pysenxor

A Python Library for Meridian Innovation's thermal imaging devices.

• Github repository: https://github.com/MeridianInnovation/pysenxor-lite/
• Documentation https://MeridianInnovation.github.io/pysenxor-lite/

Overview

This package can let users interact with Meridian Innovation's thermal imaging devices.

Features

• Device discovery and listing
• Multiple interfaces supported(USB serial, TCP/IP, GPIO, etc.)(Coming soon)
• Easy device connection and management
• Configuration and status read and write
• Non-blocking mode for frame reading
• Thermal data processing utilities
• Thread-safe for multi-threaded use
• Lightweight and minimal dependencies(no cv2 or matplotlib required)

Installation

We strongly recommend using uv to manage the virtual environment.

To install uv, please refer to the official documentation.

On Linux, if you plan to use pysenxor-lite together with opencv, qt, tkinter, etc., uv can help avoid many system dependency issues.

1. Recommended: use uv

Create a virtual environment and install the package with uv:

uv init   # create a pyproject.toml file, if you already have one, skip this step
uv add pysenxor-lite # Add the package to the virtual environment

or without pyproject.toml:

uv venv --seed   # Create a virtual environment and install the dependencies
uv pip install pysenxor-lite # Install the package

2. Use pip

python -m pip install pysenxor-lite

3. Linux notes

If you are using SenXor via USB serial on Linux, ensure you have the proper permissions to access the serial port. You can add your user to the dialout group and reboot to apply the changes:

sudo usermod -aG dialout $USER

4. Development installation

For development, clone the repository and sync the environment with uv:

git clone https://github.com/MeridianInnovation/pysenxor-lite.git
cd pysenxor-lite
uv sync

This will create a virtual environment and install pysenxor-lite in editable mode.

If you prefer pip, you can use:

git clone https://github.com/MeridianInnovation/pysenxor-lite.git
cd pysenxor-lite
python -m pip install -e .

Usage

This section gives a short overview of how to use the pysenxor library. For more detailed examples, see the documentation.

Connect to a device

You can list available devices and connect to the first one:

from senxor import connect, list_senxor

devices = list_senxor("serial")
if not devices:
    raise ValueError("No devices found")

dev = connect(devices[0])

To check the connection and device info:

print(f"Connected to {dev.name}, is_streaming: {dev.is_streaming}")
print(f"Module: {dev.get_module_type()}, FW: {dev.get_fw_version()}, SN: {dev.get_sn()}")

When you are done, close the device:

dev.close()

Stream and read frames

After connecting, you can start the stream and read frames. read() returns (header, frame): header is a uint16 array or None, and frame is a 2D array of temperature in Celsius (float32) by default.

dev.start_stream()

header, frame = dev.read()
print(f"Frame shape: {frame.shape}, dtype: {frame.dtype}")
print(f"Min: {frame.min()} C, max: {frame.max()} C, mean: {frame.mean():.1f} C")

By default read() blocks until a frame is available. If you pass block=False, it returns immediately and header and frame may be None when no frame is ready:

header, frame = dev.read(block=False)
if frame is None:
    print("No new frame")

To stop streaming but keep the connection open, call stop_stream(). To stop and disconnect, use close().

Process and visualize data

Frames are NumPy arrays, so you can index, slice, and use methods like .min(), .max(), and .mean(). The senxor.proc module provides normalization, scaling, and colormaps (without requiring cv2 or matplotlib):

import numpy as np
from senxor.proc import normalize, enlarge, colormaps, apply_colormap

uint8_image = normalize(frame, dtype=np.uint8)
float32_image = normalize(frame, dtype=np.float32)
enlarged = enlarge(frame, scale=2)

To get an RGB image, you can apply a built-in colormap such as inferno, jet, viridis, magma, plasma, or turbo:

cmap = colormaps["inferno"]
normalized = normalize(frame, dtype=np.float32)
colored_image = apply_colormap(normalized, lut=cmap)

You can display or save the result with cv2 or matplotlib. Note that cv2 uses BGR order:

import cv2
bgr = cv2.cvtColor(colored_image, cv2.COLOR_RGB2BGR)
cv2.imshow("senxor", bgr)
cv2.waitKey(0)
cv2.destroyAllWindows()

import matplotlib.pyplot as plt
plt.imshow(colored_image)
plt.show()

Learn more

• Documentation

License

This project is licensed under the Apache License 2.0.

You may freely use, modify, and distribute this software for both open-source and commercial purposes, subject to the terms of the license.

Copyright

Unless otherwise specified, all files in the source code directory(senxor/) are copyrighted by Meridian Innovation.

Copyright (c) 2025 Meridian Innovation. All rights reserved.

Contributing

We welcome contributions from the community.

By submitting a pull request, you certify compliance with the Developer Certificate of Origin (DCO). This means you assert that:

• You wrote the code or have the right to submit it;
• You grant us the right to use your contribution under the project license.

Please add the following line to your Git commit message to confirm DCO compliance:

Signed-off-by: Your Name your.email@example.com

You can automate this with git commit -s.

See more details in Contributing Guide.

Contact

For support or inquiries, please contact:

• Email: info@meridianinno.com