Package ai.tech5.sdk.abis.finger

Class T5FingerLTMatcher

java.lang.Object

ai.tech5.sdk.abis.finger.T5FingerLTMatcher

public class T5FingerLTMatcher
extends java.lang.Object

Field Summary

Fields

Modifier and Type	Field	Description
static int	FAST_IDENTIFY	Templates for both stages of matching (accelerator matching and precise matching) keeps in memory and identification performs in one step.
static int	SLOW_IDENTIFY	Only templates for 1st stage of matching (accelerator matching) keeps in memory templates for precise matching load in memory by request based on result of 1st stage of matching and identification performs in two steps.

Constructor Summary

Constructors

Constructor	Description
T5FingerLTMatcher()	create object of Matcher.

Method Summary

Modifier and Type	Method	Description
void	cancel()	delete object of T5FingerLTMatcher
boolean	create(int identify_mode, boolean check_doubles)	create object of T5FingerLTMatcher
int	delete_record(java.lang.String id)	Function delete record from memory database Parameters:
int	get_max_template_size()
long	getFinalGallerySize()	Function return current gallery size
java.lang.String	getVersion()	The function returns version of Tech5_GalleryMatcher_ServerE Parameters:
int	identify_LT(ai.tech5.sdk.abis.finger.MatchingParameters matchingParameters, byte[] latent_template, byte[] fingerMask, int numberOfThreads, int candidate_list_length, java.util.ArrayList<ai.tech5.sdk.abis.finger.CandidatesString> candidateList)	Function perform identification Parameters:
boolean	initLicense(java.lang.String licensePath)	Function sets license
int	insert_record(java.lang.String id, ai.tech5.sdk.abis.finger.TpTemplate tpTemplate)	Function insert record in memory database Parameters:
boolean	isRecordExist(java.lang.String id)	Function checks record state
long	verify_latent(ai.tech5.sdk.abis.finger.MatchingParameters matchingParameters, byte[] latent_template, byte[] finger_template, java.lang.Double score, java.lang.Double logFAR)	Function verify latent template with fingerprint template

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

FAST_IDENTIFY

public static int FAST_IDENTIFY

Templates for both stages of matching (accelerator matching and precise matching) keeps in memory and identification performs in one step. It's more quick way, but required more RAM

SLOW_IDENTIFY

public static int SLOW_IDENTIFY

Only templates for 1st stage of matching (accelerator matching) keeps in memory templates for precise matching load in memory by request based on result of 1st stage of matching and identification performs in two steps. It's slower, but required less RAM

Constructor Detail

T5FingerLTMatcher

public T5FingerLTMatcher()

create object of Matcher. Return NULL in case of error

Method Detail

getVersion

public java.lang.String getVersion()

The function returns version of Tech5_GalleryMatcher_ServerE Parameters:

Returns:

value of current version

create

public boolean create(int identify_mode,
                      boolean check_doubles)

create object of T5FingerLTMatcher

Parameters:

identify_mode - (input) - identification mode ( @see IDENTIFY constants)

check_doubles - (input) - if true, check if record with such id already in exist will be performed while insert

Returns:

Return false in case of error

cancel

public void cancel()

delete object of T5FingerLTMatcher

get_max_template_size

public int get_max_template_size()

Returns:

Function returns maximum size for template

initLicense

public boolean initLicense(java.lang.String licensePath)

Function sets license

Parameters:

licensePath - (input) - path to the license file

Returns:

true if success and false otherwise

verify_latent

public long verify_latent(ai.tech5.sdk.abis.finger.MatchingParameters matchingParameters,
                          byte[] latent_template,
                          byte[] finger_template,
                          java.lang.Double score,
                          java.lang.Double logFAR)

Function verify latent template with fingerprint template

Parameters:

matchingParameters - ( input) - matching parameters

latent_template - ( input) - latent template

finger_template - ( input) - fingerprint template

score - (output) - similarity of two records in a range 0...1

logFAR - (output) - log10(FAR) corresponded to getting score

Returns:

SUCCESS if success and error code otherwise

insert_record

public int insert_record(java.lang.String id,
                         ai.tech5.sdk.abis.finger.TpTemplate tpTemplate)

Function insert record in memory database Parameters:

Parameters:

id - ( input) - unique record ID. NOTE: id - should be numerical only!

tpTemplate - ( input) - 20-finger template record. It can have from 1 till 20 fingerprint templates.

Returns:

Returns SUCCESS if success and error code otherwise

delete_record

public int delete_record(java.lang.String id)

Function delete record from memory database Parameters:

Parameters:

id - ( input) - unique record ID. NOTE: id - should be numerical only!

Returns:

Returns number of removed records

getFinalGallerySize

public long getFinalGallerySize()

Function return current gallery size

Returns:

current gallery size

isRecordExist

public boolean isRecordExist(java.lang.String id)

Function checks record state

Parameters:

id - ( input) - unique record ID. NOTE: id - should be numerical only!

Returns:

true if record with this id already exist in gallery and false otherwise

identify_LT

public int identify_LT(ai.tech5.sdk.abis.finger.MatchingParameters matchingParameters,
                       byte[] latent_template,
                       byte[] fingerMask,
                       int numberOfThreads,
                       int candidate_list_length,
                       java.util.ArrayList<ai.tech5.sdk.abis.finger.CandidatesString> candidateList)

Function perform identification Parameters:

Parameters:

matchingParameters - ( input) - matching parameters

latent_template - ( input) - latent template

fingerMask - ( input) - define what fingers will be used in matching

numberOfThreads - ( input) - number of threads that will be used in matching. 0 - means use all logical CPU (the value returned by omp_get_num_procs())

candidate_list_length - ( input) - size of output candidate list

candidateList - (output) - candidate list of matching if success and zero otherwise

Returns:

Returns SUCCESS if success and error code otherwise