flashcam

Project flashcam

FLASk powered Hybrid webCAM

IN DEVELOPMENT

Why

Because any idea can be implemented in the code without a dependence on a closed source. And there are few ideas implemented by now.

Start

Production web interface:

export PATH=$PATH:$HOME/.local/bin
export DISPLAY=:0
killall -9 gunicorn
$( sleep 5;flashcam ask;)&
flashcamg

Just run on local network, force a port 8000

export PATH=$PATH:$HOME/.local/bin
flashcam flask -n 8000 # -p  00.0-usb-0:1.3:1.0

Select a camera

If there are two or more connected, try flashcam ls and select a proper IDpath like flashcam flask -p 00.0-usb-0:1.3:1.0.

It is possible to run with two cameras, but with local network mode and pay attention to MJPG pixelformat and using different ports.

When config is already fine-tuned for something else

Deactivate blur, accumulate, threshold, telegram, set basic resolution, another port:

flashcam flask -p clock -a 10 -b 3  -r 640x480 -n 5000 -j False -t 0 -a 0 -b 1

Config

Located at $HOME/.config.flashcam/cfg.conf it allows to setup basically everything.

A lot of options is possible to set and override from commandline.

Basic scenarios

Just use

Probably the option 'PIXELFORMAT':'MJPG' to read physical camera as mjpg is fine.

High quality

When trying e.g. for astrophotography, you may prefer 'PIXELFORMAT':'YUYV' to read camera without artifacts. While it works for 640x480, this may not be possible for higher resolution.

Motion detection

Technical details

• web.py module is the one that is called by gunicorn and provides the interface. See frames(), where Camera() class (defined in realcamera.py) is called first.

Usage

See README.howto.org

Some references that may or may not be related

https://longervision.github.io/2017/03/12/ComputerVision/OpenCV/opencv-external-posture-estimation-ArUco-board/

https://github.com/LongerVision/Examples_OpenCV/blob/master/01_internal_camera_calibration/chessboard.py

https://markhedleyjones.com/projects/calibration-checkerboard-collection