# Palm Recognition SDK
### Code
{% embed url="" %}
### Overview
**World's First Palm Recognition SDK for Any RGB Camera – On-Premise, Real-Time, and High Accuracy**
Introducing the world’s first **Palm Recognition SDK** that works seamlessly with images captured by any standard **RGB camera**. Designed for developers, this powerful SDK enables easy integration of **real-time, high-accuracy palm recognition** into your applications using advanced **deep learning models**.
Unlike cloud-based solutions, our SDK is fully **on-premise**, ensuring that **all data is processed locally on your server** — delivering maximum **data privacy and security**. No internet connection is required, and **no biometric data ever leaves your machine**.
Perfect for secure authentication, access control, and enterprise-grade biometric solutions.
### 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
```
2. Activate the environment
```
conda activate palm
```
3. Install dependencies
```
pip install torch torchvision torchaudio
pip install opencv-python
pip install tqdm
pip install scikit-image
pip install mediapipe
```
4. 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.
---
# Agent Instructions: Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://doc.faceplugin.com/palm-recognition-sdk/open-source-palm-recognition-sdk.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.