrix bar code symbols
Annex D (Normative) Symbology-specific parameters and values for symbol grading
Annex E (Informative) Interpreting the scan and symbol grades
Annex F (Informative) Substrate characteristics
Annex G (Informative) Examples of two-dimensional bar code symbol test report
Bibliography

二维条码符号印制质量的检验
1 范围
本标准规定了层排式和矩阵式二维条码符号的检验、分级以及符号整体质量评价的方法，给出了造成偏离最佳等级的可能原因及相应的纠错措施。
本标准适用于二维条码码制规范已给出参考译码算法的二维条码符号印制质量的检验，其方法也可部分或全部适用于其他码制二维条码符号的检验。
2 规范性引用文件
下列文件对于本文件的应用是必不可少的。凡是注日期的引用文件，仅注日期的版本适用于本文件。凡是不注日期的引用文件，其最新版本(包括所有的修改单)适用于本文件。
GB／T 2828.1 计数抽样检验程序 第1部分：按接收质量限(AQL)检索的逐批检验抽样计划(GB／T 2828.1-2012，ISO 2859-1：1999，IDT)
GB／T 6378.1 计量抽样检验程序 第1部分：按接收质量限(AQL)检索的对单一质量特性和单个AQL的逐批检验的一次抽样方案(GB／T 6378.1-2008，ISO 3951-1：2005，1DT)
GB／T 11186.2 漆膜颜色的测量方法 第二部分：颜色测量(GB／T 11186.2-1989，idt ISO 7724-2：1984)
GB／T 12905 条码术语
GB／T 14258 信息技术 自动识别和数据采集技术 条码符号印制质量的检验(GB／T 14258-2003，ISO／IEC 15416：2000，MOD)
GB／T 18284 快速响应矩阵码(GB／T 18284-2000，neq ISO／IEC 18004：2000)
GB／T 21049 汉信码
GB／T 35402 零部件直接标记二维条码符号的质量检验(GB／T 35402-2017，ISO／IEC TR29158：2011， MOD)
ISO／IEC 16022信息技术自 动识别和数据采集技术Data Matrix条码码制规范( Information technology-Automatic identification and data capture techniques-Data Matrix bar code symbology specification)
ISO／IEC 16023信息技术 自动识别和数据采集技术 条码码制规范 Maxi Code (Information technology-Automatic identification and data capture techniques-Bar code symbology specification-Maxi Code)
ISO／IEC 19762 信息技术 自动识别和数据采集(AIDC)技术 协调的词汇(Information technology-Automatic identification and data capture(AIDC) techniques-Harmonized vocabulary)
ISO／IEC 24778 信息技术 自动识别和数据采集技术 条码码制规范 Aztec 码码制规范(In-formation technology—Automatic identification and data capture techniques—Aztec Code bar code symbology specification)
3 术语和定义
GB／T 12905、GB／T 14258和ISO／IEC 19762界定的以及下列术语和定义适用于本文件。
3.1
像素 pixel
在一个图像采集器件(如CCD或CMOS器件)的阵列中的单个光敏单元。
3.2
有效分辨率 effective resolution
测量仪器从被测符号表面采集图像的分辨率，以每毫米的像点数或每英寸的像点数表示。其计算方法为：图像采集元件的分辨率乘以测量仪器光学系统的放大系数。
3.3
纠错容量 error correction capacity
二维条码符号(或纠错块)中用来对拒读错误和替代错误进行纠正的码字数目减去用于探测错误的码字数目。
3.4
检测区 inspection area
包括被测二维条码及其空白区的整个矩形区域。
3.5
等级阈值 grade threshold
区分某一参数两等级的分界值，其值本身是上一等级的下限值。
3.6
模块错误 module error
在二值化图像中，模块深色或浅色的状态和设计的状态发生倒置的情况。
3.7
原始图像 raw image
在X和Y坐标中，由光敏阵列每个像素所对应的实际反射率值所构成的图像。
3.8
参考灰度图像 reference grey-scale image
在X和Y坐标中，用圆形的测量孔径对原始图像进行卷积得到的图像。
3.9
二值化图像 binarised image
用整体阈值对参考灰度图像进行处理而得到的黑白两色的图像。
3.10
采样斑 sample area
直径为0.8X的圆形图像区域。X的值为被测符号经参考译码算法计算得到的平均模块宽度。如果具体应用许可的X尺寸为一个取值范围时，则计算采样斑直径时X取其中的最小值。
3.11
扫描等级 scan grade
对矩阵式二维条码符号单次扫描获得的等级，其值为由参考灰度图像和二值化图像得到的参数等级中的最低值。
3.12
模校调制比 reflectance margin
用已知模块深浅性质的正确性校正的调制比。
4 符号和缩略语
下列符号和缩略语适用于本文件。
AN：轴向不一致性(Axial Nonuniformity)
DPM：零部件直接标记(Direct Part Marking)
Ecap：纠错容量。
e：拒读错误的数目。
FPD：固有图形污损(Fixed Pattern Damage)
GN：网格不一致性(Grid Nonuniformity)
GT：整体阈值(Global Threshold)
MOD：调制比。
MARGIN：模块的模校调制比。
RM：模校调制比(Reflectance Margin)
Rmax：最高反射率，在一次扫描反射率曲线中，各单元(包括空白区)的最高反射率值，或者在矩阵式二维条码符号中所有采样斑反射率的最高值。
Rmin：最低反射率，在一次扫描反射率曲线中，各单元的最低反射率值，或者在矩阵式二维条码符号中所有采样斑反射率的最低值。
SC：符号反差。(SC＝Rmax－Rmin)
t：替代错误数目。
UEC：未使用的纠错(Unused Error Correction)
5 质量分级
5.1 概述
检测二维条码符号可得出符号质量等级。该符号等级用于符号的质量判定和过程控制，并可预测在不同环境中的识读性能。
在实际应用中，由于使用条件不同，识读设备的类型不同，可接受的二维条码符号质量等级不同。应参见附录A中A.4的内容，按本标准规定的符号等级形式，定出所需的符号等级。
应根据统计上有效的样本数量从被测样本批次中抽样，并确定可接受的最低符号等级。如果在质量控制过程或在双方的协议中没有规定抽样方案，可按GB／T 2828.1或GB／T 6378.1选择适当的抽样方案。
5.2 参数的质量等级
参数的质量等级可用数字或字母两种形式表示。数字形式用4到0表示不同的质量等级，其中4代表最高等级，0表示失败等级。字母形式用字母A、B、C、D、F表示，其中F表示失败等级。
表1给出了数字等级和字母等级的对应关系。
表1 参数数字等级和字母等级的对应关系
数字等级 字母等级
4 A
3 B
2 C
1 D
0 F
5.3 符号等级值
符号等级值按照6.2.6或7.10的规定进行计算。符号等级值保留一位小数，以4.0到0.0表示由高到低的质量等级。
符号等级值也可以用字母的形式表示。字母符号等级和数字符号等级的关系见图1。例如，数字符号等级值域在[1.5，2.5) 区间时，对应的字母等级为C。

图1 字母符号等级和数字符号等级的关系图
5.4 符号等级的表示形式
符号等级应与检测的光照条件及孔径相关联。它的表示形式为：等级／孔径／测量光波长／角度，其中：
——“等级”为5.3确定的符号等级值。
——“孔径”为孔径标号(一维扫描方式的孔径标号见GB／T 14258，矩阵式二维条码的孔径标号见7.3.3)。
——“测量光波长”为窄带照明光源峰值波长以纳米为单位的整数值，如果测量采用宽带照明光源(白光)，用字母W表示，此时应明确规定此照明的光谱响应特性，或给出光源的规格。
——“角度”为测量光的入射角，缺省值为45°。如果入射角不是45°，那么入射角度应包含在符号等级的表示中。
注：除了默认照明角度为45°外，还可选用30°和90°的照明角度。在DPM检测中照明角度需使用结合数字与字母的角度指示符，见GB／T 35402。
对于矩阵式二维条码符号，在“等级”后面加有星号表示符号周围存在反射率极值。这种情况可能干扰符号的识读，见7.7。
示例1：
3.0／05／660表示符号等级为3.0，使用的孔径为0.125 mm(孔径标号05)，测量光波长为660 nm，入射角为45°。
示例2：
3.0／10／W／30表示符号设计在宽带光照明条件下识读，测量光入射角为30°，孔径为0.250 mm(孔径标号10)。此情况下，需要给出所引用的对用于测量的光谱特性进行规定的应用标准，或给出光谱的自身特性。
示例3：
3.0*／10／670表示符号等级是在孔径为0.250 mm(孔径标号10)、光源波长670 nm情况下测量的，并且符号周围存在有潜在干扰作用的反射率极值的情况。
6 层排式二维条码符号的检测方法
6.1 概述
6.2.2和6.3规定了层排式二维条码符号质量评价方法，该方法基于GB／T 14258的规定。在符号码制适用的情况下，根据6.2.3、6.2.4和6.2.5中的方法导出符号等级。检测时应对环境光进行控制，确保其对检测结果不造成影响。测量时使用的波长和孔径应和适用的应用标准的要求一致。测量时，扫描线应和起始字符及终止字符中条的高度方向垂直，并尽量使扫描光束水平扫过行的中心，以避免跨行扫描造成的影响。在使用平面成像技术时，应通过一定的测量孔径对原始图像进行卷积，合成一定数量的、和条高方向垂直的并能足以覆盖符号中所有行的扫描线。
6.2 允许跨行扫描的符号
6.2.1 分级基础
允许跨行扫描符号的特点是扫描线出现跨行时数据仍能被识读。这类符号另一个特征是各行的起始符和终止符(或符号的等效图形，如微四一七条码的行指示符)相同，或这些图形中只有一个边的位置在相邻行间有小于1X的变化量。这些符号应根据以下几个方面进行分级：
——扫描反射率曲线分析(见GB／T 14258和6.2.2)；
——码字读出率(见6.2.3)；
——未使用的纠错(见6.2.4)；
——码字印制质量(见6.2.5)。
6.2.2 基于扫描反射率曲线分析的等级
符号的起始符、终止符或等效图形(如行指示符)根据GB／T 14258进行评价。对于数据内容所在的区域，按6.2.3、6.2.4和6.2.5所述的方法进行评价。在起始符和终止符的等级确定中应使用GB／T 14258标准中规定的所有参数。测量孔径的大小由适用的应用标准确定，或者取GB／T 14258标准中根据符号X尺寸给出的默认孔径。
扫描的次数应为10和符号的高度除以测量孔径所得的商(取整数部分)这两个数值中的较小者。应尽可能使扫描线在符号高度方向上均匀分布。例如，对于一个20行的符号，应按一定间隔对其进行10次扫描；对于一个两行的符号，对一个行可能需要在条的不同高度位置上进行多达5次扫描。针对扫描次数的选择，具体的码制规范可能会给出更详细的指导。
为了辨别条和空，每次扫描都应确定一个整体阈值。整体阈值等于最高反射率与最低反射率之和的1／2。整体阈值以上的区域应认定为空(或空白区)，整体阈值以下的区域应认定为条。
单元边缘的位置位于扫描反射率曲线上邻接单元(包括空白区)最高反射率与最低反射率的中间点处。
应使用参考译码算法评价“参考译码”和“可译码度”参数。
每次扫描中各个参数等级的最低等级值作为该次扫描的等级。扫描反射率曲线的等级应为各次扫描等级的算术平均值。
为了生产过程控制，可能需要测量条宽的平均增减量。当印刷方向和起始符和终止符高度方向一致时，印刷增量较小。如果希望全面分析印刷增量的影响，宜分别在两种方向上印制和测试符号。
6.2.3 码字读出率的等级
码字读出率(CY)衡量一维扫描从层排式二维条码中识读数据的能力。码字读出率以有效译码的码字数目占应能够译码(在调整识读角度后)的码字最大数目的百分比来表示。如果某符号其他参数等级高，而码字读出率等级低，则表明在符号的高度方向上印刷质量存在问题(如附录E中表E.1所示)。
在完成“未使用的纠错”计算(见6.2.4)后，可以得出一个正确的符号字符值表。在下面确定正确译码码字的步骤中，此符号字符值表将被用作“最终译码字符值表”。
如果某一次扫描满足下面两条件之一时，便可被纳入到码字读出率的计算：
——此扫描没有包括符号顶行或底行。通过此次扫描，至少起始符／终止符(或行指示符)其中之一以及至少一个码字或另一个终止符／起始符(或行指示符)已经被成功译码。
——扫描线包含了符号顶行或底行的识读区域，此扫描中的起始和终止符应已被成功译码。
应注意到，符号参考译码算法需要有一个扩展，以便当与起始符／终止符相邻的码字都不能译码时，探测一对起始符和终止符并译码。例如，扫描时参考译码算法本身不对四一七条码符号的起始符和终止符或微四一七条码符号中一对匹配的行指示符译码，在对这两种图形的扫描搜索时，会需要这种扩展；这样，此扩展能将没有码字(匹配的尾部图形除外)被译码的扫描纳入码字识读率的计算。但是应注意到，如果一次扫描仅扫描译码出一个起始符或终止符，而同时没有相应的第二个起始或终止符、任何其他码字或行指示符被译码，这个扫描不能作为合格的扫描。
对整个符号译码并构建符号字符值表。
对于每一次合格的扫描，将实际译出的码字和符号“最终译码字符值表”中的码字作比较，记录匹配的码字数目。累计正确译码的码字的总数，更新符号中每一个码字已被译码的次数，以及每一个行已被探测的次数。同样要记录每次扫描探测到的跨行数目(如果一个扫描线同时出现正确译码的相邻行的码字，则称为跨行)。
在处理完每次扫描后，计算目前应能够被译码的码字的最大数目：合格扫描的数目乘以符号中列单元数的乘积(不包括固定的图形，例如四一七条码的起始符和终止符，以及微四一七条码中的行指示符)。
在满足以下三个条件前，应持续对整个符号进行扫描：
a) 已译码码字的最大数目至少是符号中码字数目的十倍；
b) 符号中最高和最低的可译码行(它们并不一定是第一行和最后一行)至少被扫描三遍；
c) 已被成功识读两遍以上的码字(数据码字或纠错码字)数至少为0.9n个，其中n为符号中数据码字(非纠错码字)的个数。
示例：
一个四一七条码符号，6行16列，纠错等级为4，总码字数目为96个，其中数据码字为64个，纠错码字为32个。为了满足第一个条件，码字已被译码的最大数目至少为960。因为n等于64，为了满足第三个条件，至少应有58个码字被识读两次以上(0.9×64＝57.6)。
如果有效译码的码字总数与探测到的跨行数之比小于10：1应放弃所得的测量结果，然后调整扫描线的角度以减少跨行，重复此测量步骤。如果有效译码的码字总数与探测到的跨行数之比大于或等于10：1，要从能够识别的码字的最大数目中减去探测到的跨行数目，以补偿倾斜的影响。
码字读出率的分级方法见表2。
表2 码字读出率的分级
码字读出率(CY) 等级
CY≥0.71 4
0.64≤CY＜0.71 3
0.57≤CY＜0.64 2
0.50≤CY＜0.57 1
CY＜0.50 0
6.2.4 未使用的纠错的等级
持续扫描整个符号并译码，直至译码的码字数目趋于稳定。按下列公式计算出未使用的纠错(UEC)。
UEC＝1－(e＋2t)／Ecap
式中：
e——拒读错误数；
t——替代错误数；
Ecap——符号的纠错容量。
如果没有使用任何纠错码字，且符号能够译码，则UEC＝1；如果(e＋2t)大于Ecap，则UEC＝0。如果一个符号中有多个纠错块，应分别计算每一个纠错块中的UEC值，用其中的最小值来进行等级判定。
未使用的纠错的分级方法见表3。
表3 未使用的纠错的分级
未使用的纠错(UEC) 等级
UEC≥0.62 4
0.50≤UEC＜0.62 3
0.37≤UEC＜0.50 2
0.25≤UEC＜0.37 1
UEC＜0.25 0
6.2.5 码字印制质量的等级
本条款给出了评价可译码度、缺陷度、调制比参数的方法。该方法基于GB／T 14258中的扫描反射率曲线参数分级，同时考虑了纠错对符号质量参数可译码度、缺陷度、调制比的修正。修正方法参见附录B。
使用以下过程对这三个参数中的每一个参数进行质量评价。如果符号中存在不止一个纠错块，对于每个纠错块，都应分别进行这一过程，其中的最小值用于符号分级。
持续扫描整个符号，直到0.9n个码字(n的含义见6.2.3)已经被译码的次数大于10，或可以确认，每一个码字至少被扫描了一次而没有受到跨行的干扰。在每次扫描中，可译码度、缺陷度和调制比参数应以符号字符为单位按照GB／T 14258的规定进行测量。以上三个参数的计算应基于该次扫描反射率曲线中Rmax和Rmin值所得出的符号反差值。对于每个参数，每个码字的临时参数等级为所有扫描该码字而获得的参数等级的最高值。
如果扫描行包括不被纠错的标头字符(除了起始符、终止符及其等效图形之外)，例如四一七条码的行指示符，对于每行，首先应结合此行的上下相邻行的相应字符，对这些标头字符进行评价。这六个(对于顶行或底行，为四个)字符中最高的临时码字的等级为标头等级，这个等级用于修正此行中临时的码字等级。如果一个数据码字的临时等级比得到的头部字符的等级高，应将这个数据码字的临时等级降低到标头字符的等级。然后按照下面的方法，对由此得到的这些临时参数等级进行修正，以便将纠错的影响考虑进去。
对于每个参数，按照4级至0级和不译码的次序分别统计各级别的码字数，并累计统计大于或等于各级别的码字总数。按照如下方法将这些数目和符号的纠错能力进行比较。
对于每一个参数等级，假定低于这个等级的所有符号字符都是拒读错误，按照6.2.4的方法，根据表3所给出的阈值，导出一个假定的未使用的纠错(UEC)的等级。临时的码字参数等级应为每一个等级与其对应的假定的UEC等级的较低值。符号最终的码字参数等级应为所有等级水平中临时的码字参数等级的最高值。
注1：此假定等级和根据6.2.4计算出的符号的未使用的纠错参数不相关，也对其不影响。错误纠正能力在一定程度上可以弥补符号缺陷的影响。这种假定等级标志着弥补的程度。如果一个符号比另一个符号的纠错能力高，那么高纠错能力的符号能容忍数目更多的、参数值有问题的码字。附录F对此方法有着更详尽的说明。
表4给出了码字参数分级的示例。此例中，符号包含100个符号字符(码字)，其中数据码字为68个，纠错码字为32个。纠错码字中3个码字用于错误探测，29个码字用于错误纠正，纠错能力为29。此符号最终的码字参数等级为1级(右边列中的最高值)。
注2：调制比、缺陷度和可译码度三个参数需分别进行此计算。