Skip to content

Latest commit

 

History

History
134 lines (103 loc) · 4.51 KB

readme.md

File metadata and controls

134 lines (103 loc) · 4.51 KB

Python driver for fingerprint sensors by Zhiantec - ZFM-20 series

PyPI version PDD status

Is your scanner can be managed by this library?

If your scanner is ZFM-20 or his cheap clone then it possible. I made this library inspired by pyfingerprint, so it may also work with ZFM-60, ZFM-70, ZFM-100, R303 and R305.

Origin manuals

Fingerprint protocol specs are taken from original ZHM-20 user manual.

Terminology

Image

Image is a Fingerprint scanned grayscale image. Image can be scanned and stored in volatile image buffer.

Characteristic

Characteristic is a fingerprint characteristic. It represented by a bytearray. Characteristic can be created from fingerprint image and stored in volatile characteristic buffer.

Template

Template is a registered fingerprint model stored in scanner nonvolatile memory. We can not see or feel the template, we only can ask scanner if characteristic look like some template in scanner memory.

Working with sensor

Handshake

First of all you need make a handshake to verify connection:

with SerialPort(Serial(port='<COM1 or /dev/ttyUSB0>', baudrate=9600 * 6, timeout=2)) as port:
   rq = RqCommand(port)
   rs = RsSimple(port)
   Handshake(rq, rs).make()

As you see you will need to know serial port name of your scanner device. For windows users it may looks like COM1, for unix users it may looks like /dev/ttyUSB0.

Fingerprint image

More complex task - make an image of your fingerprint:

with SerialPort(Serial(port='...', baudrate=9600 * 6, timeout=2)) as port:
    rq = RqCommand(port)
    rs = RsSimple(port)
    print 'Waiting for finger...'
    while not Scan(rq, rs).is_scanned():
        pass
    print 'Finger has been scanned! Downloading the finger image...'
    image = UpImage(rq, rs).image()
    image.show()

Matching characteristics

Another more complex task - match characteristics of two fingerprints. Fingerprint scanner can matching only two fingerprints and it has two buffers for that operation - RqCharBuffer1 and RqCharBuffer2.

with SerialPort(Serial(port='...', baudrate=9600 * 6, timeout=2)) as port:
    rq = RqCommand(port)
    rs = RsSimple(port)
    print 'Wait for finger...'
    while not Scan(rq, rs).is_scanned():
        pass
    Img2Tz(rq, rs, RqCharBuffer1()).execute()

    print 'Once again...'
    while not Scan(rq, rs).is_scanned():
        pass
    Img2Tz(rq, rs, RqCharBuffer2()).execute()

    print 'Score {0}'.format(Match(rq, rs).score())

Enroll fingerprint

Another complex task is enroll a finger.

with SerialPort(Serial(port='...', baudrate=9600 * 6, timeout=2)) as port:
    rq = RqCommand(port)
    rs = RsSimple(port)
    print 'Wait for finger...'
    while not Scan(rq, rs).is_scanned():
        pass

    Img2Tz(rq, rs, RqCharBuffer1()).execute()
    searchResult = Search(rq, rs, start=0, count=TemplateCount(rq, rs).as_int()).execute()

    if searchResult.code() == 0:
        print 'Template already exist'
        exit(1)

    print 'Once again...'
    while not Scan(rq, rs).is_scanned():
        pass

    Img2Tz(rq, rs, RqCharBuffer2()).execute()
    score = Match(rq, rs).score()
    RegModel(rq, rs).execute()
    Store(rq, rs, RqCharBuffer1(), 1).execute()
    print 'Stored success!'

Match fingerprint with it db saved version

with SerialPort(Serial(port='...', baudrate=9600 * 6, timeout=2)) as port:
    rq = RqCommand(port)
    rs = RsSimple(port)
    while not Scan(rq, rs).is_scanned():
        pass
    Img2Tz(rq, rs, RqCharBuffer1()).execute()  # save scanned finger to the buffer 1
    
    templatesCount = TemplateCount(rq, rs).as_int()  # get all templates count
    searchResult = Search(rq, rs, start=0, count=templatesCount).execute()  # find finger in the db
    
    if searchResult.code() != 0:
        print('Finger has not been saved!')
    else:
        number = searchResult.number()  # finger index in the db
        print('Finger found in index {0}'.format(number))
        LoadChar(rq, rs, RqCharBuffer2(), number).execute()  # extract original template to the buffer 2
    
        print('Score {0}'.format(Match(rq, rs).score()))  # Match it!

@todo #1:30m Add deletion enrolled fingers