4 December 2008
|
Home |
Search |
Contact |
Help
| |
 | |
|
Data Format for the Interchange of Fingerprint, Facial &
SMT Information Interpol implementation ANSI/NIST-ITL 1a-1997 prepared by The Interpol AFIS Expert Working Group Version No 3 – June 2001 GLOSSARY
ANSI/NIST STANDARD: DATA FORMAT FOR THE INTERCHANGE OF FINGERPRINT IMAGE INFORMATION
In 1986 the American National Bureau of Standards published a standard to facilitate the interchange of fingerprint image information entitled "Data Format for the Interchange of Fingerprint Information" (ANSI/NBS-ICST 1-1986). Following a relatively exhaustive review procedure which included the UK Home Office and other US and Canadian law enforcement agencies, this was revised by the American National Institute of Standards and Technology (NIST) and issued as ANSI/NIST-CSL 1-1993. In 1997 the standard has been expanded to handle facial images and scar, mark and tattoo (SMT) image data. This expansion is issued as ANSI/NIST-ITL 1a-1997). As is often the case with standards such as these, it is defined fairly broadly so as to appeal to a large set of potential users. Hence, the standard provides some features not needed by some organisations. In addition, the standard also includes two user defined record types ("Type-2" and "Type-7") which are intentionally not defined within the standard, but rather are to be "user defined". The present document, the Interpol Implementation (INT-I), has been written with the intention of supplementing the ANSI/NIST publication for the guidance of members of the International Criminal Police Organisation. The INT-I has been drafted noting the following general points:
It should be noted that the records described in the ANSI/NIST standard and INT-I are not intended for manual entry and interpretation: rather they are intended for transmission of information between computers. It is also important to note that some TOTs, and fields within records, may not be appropriate for certain transactions between particular agencies. For example, many agencies may not allow a remote site to add a record to its database, or there may be national legal objections to sending respondent images over a wide area network before they had been verified. However, in the spirit of open standards, and with the aim of excluding only the absolute minimum of information exchange, all such transactions are specified in INT-I but with the expectation that they would be blocked by the systems involved. The following section describes the general structure of the ANSI/NIST standard and goes on to describe the various record types (Type-1, Type-2, ...). In addition this section also details the use of each of the record types.
A fingerprint file, as specified in the ANSI/NIST-CSL 1-1993 and ANSI/NIST-ITL 1a-1997 standards, consists of several logical records. There are nine types of record. Appropriate ASCII separation characters are used between each record and the fields and subfields within the records. 1. File header
2. Descriptive text (user defined)
3. Low resolution grey-scale image
4. High resolution grey-scale image
5. Low resolution binary image
6. High resolution binary image
7. User-defined image
8. Signature image
9. Minutiæ record
10. Type-10 Logical Record: Facial and/or SMT Binary Image Record
This record describes the structure of the file, the type of the file, and other important information. The character set used for Type-1 fields shall contain only the 7-bit ANSI code for information interchange. 1.1 Fields for Type-1 Logical Record 1.1.1 Field 1.01: Logical Record Length (LEN) This field contains the total count of the number of bytes in the whole Type-1 logical record. The field begins with "1.01:", followed by the total length of the record including every character of every field and the information separators. 1.1.2 Field 1.02: Version Number (VER) To ensure that users know which version of the ANSI/NIST standard is being used, this four byte field specifies the version number of the standard being implemented by the software or system creating the file. The first two bytes specify the major version reference number, the second two the minor revision number. For example, the original 1986 Standard would be considered the first version and designated "0100" while the present standard is "0200". 1.1.3 Field 1.03: File Content (CNT) This field lists each of the records in the file by record type and the order in which the records appear in the logical file. It consists of one or more subfields, each of which in turn contains two information items describing a single logical record found in the current file. The subfields are entered in the same order in which the records are recorded and transmitted. The first information item in the first subfield is "1", to refer to this Type-1 record. It is followed by a second information item which contains the number of other records contained in the file. This number is also equal to the count of the remaining subfields of field 1.03. Each of the remaining subfields is associated with one record within the file, and the sequence of subfields corresponds to the sequence of records. Each subfield contains two items of information. The first is to identify the Type of the record. The second is the record's IDC which is generally in the range 0-16 (one Type-1, one Type-2, and 14 Type-4), but could be much higher if Type-7, Type-8, Type-9 and/or Type-10 records are included. The "US" character shall be used to separate the two information items. 1.1.4 Field 1.04 Type of Transaction (TOT) This field contains a three letter mnemonic designating the type of the transaction. These codes are different from those used by other implementations of the standard.
The definition of these transactions implies that what appears to the officer performing a search as one transaction may, in fact, involve a number of separate transactions between the officer’s workstation and the remote site. It is likely that a system would be designed to block transactions initiated by a remote agency unless it had been specifically authorised by a senior user of the receiving agency. One limitation of the standard is that it is not permissable for the file to have more than one transaction field. Thus if, say, a latent is to be searched against both the latents database and the prints database, two separate files must be sent. Table 1 lists which records are permissible in the various transactions.
Key:
1.1.5 Field 1.05 Date of Transaction (DAT) This field indicates the date on which the transaction was initiated and must conform to the ISO standard notation of
where YYYY is the year, MM is the month and DD is the day of the month. Leading zeros are used for single figure numbers. For example, "19931004" represents the 4 October 1993. 1.1.6 Field 1.06 Priority (PRY) This optional field defines the priority, on a level of 1 to 4, with which the request is to be treated. "1" is the highest priority and "4" (the default if no priority field is present) the lowest. It is up to the receiving agency to define its policy on how each priority level is interpreted. 1.1.7 Field 1.07 Destination Agency Identifier (DAI) This field specifies the destination agency for the transaction. It consists of two information items in the following format
The first information item contains the Interpol Country Code, defined in ISO 3166, two alpha-numeric characters long. The second field, agency, is a free text identification of the agency, up to a maximum of 32 alpha-numeric characters. 1.1.8 Field 1.08 Originating Agency Identifier (ORI) This field specifies the file originator and has the same format as the DAI (Field 1.07). 1.1.9 Field 1.09 Transaction Control Number (TCN) This is a control number for reference purposes. It should be generated by the computer and have the following format:
where YY is the year of the transaction, SSSSSSSS is an eight-digit serial number, and A is a check character generated by following the procedure given in Appendix 2. Where a TCN is not available, the field, YYSSSSSSSS, is filled with zeros and the check character generated as above. 1.1.10 Field 1.10 Transaction Control Response (TCR) Where a request was sent out, to which this is the response, this optional field will contain the transaction control number of the request message. It therefore has the same format as TCN (Field 1.09). Where a TCR is not available, the field, YYSSSSSSSS, is filled with zeros and the check character generated as in TCN (Field 1.09). 1.1.11 Field 1.11 Native Scanning Resolution (NSR) This field specifies the normal scanning resolution of the system supported by the originator of the transaction. It allows the recipient of a search request to send the response(s) at either the minimum (or default) scanning rate of 19.68 pixels/mm (500 pixels/inch) or, if it has the ability, at the scanning rate of the system which made the request. The resolution is specified as two numeric digits followed by the decimal point and then two more digits (e.g. "20.00"). If both recipient and sender use the same native sampling resolution it may be more efficient and less error prone if both systems exchange images at their native sampling resolution rather than using the default rate specified in the standard. The current ANSI/NIST standard allows any sampling rate from 500 to 520 pixels/inch, but the intention is for new systems to adopt 500 pixels/inch or 19.68 pixels/mm. 1.1.12 Field 1.12 Nominal Transmitting Resolution (NTR) This five-byte field specifies the nominal transmitting resolution for the images being transmitted. The resolution is expressed in pixels/mm in the same format as NSR (Field 1.11)
The structure of most of this record is not defined by the ANSI/NIST standard. The record contains information of specific interest to the agencies sending or receiving the file. To ensure that communicating fingerprint systems are compatible the INT-I requires that only the fields listed below are contained within the record. This document specifies which fields are mandatory and which optional, and also defines the structure of the individual fields. Currently the numbers 001 to 074 have been assigned to specific fields. Numbers 075 to 199 are reserved for future additions to the INT-I. The fields 200 to 999 are outside the scope of the INT-I and may be used for national requirements or by system implementers for information specific to their systems. A file may contain only a small subset of these fields, depending on the transaction taking place. The character set used for Type-2 fields shall contain only the 7-bit ANSI code for information interchange. 2.1 Fileds for Type-2 Logical Record Fields 2.001 to 2.003 are mandatory in all records. They give essential information about the record. 2.1.1 Field 2.001 Logical Record Length (LEN) This mandatory field contains the length of this Type-2 record, and specifies the total number of bytes including every character of every field contained in the record and the information separators. 2.1.2 Field 2.002 Image Designation Character (IDC) The IDC contained in this mandatory field is an ASCII representation of the IDC as defined in the file content field of the Type-1 record. 2.1.3 Field 2.003 System Information (SYS) This field is mandatory and contains four bytes which indicate which version of the INT-I this particular Type-2 record complies with. This feature gives the INT-I the ability to evolve as necessary while still allowing a system to process transactions generated by a system complying with an older version of the INT-I. The first two bytes specify the major version number, the second two the minor revision number. For example, this implementation is version 3 revision 0 and would be represented as «0300». Fields 2.004 and 2.005 contain general information regarding the file. Their use is optional in most transactions. 2.1.4 Field 2.004 Date of Record (DAR) This specifies the date, in ISO format, on which the record was first created, and is formatted according to the ISO standard
where YYYY is the year, MM is the month, and DD is the day, as explained in DAT (Field 1.05). This field will probably be generated automatically. 2.1.5 Field 2.005 Date of Last Update (DLU) This specifies the most recent date on which the data were changed in the record. The field is formatted in the ISO standard of
where YYYY is the year, MM is the month, and DD is the day, as explained in DAT (Field 1.05). Like Date of Record (see above) this field would be generated by the system when the fingerprint record is amended. Fields 2.006 to 2.016 are reference information which give information about the nature of the file and its contents. They are, in general, optional, although some are mandatory for certain transactions. 2.1.6 Field 2.006 Send Copy To (SCT) This field indicates to the receiver to send the response of the transaction to other stations. It consists of one or more subfields, each having the format of DAI (Field 1.07), namely up to two alpha-numeric characters for the Interpol Country Code and up to 32 alpha-numeric characters of free text. 2.1.7 Field 2.007 Case Number (CNO) This is a number assigned by the local fingerprint bureau to a collection of latents found at a scene-of-crime. The following format is adopted:
where CC is the Interpol Country Code, two alpha-numeric characters in length, and the number complies with the appropriate local guidelines and may be up to 32 alpha-numeric characters long. This field allows the system to identify latents associated with a particular crime. 2.1.8 Field 2.008 Sequence Number (SQN) This specifies each sequence of latents within a case. It can be up to four numeric characters long. A sequence is a latent or series of latents which are grouped together for the purposes of filing and/or searching. This definition implies that even single latents will still have to be assigned a sequence number. In the case of search requests the field is included for identification purposes: if the remote system is an AFR system it can use the case number, sequence number and latent identifier to determine whether it already has an AFR encoding of the latent. This field together with MID (Field 2.009) may be included to identify a particular mark within a sequence. 2.1.9 Field 2.009 Latent Identifier (MID) This specifies the individual latent within a sequence. The value is a single letter, with 'A' assigned to the first latent, 'B' to the second, and so on up to a limit of 'J'. This field is used analog to the latent sequence number discussed in the description for SQN (Field 2.008). 2.1.10 Field 2.010 Criminal Reference Number (CRN) This is a unique reference number assigned by a national agency to an individual who is charged for the first time with committing an offence. Within one country no individual ever has more than one CRN, or shares it with any other individual. However, the same individual may have Criminal Reference Numbers in several countries, which will be distinguishable by means of the country code. The CRN field consists of at least one subfield, which in turn consists of two information items. The following format is adopted for each subfield:
where CC is the Interpol Country Code, two alpha-numeric characters in length, and the number complies with the appropriate national guidelines of the issuing agency, and may be up to 32 alpha-numeric characters long. In the case of CPS transactions, the fingerprint officer may believe that he/she already knows the CRN (e.g. as the result of a name check). This field will then specify that CRN, and will allow a verification to be carried out before a print-to-print search on the complete database is initiated. In the case of PMS searches this field may be used to specify the CRN of the person whose ten-prints are to be searched against an Unidentified Latents collection. This may be useful if the remote system has already encoded and filed the individual’s prints. In an SRE transaction in which the identy of the subject is certain, the CRN is of that individual. For example, Agency A might initiate an SRE transaction, in response to a CPS from Agency B, after an Agency A fingerprint expert has examined the fingerprints and identified the individual. System design should ensure that the response makes clear what has been done. Similarly, for an IRQ the field may be used to find a given individual’s prints in the collection. In this case the responding IMR may contain the same CRN. 2.1.11 Field 2.011 Other Reference Number (ORN) This is a unique reference number for a ten-print set which does not have a CRN. It is very similar in format and function to CRN (Field 2.010). The field consists of at least one subfield, which in turn consists of two information items. The following format is adopted for each subfield:
where CC is the Interpol Country Code, two alpha-numeric characters in length, type_number consists of up to 32 alpha-numeric characters of free text defining the type of reference number, and ref_number complies with the appropriate national guidelines of the issuing agency, and may be up to 32 alpha-numeric characters long. 2.1.12 Field 2.012 Miscellaneous Identification Number (MN1) Any miscellaneous identification numbers may be entered in this and the following four fields (MN1 to MN5). Each of these fields may have a maximum length of 32 alpha-numeric characters. 2.1.13 Field 2.013 Miscellaneous Identification Number (MN2) 2.1.14 Field 2.014 Miscellaneous Identification Number (MN3) 2.1.15 Field 2.015 Miscellaneous Identification Number (MN4) 2.1.16 Field 2.016 Miscellaneous Identification Number (MN5) Fields 2.017 to 2.025 are used to give information about specific images involved in the transaction. For this reason, if the fingerprints were not taken at the same session each must have as many subfields as there are images in the file. These subfields form a list of data, each consecutive element relating to the respective image in the file. The requirement for these fields depends upon the transaction being undertaken. 2.1.17 Field 2.017 Finger Number (FNU) This field consists of a number of subfields. The first subfield consists of one of the letters T, F or I, which have the following meaning:
If the first subfield is I then any of Fields 2.018 to 2.025 inclusive which are used will contain a number of subfields, each relating to the respective image in this FNU field. 2.1.18 Field 2.018 Fingerprint Identification Byte (FIB) This field consists of one subfield for each corresponding subfield in FNU (Field 2.017) Each subfield contains two characters which are used to indicate the reason for fingerprinting. It has one of the following values:
In the case of "10", RFP (Field 2.021) can be used to give a more detailed description. In search transactions the field specifies the nature of a transmitted set of ten-prints. In SRE and IMR responses the field will specify the nature of the ten-prints being examined and will therefore effectively echo the original search or image retrieval request. 2.1.19 Field 2.019 Date Fingerprinted (DPR) This field consists of one subfield for each corresponding subfield in FNU (Field 2.017), and is intended to be used for both prints and latents. For prints this field contains the date on which the subject was fingerprinted and refers to the date on which the prints included in the transaction were taken. For latents this field specifies the date on which the latent was inspected at the scene of the crime by a scene examiner. The format is the ISO standard of
where YYYY is the year, MM is the month, and DD is the day, as explained in DAT (Field 1.05). 2.1.20 Field 2.020 Time of Fingerprinting (TOF) This field consists of one subfield for each corresponding subfield in FNU (Field 2.017), and is intended for use with prints. It specifies the time at which fingerprints were taken. The format is
where HH is a two digit hour reference and MM a similar minute reference. Standard twenty-four hour clock notation will be used (eg "0730", "1752" etc). Midnight should be recorded as either "2359" or "0001", instead of "2400" or "0000". 2.1.21 Field 2.021 Reason Fingerprinted (RFP) This field consists of one subfield for each corresponding subfield in FNU (Field 2.017). It is an alpha-numeric field with a maximum length of 128 alpha-numeric characters and is to allow the human operator to enter an extra message, for example giving further details of the reason for fingerprinting or information about how a search is to be carried out. 2.1.22 Field 2.022 Place Of Arrest (POA) This field consists of one subfield for each corresponding subfield in FNU (Field 2.017). Each subfield specifies the place of arrest, or the place where the fingerprints were taken, in the same format as DAI (Field 1.07). 2.1.23 Field 2.023 Owning Bureau (OBU) This field consists of one subfield for each corresponding subfield in FNU (Field 2.017), and is intended for use with both prints and latents. The format of the field is the same as that used in POA (Field 2.022) and DAI (Field 1.07). 2.1.24 Field 2.024 Date of Notice (DON) This field consists of one subfield for each corresponding subfield in FNU (Field 2.017). It specifies the Date of Notice of the record. The format is according to the ISO standard of
where YYYY is the year, MM the month and DD the day, as explained in DAT (Field 1.05). 2.1.25 Field 2.025 Station Inputting Mark (SIM) This field specifies the local office inputting the latent. Its format is the same as DAI (Field 1.07). Fields 2.026 to 2.028 are used to specify information about image quality and the type of pattern classification used. The precise definition of these fields is to be determined and will be published in a future revision of the INI-I. 2.1.26 Field 2.026 Quality Measure (QLM) This field contains at least one subfield. Each subfield contains two information items: the first is the IDC code of the finger to which this subfield refers, and the second is a quality measure assigned either by the fingerprint examiner or automatically by the system. The format of the quality measure has not yet been determined. 2.1.27 Field 2.027 Coarse Classification of Patterns (CCP) This field contains at least one subfield. Each subfield contains at least two information items: the first is the IDC code of the finger to which this image refers, and the second is the character representing the coarse classification. Each subsequent information item specifies a further coarse classification of the fingerprint, which permits allowances. The coarse classification is represented by an alpha-numeric character, but the precise classification to be employed is yet to be determined. 2.1.28 Field 2.028 Fine Classification of Patterns (FCP) This represents the detailed code for the fingerprint pattern class. It contains a number of subfields where each subfield contains at least three information items. The first information item is the IDC of the relevant image. The second is either S or F, recommending whether the pattern is for filing or searching. The third is a character string with between 1 and 4 alpha-numeric characters containing the fine classification. If the fine classification is doubtful, more than one fine classification may follow the IDC code. The precise classification to be employed is yet to be determined. Fields 2.029 to 2.051 are used in several transactions to convey personal information relating to the file. This information is generally used for filing purposes, although there are some cases where this personal information may be used for searches. 2.1.29 Field 2.029 Nominal File (NLF) This field consists of two subfields. The first subfield is a single byte. The value of the first subfield is « 0 » if the data contained in fields 2.030 to 2.051 is transmitted within this Type 2 record. The value of the first subfield is « 1 » if the data is transmitted using a Formatted Message. If the first subfield is « 0 » then the second subfield is blank. If the first subfield is « 1 » then the second contains a reference to the Formatted Message in free text, up to 32 alpha-numeric characters. This field contains the names of the subject. The format is:
For instance Charles Peter Bell would appear as "BELL/CHARLES/PETER". If only the family name (surname) is known then this is followed by a single slash. The entire field is limited to 64 characters including the slashes. Spaces, apostrophes, hyphens and full stops that occur within a component name should be entered as such. If the name is longer than 64 characters the 64th character should be a plus sign. The plus sign can only be used in the final position. 2.1.31 Field 2.031 Maiden Name (MNA) The format of this field is identical to NAM (Field 2.030), and is limited to 64 characters including the slashes. Spaces, apostrophes, hyphens and full stops that occur within a component name should be entered as such. If the name is longer than 64 characters the 64th character should be a plus sign. The plus sign can only be used in the final position. 2.1.32 Field 2.032 Address (ADD) This field contains the address of the subject, in free text up to 128 alpha-numeric characters. 2.1.33 Field 2.033 True Identity (TRU) This field contains information on how the individual's true identity was determined. It consists of two information items in the following format:
The first subfield contains a single binary digit, A, which is "0" (the default value) if true identity has not been established, or "1" if a positive result to an investigation has been obtained. The description consists of 128 characters of free text, describing the manner in which the true identity was established. 2.1.34 Field 2.034 Aliases (AKA) If present this field consists of at least one subfield. Each subfield is formatted as NAM (Field 2.030). Its use is identical to NAM. 2.1.35 Field 2.035 Date of Birth (DOB) This field specifies the date of birth in ISO format:
where YYYY is the year, MM is the month, and DD is the day, as explained in DAT (Field 1.05). 2.1.36 Field 2.036 Date of Birth Range (DBR) Sometimes it will not be possible to specify the date of birth exactly. In
such circumstances a date of birth range may specified. The Format of the
field is YYYYMMDDQYYYYMMDD where the two strings YYYYMMDD are the two ISO dates defining the range, and
Q is qualifier, whose value is always 4, separating the two dates. Thus if the
range is between 1st December 1995 and 31st January 1996, the field value will
be «19951201419960131». 2.1.37 Field 2.037 Place of Birth (POB) This field consists of up to three information items and specifies the place of birth. The format is
where CC is the Interpol Country Code, two alpha-numeric characters long, country is the free text equivalent, up to 32 characters long, and town is the free text name of the town of birth, up to 32 characters long. 2.1.38 Field 2.038 Nationality (NAT) This field consists of up to two information items and specifies the nationality of the fingerprinted subject. The format is
where CC is the Interpol Country Code, two alpha-numeric characters long, and nationality is the free text equivalent, up to 32 characters long. This is a single letter code representing the sex of the subject
2.1.40 Field 2.040 Colour (COL) This is a single letter code representing the colour of the fingerprinted subject. It is a one letter code:
2.1.41 Field 2.041 Height (HGT) This field specifies the subject’s height. The first letter indicates whether the height is in feet and inches or in centimetres:
The first letter is followed by a three-digit number (including leading zeros) representing the height. For an imperial measure the first digit indicates the feet whilst the second and third digits represent the inches which may range from 00-11. For instance, for a person 5ft 8in (173 cm) tall this field would be either "F508" or "M173". 2.1.42 Field 2.042 Build (BLD) This field is up to 32 characters long, and contains a free-text description of the subject's build. This field is up to 32 characters long, and contains a free-text description of the colour and style of the subject's hair. This field is up to 256 characters long, and contains a free-text description of the subject's face. 2.1.45 Field 2.045 Languages Spoken (LAN) This field consists of up to ten subfields, each of which consists of two information items, and specifies the languages spoken by the fingerprinted subject. The format of each subfield is free text up to 32 characters long. 2.1.46 Field 2.046 Photograph Number (PHO) This field consists of two subfields, the first subfield being a single character. If no photograph of the subject is available then the subfield contains "0" and the second subfield is empty. If a photograph is available then the first subfield contains "1". The second subfield is up to 32 alpha-numeric characters long, and contains the reference number of the photograph. 2.1.47 Field 2.047 Passport Number (PSP) This field contains a passport number and is up to 32 alpha-numeric characters long. 2.1.48 Field 2. 048 Marks etc (MAR) This field consists of up to 16 subfields, each being 64 alpha-numeric characters of free text describing marks, scars and tattoos of the fingerprinted subject. If images of the marks, scars and/or tattoos exists they should be included as Type 10 records within the transaction. 2.1.49 Field 2.049 Occupation (OCC) This field contains a free text description of the subject's occupation, and is up to 64 characters long. 2.1.50 Field 2.050 Warning ( WNG) This is a free text field, up to 32 alpha-numeric characters, warning if the subject is dangerous ( eg. carries firearms, violent, etc.) 2.1.51 Field 2.051 Modus Operandi (MDO) This field contains a free text description of the subject's normal modus operandi, and is up to 64 characters long. Fields 2.052 to 2.063 are search criteria. In most cases, it is possible to allow a range of values to be specified within a type of criterion as well as exact values. Special fields are included to allow this. In these cases, either, but not both, of the fields may be used to search on the basis of a particular aspect. 2.052 Geographical Area of Crime (GAC) This field indicates the geographical area in which the crime was committed. The field consists of one or more subfields in the following format:
The first subfield contains the Interpol Country Code, two alpha-numeric characters long. The second field, area, is a free text identification of the area, up to a maximum of 256 alpha-numeric characters, and may include the actual address at which the offence was committed. The third (optional) subfield, up to 16 alpha-numeric characters long, can contain a reference number as generated by a Geographical Information System. The format of the number is unspecified. 2.1.53 Field 2.053 Geographical Search Area (GSA) This field is split into one or more subfields, each specifying a geographical region as two information items in the following format: where CC is the Interpol Country Code, two alpha-numeric characters long, and the second information item, area, is a free text identification of the area, up to a maximum of 32 alpha-numeric characters. If omitted in a latent search, then the system should ensure that the geographical search area defaults to the area specified in GAC (Field 2.052). The field may also be present in search and image responses, when the field would be simply a copy of that appearing in the original request transaction. 2.1.54 Field 2.054 Offence Type (OTY) This field identifies the type of crime committed. It consists of up to seven subfields, each of which is free text up to 64 alpha-numeric characters long. 2.1.55 Field 2.055 Date of Offence (DOO) This field specifies the date in ISO format (YYYYMMDD) on which the offence was committed. In searches it may be used interchangeably with DOR (Field 2.056). If present in a search it records the date on which the crime was committed. 2.1.56 Field 2.056 Date of Offence Range (DOR) Sometimes it will not be possible to specify the date of the offence exactly. In such circumstances a date of offfence range may be specified. The format of the field is
where the two strings YYYYMMDD are the two ISO dates defining the range, and Q is a qualifier, whose value is always "4", separating the two dates. Thus if the range is between 1st December 1995 and 31st January 1996, the field value will be "19951201419960131". The dates may include the wildcard character *, which can be used both if the start or end of the period is uncertain (e.g. "********419940101") and if the dates cannot be specified exactly (e.g. "1992****419930101"). The use of the field is similar to that of DOO (Field 2.055). 2.1.57 Field 2.057 Date of Offence Search Range (DSR) This field has the same format as DOR (Field 2.056), including the same wildcarding mechanism. It is analogous to GSA (Field 2.053) and can be used to specify a search range. Only prints or latents whose DOO or DOR (Fields 2.055 and 2.056 respectively) lie within the listed date range will be included in the search. 2.1.58 Field 2.058 Time of Offence (TOO) This field specifies the time at which the offence was thought to be committed. The format is the same as TOF (Field 2.020). 2.1.59 Field 2.059 Time of Offence Range (TOR) When it is not possible to specify the exact time of the crime, a range of time may be recorded, similarly to DOR (Field 2.056). The format of this range is
where the strings HHMM are the two four-figure time references defining the range, as used in TOF (Field 2.020), and Q is a qualifier, whose value is always "4", separating the two times. 2.1.60 Field 2.060 Time of Offence Search Range (TSR) This field has the same format as TOR (Field 2.059). It is analogous to GSA (Field 2.053) and can be used to specify a search range. Only prints or latents whose TOO or TOR (Fields 2.058 and 2.059 respectively) lie within the listed date range will be included in the search. 2.1.61 Field 2.061 Time Limit (TLM) If there is a time limit within which a prosecution is to be processed then this field contains the latest date by which the results must be received. Its format is the ISO standard notation, as in DAT (Field 1.05). 2.1.62 Field 2.062 ICPO/GS (ICP) This field consists of a single binary digit: "0" if the request has not been sent to the Interpol General Secretariat, or "1" if it has been sent. 2.1.63 Field 2.063 Additional Information (INF) This field, consisting of 32 alpha-numeric characters, gives a contact point (e.g. name, phone number) for further information about the request. Fields 2.064 to 2.066 contain information regarding the response to an inquiry, and their presence is highly dependent upon the type of transaction being undertaken. It should be noted that local legislation or guidelines may prohibit the transmission of images which have not been verified. Thus the procedure adopted may be for the local agency to provide resources to verify respondents before initiating a response to the remote agency, and to return only confirmed matches. 2.1.64 Field 2.064 Respondents List (RLS) This field contains at least two subfields. The first subfield describes the type of search that has been carried out, using the three-letter mnemonics which specify the transaction type in TOT (Field 1.04). The contents of the remaining subfields depend on the type of transaction.. In some implementations legal constraints or local guidelines will mandate that the field is restricted to the number of verified respondents only. If the remote AFR system does not assign scores, then a score of zero should be used at the appropriate point. 2.1.65 Field 2.065 Recipient Countries (COU) When a search is to be carried out against the databases of a number of countries, this field provides confirmation on which have actually been searched. It consists of a number of subfields, each of which consists of two information items. The first contains the two character Interpol Country Code of the country whose database was to be searched. The second is a single binary digit: "0" if the search was not carried out and "1" if the search was carried out. 2.1.66 Field 2.066 Result (RES) This field contains up to 128 alpha-numeric characters of free text, giving the address to which the response to a transaction should be sent, if this is to be done other than electronically using an ANSI/NIST message. Fields 2.067 to 2.071 are flags whose presence prompts an action to be taken regarding some part of the transaction. In all cases they are optional and their use may be limited to certain transactions. 2.1.67 Field 2.067 Alert Flag (ALF) The alert flag is to indicate who should be informed if a match is made involving the latent or print. It contains three information items in the following format:
CC is the Interpol Country Code, two alpha-numeric characters long. The second item, agency, is a free text identification of the agency, up to a maximum of 32 alpha-numeric characters. The third is a 128 alpha-numeric string in which extra information might be added, e.g. a contact person or a telephone number. This Item is not intended to be interpreted by computer. A possible use for this field is to indicate a terrorist or a violent criminal. 2.1.68 Field 2.068 Target Criminal Flag (TCF) This indicates whether the subject is considered to be somebody likely to commit an offence and whose fingerprints should always be included in a search irrespective of the defined search scope. The field has one of two possible values:
If the field is not present a default value of 0 is assumed. The main use of the field is in submitting images to an AFR system for a search or to an image database for filing. 2.1.69 Field 2.069 Identified Flag (IDF) This field indicates that a given individual has left latents in various areas which have been successfully matched. The field consists of one or more subfields each of which refers to an identified case. The first three information items in a subfield define the geographical area in which the matching latent was found, in the same format as GAC (Field 2.049). The fourth information item, which is optional, is the case number of the identified latent. 2.1.70 Field 2.070 Latent Priority Flag (MPF) This is the latent equivalent of a TCF (Field 2.068) flag. It indicates that the latent is connected with a particularly serious crime and that it should always be included in any searches of an Unidentified Latents database. The flag takes one of two values:
2.1.71 Field 2.071 Tie Up Flag (TUF) This field is used to indicate that the latent to which it refers has been connected with one or more other latents. Each subfield refers to one of these other latents and consists of up to four information items relating to it: GAC, CNO, SQN, and MID (Fields 2.049, 2.007, 2.008 and 2.009 respectively) or MN1 to MN5 (Fields 2.014 to 2.018). The flag would normally be set, in both enquiry and respondent latents, after a successful latent-to latent search. Fields 2.072 to 2.073 are included only if there is a Type-8 Logical Record, i.e. a signature image. This field consists of up to 16 alpha-numeric characters, and contains a free text description of the rank or grade of the officer providing the signature. 2.1.73 Field 2.073 Date signature (DSG) This field contains the date at which the signature was written, and will probably be supplied automatically by the system. It is in the same ISO format as DAR (Field 2.004). Finally, field 2.074 is included to allow the passing
of status and error messages. Its use is limited to Error Transactions (ERR),
during which it is mandatory. It must not be present in any other transaction. This field contains error messages resulting from image transactions, which will be sent back to the requester as part of an Error Transaction. The maximum length of the field is 128 characters. One example of the use of this field is when the CRO number specified in an image retrieval does not exist. Another is when a requested transaction is not allowed by the receiving system, for instance the search of an asylum seeker's prints against the criminal ten-print database. The field is not intended to be used to transfer information if a search of the database fails to make any matches. This field is mandatory for error transactions.
The INT-I does not allow a record of this format.
It should be noted that Type-4 records are binary rather than ASCII in nature. Therefore each field is assigned a specific position within the record, which implies that all fields are mandatory. The standard allows both image size and resolution to be specified within the record. It requires Type-4 Logical Records to contain fingerprint image data that are being transmitted at a nominal pixel density of 500 to 520 pixels per inch. The preferred rate for new designs is at a pixel density of 500 pixels per inch or 19.68 pixels per mm. 500 pixels per inch is the density specified by the INT-I, except that similar systems may communicate with each other at a non-preferred rate, within the limits of 500 to 520 pixels per inch. For a system to comply with the INT-I it is necessary (although not sufficient) that it can send and receive fingerprints as Type-4 records. 4.1 Fields for Type-4 Logical Record 4.1.1 Field 4.01 Logical Record Length (LEN) This four-byte field contains the length of this Type-4 record, and specifies the total number of bytes including every byte of every field contained in the record. 4.1.2 Field 4.02 Image Designation Character (IDC) This is the one-byte binary representation of the IDC number given in the header file. 4.1.3 Field 4.03 Impression Type (IMP) The impression type is a single-byte field occupying the sixth byte of the record. The permitted codes are:
4.1.4 Field 4.04 Finger Position (FGP) This fixed-length field of 6 bytes occupies the seventh through twelfth byte positions of a Type-4 record. It contains possible finger positions beginning in the left most byte (byte 7 of the record). The known or most probable finger position is taken from the following table. Up to five additional fingers may be referenced by entering the alternate finger positions in the remaining five bytes using the same format. If fewer than five finger position references are to be used the unused bytes are filled with binary 255. To reference all finger positions code 0, for unknown, is used. The finger positions are encoded as:
For scene of crime latents only the codes 0 to 10 should be used. 4.1.5 Field 4.05 Image Scanning Resolution (ISR) This one-byte field occupies the 13th byte of a Type-4 record. If it contains "0" then the image has been sampled at the preferred scanning rate of 19.68 pixels/mm (500 pixels per inch). If it contains "1" then the image has been sampled at an alternative scanning rate as specified in the Type-1 record. 4.1.6 Field 4.06 Horizontal Line Length (HLL) This field is positioned at bytes 14 and 15 within the Type-4 record. It specifies the number of pixels contained in each scan line. The first byte will be the most significant. 4.1.7 Field 4.07 Vertical Line Length (VLL) This field records in bytes 16 and 17 the number of scan lines present in the image. The first byte is the most significant. 4.1.8 Field 4.08 Grey-scale Compression Algorithm (GCA) This one-byte field specifies the grey-scale compression algorithm used to encode the image data. A binary zero indicates that no compression algorithm has been used. In this case pixels are recorded in left to right, top to bottom fashion. The FBI will maintain a registry relating non-zero numbers to compression algorithms. The INT-I will use the same allocation of numbers. This field contains a byte stream representing the image. Its structure will obviously depend on the compression algorithm used.
The INT-I does not allow a record of this format.
The INT-I does not allow a record of this format.
Type-7 records are intended for user-defined image information relating to the subject of a transaction. For Interpol purposes this logical record is likely to fall into two categories, either:
The category to which each record belongs is defined by IMT (Field 7.03). When used for Category-1 data, there may be up to twelve of these Type-7 records in a file: palm, outside edge, inside edge and wrist of each hand plus sole and toes of each foot. These records contain high resolution image data that have been captured at the minimum scanning resolution (500 pixels per inch). They are quantised to eight-bits (ie. 256-level grey scale), a value of zero being used to define a black pixel and an unsigned value of 255 to define a white pixel. When used for Category-2 data there may be up to six of these Type-7 records in a file. The scanning resolution used to capture the data is specified by IMR (Field 7.06). When there are one or more Type-7 logical records, entries are provided in ten ordered and unnumbered mandatory fields. The first nine are fixed in length and total 33 bytes. These nine fields precede the image data contained in field 7.10 7.1 Fields for Type-7 Logical Record 7.1.1 Field 7.01 Logical Record Length (LEN) This four-byte field contains the length of the logical record, specifying the total number of bytes including every byte of all the fields contained in the record. 7.1.2 Field 7.02 Image Designation Character (IDC) The fifth byte contains the one-byte binary representation of the IDC recorded in CNT (Field 1.03). It is used to identify the image data. 7.1.3 Field 7.03 Image Type (IMT) The sixth byte contains a one-byte identifier which specifies whether the image is of Category-1 (palm, sole and toe data) or Category-2 (other) data. The permissable values of this field are:
7.1.4 Field 7.04 Image Description (IMD) This one-byte field occupies the seventh byte position of a Type-7 record. The field contains a code selected from the following table. Table 7.1 Image Description
7.1.5 Field 7.05 Pattern Classification (PCN) When the Type-7 record is used for Category-1 image data, this field, of length ten bytes, contains the pattern classification of the image, in any agreed format. If no classification is to be included in the record, then these ten bytes contain binary zeros. When the Type-7 record is used for Category-2 image data, these ten bytes contain zeros. 7.1.6 Field 7.06 Image Capture Resolution (IMR) This field of length eleven bytes consists of three subfields, starting at byte number 18. When the Type-7 record is used for Category-1 image data, this field contains binary zeros. When the Type-7 record is used for Category-2 image data, the three subfields consist of:
Scanning resolution is specified in units of pixels per 100mm using two bytes, the first being the most significant. For example 7,176 represents (256*7)+176 = 1968, equivalent to 19.68 pixels per mm (500 pixels per inch). 7.1.7 Field 7.07 Horizontal Line Length (HLL) This two-byte field occupies the 29th and 30th byte positions of the Type-7 record. It is used to specify the number of pixels contained in a single line scan of the image. The first byte is the most significant. 7.1.8 Field 7.08 Vertical Line Length (VLL) This two-byte field occupies the 31st and 32nd byte positions of the Type-7 record. It is used to specify the number of scan lines contained in the image. The first byte is the most significant. 7.1.9 Field 7.09 Grey-scale Compression Algorithm (GCA) This one-byte field occupies the 33rd byte of the record. It is used to specify the type of grey-scale compression algorithm used (if any). A binary zero denotes no compression. In this case the scan sequence is left to right and top to bottom. Otherwise, the contents of this field is a binary representation of the number allocated to the particular compression technique used by the interchange parties. The FBI will maintain a registry relating these numbers to the compression algorithms. Interpol will use the same allocation of numbers. This field contains a byte stream representing all of the greys-cale image data. It commences at the 34th byte of the record.
The Type-8 record is used to transmit a signature in the record. This signature may be either of the officer taking the fingerprints or of the fingerprinted subject. The signature is represented in uncompressed binary, compressed binary or vectorized format. Like the other records containing image information the structure is binary, which implies that record fields take a fixed position and all fields are mandatory. The following explains the meaning of the individual fields. 8.1 Fields for Type-8 Logical Record 8.1.1 Field 8.01 Logical Record Length (LEN) The first four bytes of the Type-8 record contain the length of the record expressed as the total number of bytes, including every byte of all eight fields of the record. 8.1.2 Field 8.02 Image Designation Character (IDC) The fifth byte of the record contains a binary representation of the IDC recorded in CNT (Field 1.03). 8.1.3 Field 8.03 Signature Type (SIG) The sixth byte contains the signature type field. The permissable values of this field are:
8.1.4 Field 8.04 Signature Representation Type (SRT) This field indicates how the signature is stored, and is located at the seventh byte of the record. The permissable values of this field are:
8.1.5 Field 8.05 Image Scanning Resolution (ISR) This field gives the image scanning resolution in pixels per mm. One byte is required, in the eighth position of the Type-8 record. The format is a binary zero if the minimum scanning resolution is used and a one if the native scanning resolution is used. A zero shall also be recorded if the image is in vector format. 8.1.6 Field 8.06 Horizontal Line Length (HLL) This field occupies the ninth and 10th byte of the Type-8 record. For binary images it specifies the number of pixels per scan line in the image. For vectorized signature data both bytes contain the value "0000 0000". The first byte is the most significant. 8.1.7 Field 8.07 Vertical Line Length (VLL) This two-byte field indicates the number of scan lines present in a binary image and is positioned at bytes 11 and 12 within the record. As with HLL (Field 8.06) it contains zeros if the signature is in vector representation. The first byte is the most significant. 8.1.8 Field 8.08 Signature Data This field contains the image data in uncompressed binary, compressed binary or vectorized form according to the entry in SRT (Field 8.04). Binary images are compressed according to the ANSI/EIA-538-1988 facsimile compression algorithm (FAX Group 4 standard). Vectorized image data specify a list of vectors describing the pen position and pen pressure of line segments within the signature. Each vector is five bytes in length and contains the unsigned binary X position (two bytes, most significant first), the unsigned binary Y position (two bytes, most significant first) and the pen pressure (1 byte). A pressure value of "0000 0000" indicates the end of a line (i.e. pen up), while "0000 0001" to "1111 1110" indicate a range of pressures from the least recordable up to the maximum recordable pressure for the input device. The end of the vector list is indicated by a value of "1111 1111". The origin of the image is the bottom left hand corner and X,Y positions are expressed units of 0.01mm.
Type-9 records shall contain ASCII text describing minutiæ and related information encoded from a finger or palm. For a tenprint search transaction, there may be up to ten of these Type-9 records in a file, each of which shall be for a different finger. There may be up to four of these records for palm print searches. The Type-9 record shall also be used to exchange the minutiæ information from latent finger or palm images between similar or different systems. 9.1 Minutiæ and Other Information Descriptors 9.1.1 Minutia Type Identification This standard defines four identifier characters that are used to describe the minutia type. These are listed in Table 9.1. A ridge ending shall be designated Type A. It occurs at the point on a fingerprint or palm print that a friction ridge begins or ends without splitting into two or more continuing ridges. The ridge must be longer than it is wide. A bifurcation shall be designated Type B. It occurs at the point that a ridge divides or splits to form two ridges that continue past the point of division for a distance that is at least equal to the spacing between adjacent ridges at the point of bifurcation. A minutia shall be designated Type C, a compound type, if it is either a trifurcation (a single ridge that splits into three ridges) or a crossover (two ridges that intersect). If a minutia cannot be clearly categorized as one of the above three types, it shall be designated as undetermined, Type D.
Each minutia shall be identified by an index number that is assigned to it. The numbering shall begin at «1» and be incremented by «1» for as many times as there are minutiæ encountered. This allows each minutia to be uniquely identified. Ridge counts may be made from each minutia in a fingerprint or palm print to certain other neighboring minutiæ. When this occurs, ridge counts between designated minutiæ shall be associated with the applicable index numbers so as to ensure maintenance of the proper relationships. Rules for identifying neighboring minutiæ an | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
