Open Source Palm Recognition SDK

Completely Free and Open Source Palm recognition Windows SDK

Code

GitHub - Faceplugin-ltd/Palm-Recognition: The worlds's 1st free and open source palm recognition SDK for Windows and Linux (Palm detection, ROI extraction, Template extraction, Template mathcing)GitHub

Overview

The world's 1st Completely Free and Open Source Palm Recognition SDK for developers to integrate palm recognition capabilities into applications. Supports real-time, high-accuracy palm recognition with deep learning models. This is on-premise palm recognition SDK which means everything is processed in your server and NO data leaves the machine. Please contact us if you need the SDK with higher accuracy.

Setup

Please download anaconda on your computer and install it. We used Windows machine without GPU for testing

1. Create anaconda environment

conda create -n palm python=3.9

1. Activate the environment

conda activate palm

1. Install dependencies

pip install torch torchvision torchaudio
pip install opencv-python
pip install tqdm
pip install scikit-image
pip install mediapipe

1. Compare two palm images in the test_images directory

python main.py

APIs and Parameters

• classify_hand(mp_hands, hand_landmarks, image_width): determine if the hand is left hand or right hand

• extract_roi(hands, mp_hands, img_path): extract region of interest from the palm image for template matching

• extract_features(mp_hands, hands, path: str): extract template from the plam image specified by the path parameter

• compare_two_images(mp_hands, hands, image_path1, image_path2, similarity_threshold=0.8): compare two hand images to determine if they are the same hand or not.