Valid

GA/T 1012-2019Technical specifications for fingerprint acquisition and comparison of Resident Identity Card (English PDF)

居民身份证指纹采集和比对技术规范

Preview PDF

This is a limited preview

Buy now to download the full PDF (15 pages)

Need quotation for your management?

Get a formal quotation for GA/T 1012-2019 in 30 minutes.

Issued by

Ministry of Public Security

Level / Type

National · Recommended

Issue date

January 1, 1970

Implementation date

June 27, 2019

Scope

GA/T 1012-2019 is the English-translated version of 居民身份证指纹采集和比对技术规范.

GA/T 1012-2019 — This standard specifies the requirements for fingerprint collection and comparison of resident ID cards, technical indicators and test methods for fingerprint algorithms.

This standard applies to fingerprint collection and comparison of resident ID cards.

Found what you need?

Document preview — GA/T 1012-2019

National Standard of the People's Republic of China

ICS: 35.240.15
Classification: A 90

Issued by: Ministry of Public Security

Foreword
1 Scope
2 Normative references
3 Terms and definitions
4 Fingerprint collection requirements
4.1 Fingerprint data
4.2 Collection methods
4.3 Collection Quantity
4.4 Acquisition Finger Position
4.5 Collection Quality Requirements
4.6 Collection results
5 Fingerprint comparison requirements
5.1 Comparison method
5.2 Comparison results
6 Technical indicators and test methods of fingerprint algorithm
6.1 Technical indicators
6.2 Test methods
Appendix A (Normative Appendix) Fingerprint Data Structure
Appendix B (Normative Appendix) Application Program Interface Function Description
References

Foreword

This standard was drafted in accordance with the rules given in GB/T 1.1-2009.

This standard replaces GA 1012-2012 "Technical Specifications for Fingerprint Collection and Comparison of Resident Identity Cards", and the main changes compared with GA 1012-2012 as follows.

-Amended to a recommended standard.

Please note that some elements of this document may involve patents. The issuer of this document is not responsible for identifying these patents.

This standard is proposed by the Public Security Administration of the Ministry of Public Security.

This standard is under the jurisdiction of the Computer and Information Processing Standardization Technical Committee of the Ministry of Public Security.

This standard was drafted. The First Research Institute of the Ministry of Public Security, the Public Security Administration of the Ministry of Public Security, and the Ministry of Public Security's safety and police electronic product quality testing Xin, Beijing China Shield Security Technology Development Corporation.

1 Scope

This standard specifies the requirements for fingerprint collection and comparison of resident ID cards, technical indicators and test methods for fingerprint algorithms.

This standard applies to fingerprint collection and comparison of resident ID cards.

2 Normative references

The following documents are essential for the application of this document. For dated references, only the dated version applies to this document.

For undated references, the latest version (including all amendments) applies to this document.

GB/T 13000 Information Technology Universal Multi-octet Coded Character Set (UCS) GA/T 449 Resident ID Terms GA/T 625-2010 Live Fingerprint Image Collection Technical Specification GA 773-2008 Terminology for Automatic Fingerprint Identification System GA 774.1-2008 Fingerprint Feature Specification Part 1. Fingerprint Direction GA 774.3-2008 Fingerprint Feature Specification Part 3. Method of Labeling Fingerprint Center Point GA 774.5-2008 Fingerprint Feature Specification Part 5. Fingerprint Detail Feature Point Marking Method GA/T 893-2010 Security biometric identification application term GA/T 894.3-2010 Security Fingerprint Identification Application System Part 3. Fingerprint Image Quality GA/T 894.6-2010 Security Fingerprint Identification Application System Part 6. Evaluation Method of Fingerprint Identification Algorithm GA/T 1011-2012 General technical requirements for fingerprint reader of resident ID card ISO/IEC 19794-4 Information technology. Biometric data exchange format. Part 4. Fingerprint image data (Information technology – Biometric data interchange formats – Part 4. Finger image data) IAFIS-IC-0110 (V3.1) Wavelet Scalar Quantization (WSQ) Grayscale Fingerprint Image Compression Specification (WSQ GRAY-SCALE FINGERPRINT IMAGE COMPRESSION SPECIFICATION)

3 Terms and definitions

GA/T 449, GA/T 625-2010, GA 773-2008, GA/T 893-2010, GA/T 894.3-2010, GA/T 894.6-2010 and the following terms and definitions apply to this document.

Enrolment

When citizens register fingerprint information, the process of collecting finger fingerprint images, extracting their characteristics, and storing them.

Enrolment success

When citizens register fingerprint information, they can collect fingerprint images of their fingers, and the image quality meets the requirements for feature extraction.

Enrolment failure

When citizens register fingerprint information, they can collect fingerprint images of their fingers, and the image quality does not meet the feature extraction requirements.

Un-enrolled

When citizens register fingerprint information, they cannot collect fingerprint images of their fingers due to physical reasons such as finger disability.

Test database

Similar to the crowd distribution used by the tested fingerprint recognition application system, the fingerprint image or Feature file collection.

[GA/T 894.6-2010, definition 3.2]

4 Fingerprint collection requirements

4.1 Fingerprint data

4.1.1 Fingerprint image data

4.1.1.1 Fingerprint raw image data

Should be generated by a fingerprint collector that complies with GA/T 1011-2012. The original image data length of a single fingerprint should be 92 160 bytes.

The width should be equal to 256 pixels and the height should be equal to 360 pixels.

4.1.1.2 Fingerprint compressed image data

It should be generated by compressing the original fingerprint image data. The compression method should meet the requirements of IAFIS-IC-0110 (V3.1).

According to the file structure, see Appendix A. The length of a single fingerprint compressed image data is not more than 20 480 bytes.

4.1.1.3 Fingerprint reproduction image data

It should be generated by fingerprint image data reproduction, and the reproduction method should conform to the provisions of IAFIS-IC-0110 (V3.1).

4.1.2 Fingerprint Feature Data

Data describing the characteristics of the fingerprint. For the file structure, see Appendix A. The length of a single fingerprint is 512 bytes.

4.2 Collection method

The collection method is as follows.

a) single fingerprint plane live collection;

b) The orientation of the fingerprint should meet the requirements of GA 774.1-2008;

c) After the collection is completed, the original fingerprint image data and fingerprint reproduction image data should be cleared.

4.3 Collection quantity

Each person collects two fingerprints.

4.4 Acquisition finger order

The order of collecting fingers is as follows.

a) Register your fingerprint with your right hand and then your left hand;

b) Finger position in the order of thumb, index finger, middle finger, ring finger, little finger;

c) When the right thumb registration fails, the other fingers of the right hand should be registered;

d) when the left thumb registration fails, the other fingers of the left hand should be registered;

e) If the registration of the fingers of the right hand fails, the other fingers of the left hand should be registered;

f) When the registration of the left-hand finger fails, the other fingers of the right hand should be registered.

4.5 Collection quality requirements

The collection quality requirements are as follows.

a) The original fingerprint data of three fingerprints collected by the same finger should be greater than the preset quality value;

b) Perform feature extraction on the original fingerprint image data, generate fingerprint feature data, obtain three similarity values after cross-comparison, and calculate the average value, which should be greater than the preset similarity value;

c) Compress and reproduce the original fingerprint image data with the highest quality value, and reproduce the fingerprint image data and the corresponding number of fingerprint features According to the comparison, a similarity value is obtained, and the value should be greater than a preset similarity value.

4.6 Collection results

According to the provisions of 4.4 and 4.5, the two fingerprints collected are identified as fingerprint 1 and fingerprint 2 in sequence. The collection results include the following situations.

a) Fingerprint 1 is successfully registered and Fingerprint 2 is successfully registered;

b) Fingerprint 1 is registered successfully and Fingerprint 2 is failed to register;

c) Fingerprint 1 is successfully registered and Fingerprint 2 is not registered;

d) Fingerprint 1 registration failed, Fingerprint 2 registration failed;

e) Fingerprint 1 failed to register and Fingerprint 2 was not registered;

f) Fingerprint 1 is not registered and fingerprint 2 is not registered.

Fingers registered successfully generate fingerprint feature data and fingerprint compressed image data.

Fingers that fail to register generate fingerprint characteristic data, and take the original image data of the fingerprint with the highest image quality value to generate fingerprint compressed image data.

Unregistered fingers generate fingerprint feature data and no fingerprint compressed image data.

5 Fingerprint comparison requirements

5.1 Comparison method

Use a resident identity card reader to read the fingerprint characteristic data stored in the resident identity chip, obtain the registration result and the index code, According to the registration results, the comparison methods are divided into.

a) When the registration result is "successful registration", according to the index code, according to 4.2, collect the original fingerprint of the corresponding fingerprint of the holder the image data is compared with fingerprint characteristic data. After the comparison is completed, the original image data of the fingerprint should be cleared;

b) When the registration result is "registration failed" or "unregistered", no fingerprint comparison is performed.

5.2 Comparison results

The comparison result is divided into comparison success and comparison failure. The comparison similarity value is not less than the preset similarity value. The comparison is successful, otherwise the comparison is failure.

6 Technical indicators and test methods of fingerprint algorithm

6.1 Technical indicators

6.1.1 Registration Failure Rate

Should not be greater than 5%.

6.1.2 Equal error rate

Should not be greater than 0.1%.

6.1.3 False acceptance rate and false rejection rate

When the false rejection rate is 0.5%, the false acceptance rate should not be greater than 0.05%.

6.1.4 Response time

Should not be greater than 0.5s.

6.2 Test method

6.2.1 Test library building guidelines

Collection method. It shall meet the requirements of 4.2.

Collection equipment. It should meet the requirements of GA/T 1011-2012, with no less than two types.

Image parameters. shall comply with the provisions of 4.1.1.1.

Image format. BMP or other file format in accordance with ISO/IEC 19794-4.

Number of people. No less than 1,000, each person must collect no less than two fingers, each of the right and left hands must be no less than one finger, each finger Collect no less than three fingerprint images.

Gender distribution. 51% are male and 49% are female.

Age distribution. 17% under 15 years old, 70% between 15 and 59 years old, and 13% over 60 years old.

Industry distribution. The primary industry accounts for 38%, the secondary industry accounts for 28%, and the tertiary industry accounts for 34%.

6.2.2 Application Program Interface Functions

See Appendix B for a description of the API functions.

6.2.3 Test environment

Should comply with the provisions of Chapter 5 of GA/T 894.6-2010.

6.2.4 Test process

Should comply with the provisions of Chapter 7 of GA/T 894.6-2010.

Appendix A

(Normative appendix) Fingerprint data structure A.1 Fingerprint feature data A.1.1 Fingerprint center point the method of marking the center point of the fingerprint shall meet the requirements of GA 774.3-2008.

The fingerprint center point data consists of two parts.

a) abscissa. x, valid value is 0 ~ 255;

b) Vertical coordinate. y, valid value is 0 ~ 359.

The data of each fingerprint center point is represented by 3 bytes, as shown in Figure A.1. The first byte stores all 8 of the abscissa (x) of the center point of the fingerprint Bit value; the second byte stores the lower 8-bit value of the vertical coordinate (y) of the fingerprint center point; the third byte stores the vertical coordinate of the fingerprint center point (Y) Most significant bit, the remaining bits are reserved, and the reserved bit is set to 1.

The data of the fingerprint center point are arranged sequentially in units of 3 bytes.

a) 1st byte b) 2nd byte

c) 3rd byte

Figure A.1 Data representation method of fingerprint center point

A.1.2 Fingerprint detail feature points Fingerprint detail feature point marking method shall meet the requirements of GA 774.5-2008.

Fingerprint detail feature point data consists of three parts.

a) abscissa. x, valid value is 0 ~ 255;

b) ordinate. y, valid value is 0 to 359;

c) Direction. d, valid value is 0 ~ 359.

Each fingerprint detail feature point is represented by 4 bytes, see Figure A.2. The first byte stores the full coordinates (x) of the feature points of the fingerprint details 8-bit value; the second byte stores the lower 8-bit value of the vertical coordinate (y) of the fingerprint detail feature point; the third byte stores the fingerprint detail feature point Low 8-bit value of direction (d); the 4th byte, the first bit stores the highest bit of the vertical coordinate (y) of the feature point of the fingerprint details, and the second bit stores the fingerprint the most significant bit in the direction of the feature point (d) of the node, the remaining bits are reserved, and the reserve bit is set to 1.

Fingerprint detail feature point data are arranged sequentially in units of 4 bytes.

a) 1st byte b) 2nd byte

c) 3rd byte d) 4th byte

Figure A.2 Fingerprint detail feature point data representation method

A.2 Fingerprint feature coordinate system refer to GA 775-2008 for the fingerprint feature coordinate system, see Figure A.3. Take the upper left corner of the fingerprint image as the coordinate origin (0, 0), X-axis coordinates Values increase from left to right, Y-axis coordinate values increase from top to bottom, X and Y values are in pixels, X is greater than or equal to 0 and less than 256, and Y is greater than or equal to 0 and less than 360. The vector angle is expressed in degrees and expressed as a non-negative integer. The direction parallel to the X axis and increasing with X is 0 °. When turning counterclockwise the angle increases and the angle ranges from 0 ° to 359 °.

Figure A.3 Fingerprint feature coordinate system

A.3 Pointing code

The reference code refers to GA 777.1-2010, which uses two-digit numeric characters and converts it to a one-byte hexadecimal value.

Table A.1 Index code table

The value after the code finger name conversion

11 Right thumb 0BH

12 Right index finger 0CH

13 Right middle finger 0DH

14 Right hand ring refers to 0EH

15 Right Pinky 0FH

16 Left thumb 10H

17 Left index finger 11H

18 Left middle finger 12H

19 Left hand ring 13H

20 Left hand little finger 14H

97 Right hand uncertain finger position 61H

98 Left hand uncertain finger position 62H

99 Other uncertain fingers 63H

A.4 Fingerprint feature data file structure the fingerprint feature data file structure is shown in Table A.2. The examples are shown in Table A.3.

Table A.2 Fingerprint Feature Data File Structure

Sequence field relative position length (bytes)

1 Field 1 1 1 File header type, record character "C" (43H)

2 Field 2 2 1 Fingerprint algorithm version number

3 Field 3 3 1 Fingerprint collector code

4 Field 4 4 1 Fingerprint Algorithm Developer Code 5 field 5 5 1 Registration result code, 01H means "registration succeeded", 02H means "registration Failed ", 03H means" Unregistered ", 09H means" Unknown "

6 Field 6 6 1 Pointing code 7 field 7 7 1 Fingerprint quality value, expressed as 00H ~ 64H, value 01H indicates the lowest quality, A value of 64H means the highest quality, and a value of 00H means "Unknown"

8 Field 8 8 ~ 19 12 Reserved 9 field 9 20 1 Number of fingerprint detail feature points. "Registration failed" and "Not registered" fingerprint settings Set to 00H

10 Field 10 21 ~ 22 2 Length of fingerprint characteristic data, in bytes

11 Field 11 23 ~ 31 9 Fingerprint center point data

12 Field 12 32 ~ 511 480 Fingerprint detail feature point data 13 field 13 512 1 parity bit, the parity value is generated using CRC-8

Table A.3 Example of Fingerprint Feature Data File Structure

Sequence field relative position value (hexadecimal)

1 Field 1 1 43 "C" indicates fingerprint characteristic data

2 Field 2 2 03 Fingerprint algorithm version number is 3

3 Field 3 3 11 Fingerprint collector code is 17

4 Field 4 4 0A Fingerprint algorithm developer code is 10

5 Field 5 5 01 "Registration succeeded" 6 field 6 6 10 left thumb

7 Field 7 7 50 Fingerprint quality value 80

8 Field 8 8-19

FF FF FF FF FF FF FF FF FF FF FF FF Reserved item

9 Field 9 20 33 Number of fingerprint detail feature points is 51 10 field 10 21 ~ 22 00 D5 fingerprint characteristic data length is 213 bytes

11 Field 11 23 ~ 31 ... Fingerprint center point data

12 Field 12 32 ~ 511 ... Fingerprint detail feature point data 13 field 13 512 6D CRC check bit A.5 Fingerprint compressed image data file structure the file structure of fingerprint compressed image data is shown in Table A.4, and examples are shown in Table A.5.

Table A.4 Fingerprint compressed image data file structure

Sequence field relative position length (bytes)

1 Field 1 1 1 File header type, record character "Y" (59H)

2 Field 2 2 1 Fingerprint algorithm version number

3 Field 3 3 1 Fingerprint collector code

6 Field 6 6 1 Pointing code

7 Field 7 7 ~ 20 14 Reserved

8 Field 8 21 ~ 22 2 Fingerprint compressed image data length, in bytes 9 field 9 23 ~ X fingerprint compressed image data 10 field 10 Y 1 parity bit, the parity value is generated using CRC-8. Y = 23 X

Table A.5 Example of Fingerprint Compressed Image Data File Structure

Sequence field relative position value (hexadecimal)

1 Field 1 1 59 "Y" means fingerprint compressed image data

2 Field 2 2 03 Fingerprint algorithm version number is 3

3 Field 3 3 11 Fingerprint collector code is 17

4 Field 4 4 0A Fingerprint algorithm developer code is 10

5 Field 5 5 01 "Registration succeeded" 6 field 6 6 10 left thumb

7 Field 7 7-20

FF FF FF FF FF FF FF FF FF FF FF FF FF FF Reserved item 8 field 8 21 ~ 22 4F 01 compressed image data length is 20225 bytes

9 Field 9 23 ~ 20247 ... Fingerprint compressed image data 10 field 10 20248 D6 CRC check digit BB Appendix B (Normative appendix) Application program interface function description B.1 File format B.1.1 File generation the interface functions are released in the form of Windows Dynamic Link Library (DLL). Supported operating systems should include Windows XP or compatible systems.

Interface functions should be output in a dynamic link library and support multithreading. The program is compiled into a 32-bit version.

B.1.2 File name

The dynamic link library file name is. ID_Fpr.dll.

B.2 Interface functions the interface functions are shown in Table B.1.

The fingerprint image data in the interface function description does not include fingerprint compressed image data.

Table B.1 Interface functions

Number name description

1 FP_GetVersion Get version information

2 FP_Begin initialization operation

3 FP_FeatureExtract fingerprint image feature extraction

4 FP_FeatureMatch Fingerprint Feature Data Comparison

5 FP_ImageMatch Comparison of fingerprint image data and fingerprint feature data

6 FP_Compress Fingerprint image data compression

7 FP_Decompress fingerprint image data reproduction

8 FP_GetQualityScore Get fingerprint image quality value

9 FP_GenFeatureFromEmpty1 Generate fingerprint feature data of "registration failed"

10 FP_GenFeatureFromEmpty2 Generate "unregistered" fingerprint feature data

11 FP_End End operation B.3 Function description B.3.1 Obtaining version information Function prototype. int __stdcall FP_GetVersion (unsigned char code [4]).

"1201", the code is filled in as follows. code [0] = '1', code [1] = '2', code [3] = '0', code [4] = '1'. Output parameters.

Return value. If the call is successful, it returns 1; otherwise, it returns an error code. The error code value should meet the requirements of B.4.

Description. Get version information.

B.3.2 Initialization operation Function prototype. int __stdcall FP_Begin ().

Parameters. None.

Return value. If the call is successful, it returns 1; otherwise, it returns an error code. The error code value should meet the requirements of B.4.

Description. Initialization operation.

B.3.3 Fingerprint image feature extraction Function prototype. int __stdcall FP_FeatureExtract (unsigned char cScannerType, unsigned char cFingerCode, unsigned char * pFingerImgBuf, unsigned char * pFeatureData).

Parameters. unsigned char cScannerType Fingerprint collector code. Input parameters.

unsigned char cFingerCode Refers to the bit code. Input parameters.

unsigned char * pFingerImgBuf Fingerprint image data pointer. The fingerprint image is in RAW format. Input parameters.

unsigned char * pFeatureData Fingerprint feature data pointer, stores the generated fingerprint feature data.

The user allocates memory space. The structure of the fingerprint feature data file should meet the requirements of Appendix A. Output parameters.

Return value. If the call is successful, it returns 1; otherwise, it returns an error code. The error code value should meet the requirements of B.4.

Note. Feature extraction is performed on fingerprint image data to generate fingerprint feature data.

B.3.4 Comparison of fingerprint characteristic data Function prototype. int __stdcall FP_FeatureMatch (unsigned char * pFeatureData1, unsigned char * pFeatureData2, float * pfSimilarity).

Parameters. unsigned char * pFeatureData1 Fingerprint feature data pointer 1. Input parameters.

unsigned char * pFeatureData2 Fingerprint feature data pointer 2. Input parameters.

float * pfSimilarity Similarity, ranging from 0.00 to 1.00, value 0.00 means no match, value 1.00 Indicates an exact match. Output parameters.

Return value. If the call is successful, it returns 1; otherwise, it returns an error code. The error code value should meet the requirements of B.4.

Explanation. The two fingerprint feature data are compared to obtain the similarity value.

B.3.5 Comparison of fingerprint image data and fingerprint characteristic data Function prototype. int __stdcall FP_ImageMatch (unsigned char * pFingerImgBuf, unsigned char * pFeatureData, float * pfSimilarity).

Parameters. unsigned char * pFingerImgBuf Fingerprint image data pointer, the fingerprint image is in RAW format. Input parameters.

unsigned char * pFeatureData Pointer to fingerprint feature data. Input parameters.

float * pfSimilarity Similarity, ranging from 0.00 to 1.00, value 0.00 means no match, value 1.00 Indicates an exact match. Output parameters.

Return value. If the call is successful, it returns 1; otherwise, it returns an error code. The error code value should meet the requirements of B.4.

Explanation. The fingerprint image data is compared with the fingerprint characteristic data to obtain the similarity value.

B.3.6 Fingerprint image data compression Function prototype. int __stdcall FP_Compress (unsigned char cScannerType, unsigned char cEnrolResult, unsigned char cFingerCode, unsigned char * pFingerImgBuf, int nCompressRatio, unsigned char * pCompressedImgBuf, unsigned char strBuf [256]).

Parameters. unsigned char cScannerType Fingerprint collector code. Input parameters.

unsigned char cEnrolResult Registration result code. Input parameters.

unsigned char cFingerCode Refers to the bit code. Input parameters.

unsig...

......
This preview omits tables, figures, formulas and parts of the technical clauses. The complete document — 15 pages — is available in the English PDF.

Referenced standards

Normative references

GB/T 13000Information technology - Universal coded character set(UCS)

ISO/IEC 19794

How to Download GA/T 1012-2019

1
Add to cart
Click "Download PDF" on this page and choose "Buy this standard". You can add more standards before checkout.
2
Checkout
Enter your email and billing details. Payment is processed securely by Stripe (cards or bank transfer). Prefer bank details? Request a quotation (min. order $300).
3
Instant delivery (0–9 sec)
Delivery is automatic: within seconds of payment you'll receive an email with a secure download link. The link stays valid for 72 hours.
4
Invoice included
A tax invoice is attached to the confirmation email. Need a custom invoice? Contact us.