NBioBSP for Python

Beta module for integrating NBioBSP SDK with Python

This is a Python module for NBioBSP SDK. Built with C/C++ and pybind11. Works only on Windows at the moment.

All code is self contained into a single file, as it is just bindings.

- Documento em Português

Requirements

• Python 3.10 or higher
• eNBSP SDK greater or equal to 5.2
• MSVC 2019 or higher (maybe it works with another compiler, didn't test it yet)
• Cmake 20 or higher

To Do:

• Readable error output using NBioAPI types (priority)
• Handle errors properly
  • Throw translatable and proper exceptions from C++ to Python
• Handle return values properly (priority)
• Handle fingerprint purposes on capture properly
• Implement IndexSearch methods
• Implement all functions that are possible in Windows
• Implement all functions that are possible in Linux (no priority)

You can test the module with the following code on the Python interpreter:

import nbsp_python as nbsp

# Initialize the NBioBSP instance
nbsp.initialize_handle()

# Get the version of the NBioBSP SDK
nbsp.get_version()

# Enumerate the devices connected to the computer
nbsp.enumerate_device()

# Open the device using auto-detection
nbsp.open_device()

# Capture a fingerprint and assign it to handle
handle = nbsp.capture("enroll", 10000)

# Verify fingerprints contained in handle
nbsp.verify(handle)

# Extract the hash from the last fingerprint added in the handle
fingerprint_hash = nbsp.extract_fir_text(handle)

# Initialize the index search
nbsp.init_index_search()

# Add the fingerprint to the index search and pass
nbsp.add_fir_to_index_search(fingerprint_hash, 1)

# Identify the fingerprint in the index search
nbsp.identify_data_from_index_search(fingerprint_hash)

# Terminate the index search
nbsp.terminate_index_search()

# Terminate the NBioBSP instance
nbsp.terminate_handle()

# Close the device
nbsp.close_device()

Or by executing a TKinter Example with Python.

Installation

git clone https://github.com/Andoreatta/nbsp-python.git
cd nbsp-python
pip install .

Pulls all the dependencies and installs the module, if something happens, check if the dependencies in pyproject.toml are installed correctly.

Usage

• nbsp.initialize_handle() Initializes the NBioBSP instance. Must be called before any other function that interacts with the device. So you can call get_version() before initialize_handle(), but not capture(). Returns a handle just in case.

• nbsp.terminate_handle() Terminates the NBioBSP instance.

• nbsp.get_version() Returns the version of the NBioBSP SDK.

• nbsp.enumerate_device() Enumerates the devices connected to the computer. Returns a list of devices.

• nbsp.open_device() Opens the device using auto-detection. Returns the handle of the device.

• nbsp.open_specific_device(index) Opens the device using the device ID. Returns true if successful.

• nbsp.close_device() Closes the device.

• nbsp.capture("purpose", timeout) Captures a fingerprint. Returns the handle of the fingerprint. purpose can be one of the following:

  • "enroll" for registering a fingerprint
  • "verify" for verifying a fingerprint
  • "identify" for identifying a fingerprint (deprecated as of 5.2)
  • "enroll_for_verification" for registering a fingerprint for verification
  • "enroll_for_identification" for registering a fingerprint for identification The timeout parameter is the maximum time in milliseconds to wait for a fingerprint to be captured.

• nbsp.verify(handle) Verifies the last fingerprint added in handle. Returns true if successful.

• nbsp.extract_fir_text(handle) Extracts the fingerprint hash from the handle. Returns the fingerprint hash.

• nbsp.init_index_search() Initializes the index search.

• nbsp.terminate_index_search() Terminates the index search.

• nbsp.add_fir_to_index_search(fingerprint_hash, user_id) Adds a fingerprint to the index search. Returns true if successful.