https://github.com/photron-limited/pypuclib
python package for INFINICAM SDK
https://github.com/photron-limited/pypuclib
camera computer-vision machine-vision opencv python
Last synced: about 1 month ago
JSON representation
python package for INFINICAM SDK
- Host: GitHub
- URL: https://github.com/photron-limited/pypuclib
- Owner: infinicam
- License: mit
- Created: 2021-04-29T23:42:16.000Z (about 5 years ago)
- Default Branch: main
- Last Pushed: 2025-03-26T07:05:18.000Z (about 1 year ago)
- Last Synced: 2025-03-26T08:22:11.873Z (about 1 year ago)
- Topics: camera, computer-vision, machine-vision, opencv, python
- Language: C++
- Homepage: https://www.photron.co.jp/
- Size: 32.4 MB
- Stars: 9
- Watchers: 3
- Forks: 0
- Open Issues: 7
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# pypuclib
pypuclib is python package for INFINICAM SDK [PUCLIB](https://www.photron.co.jp/products/hsvcam/infinicam/tech.html).
This package allows user to control [INFINICAM UC-1](https://www.photron.co.jp/products/hsvcam/infinicam/) on python.
https://user-images.githubusercontent.com/71418024/171789047-f61ddb07-27a4-4213-bff1-26cc0676d0b6.mp4
*This is [tracking_sample](pypuclib/pypuclib_sample/tracking_sample.py) demo movie.
## How to Use
1. Install INFINICAM SDK [PUCLIB](https://www.photron.co.jp/products/hsvcam/infinicam/tech.html), driver, and Visual C++ Redistributable 2019.
2. See [Install](#Install) section to install pypuclib package.
3. See [Quick Start](#Quick-start-for-image-processing) section to start programming.
## Install
You can install it in any of the following ways:
* Open command prompt and install using pip from [pypi](https://pypi.org/).
```
pip install pypuclib
```
If you don't register environment variable for python path, use this command instead.
```
py -m pip install pypuclib
```
* Install from source code.
```
git clone repositryURL
```
Move current directory to pypuclib/pypuclib/ and install using this command.
```
pip install ./
```
## Quick Start for Image Processing
To connect the first detected camera:
```python
cam = CameraFactory().create()
```
And grab the image:
```python
xferdata = cam.grab()
```
To decode image, get decoder object from camera and use it like:
```python
decoder = cam.decoder()
img = decoder.decode(xferdata)
```
Decoded array is numpy array, so now you can use to imageprocessing package like opencv directory.
If you want to use only the resion of interest:
```python
x = 0, y = 0, w = 128, h = 128
img = decoder.decode(xferdata, x, y, w, h)
```
This can reduce the decoding time.
## For High Speed Processing
Use beginXfer and endXfer to get callback from C++.
```python
# xfercallback from c++
def callback(data) :
seq = data.sequenceNo()
if oldSeq != seq:
# avoid same sequence number.
# beginn xfer and start callback background
cam.beginXfer(callback)
~~ # some processing here
cam.endXfer()
```
## How to Run Samples
1. Install pypuclib using pip.
2. Download following sample source code.
3. Run on python.
```python
python hello_world.py
```
* [hello_world](pypuclib/pypuclib_sample/hello_world.py)
Sample of streaming a live image and saving a bitmap.
If a GPU device is available, decoding is done on the GPU. Otherwise, decoding is done on the CPU.
* [create_movie](pypuclib/pypuclib_sample/create_movie.py)
Sample of saving videos with high speed streaming.
* [tracking_sample](pypuclib/pypuclib_sample/tracking_sample.py)
Sample of real-time tracking at 1,000 fps.
* [gui_sample](pypuclib/pypuclib_sample/gui_sample.py)
GUI sample for controlling camera parameters and recording.
* [find_1000](pypuclib/pypuclib_sample/find1000/README.md)
This sample code implements the a deep learning network to detect objects such as people, cellphones etc.
## Source Code Structure
This repository consists of three projects.
* [pypuclib](pypuclib/pypuclib)
Source code of python wrapper of PUCLIB
* [pupuclib_sample](pypuclib/pypuclib_sample)
Python sample code using pypuclib
* [pypuclib_test](pypuclib/pypuclib_test)
Unit test of pypuclib
## Prerequisites
To build pypuclib, the following environment is required.
* Installed Visual Studio 2019
:warning: The MSVC compiler is required to build the package.
:warning: This package is available for Windows only.
## Issues
Don't hesitate to ask any questions or issues on GitHub.
#### developped by:
