Skip to content

Latest commit

 

History

History
78 lines (43 loc) · 3.82 KB

TUTORIAL.md

File metadata and controls

78 lines (43 loc) · 3.82 KB
Raw

time_stacker Tutorial

Overview

time_stacker creates a time-stack image show from a video or animated gif. This type of picture is best used to showcase the passage of time where the background is static and the focal object moves in a non-overlapping trajectory (e.g. ball bouncing down a hallway).

Setting up

After following the installation instructions on README.md, navigate to the time\_stacker working directory

cd <path_to_time_stacker>

Starting from a video

time\_stacker can read animated gifs and most video formats and output most image file formats.

Basic functionality: Running the script without extra arguments.

python ./script/time_stacker.py --input_file ./example/drosophila.mov --output_file ./output/drosophila_basic.png

Adding an interval: Space out the images selected to allow for better resolution between objects.

python ./script/time_stacker.py --input_file ./example/drosophila.mov --output_file ./output/drosophila_interval_10.png --interval 10

Interval with grayscale: Same image as above, but in grayscale.

python ./script/time_stacker.py --input_file ./example/drosophila.mov --output_file ./output/drosophila_interval_10_grayscale.png --interval 10 --grayscale

Options for processing video

[Required]

1. `--input\_file` - Path to movie (most formats) or animated gif file
2. `--output\_file` - Path to output file, including desired file type as the file suffix

[Optional]

3. `--method` - Method for flattening frames into a single frame. `min` is better for darker objects on a lighter background (default). `max` is better for lighter objects on a darker background
4. `--interval` - Use every <n-th/interval> frame for final image (default is 1). Higher interval numbers decrease processing time.
5. `--grayscale` - Option to output image in grayscale. This option significantly increases processing time
6. `--dpi` - Resolution to save final image in drops per inch (dpi)

--> `--help` - Provides help text in the command line to aid in what inputs are permissible

Starting from a stack of images

time\_stacker requires a video or animated gif, rather than a series of images. If you are beginning with a series of images, you can convert them to a video using FFmpeg. Depending on how the images are named, the following command will be different.

Assuming a series of images with a common prefix (img_), images named with three digits (000,001...999), and a common suffix (.png):

ffmpeg -i <path_to_image_folder>/img_%03d.png ./example/output.mov

A video file (output.mov) will be created in time_stacker's example folder. The video's frame rate is not important, though there may be discrepancies in the video quality depending on the end type of video you try converting. FFmpeg is well documented and can be used to create videos and animated gifs (among other things) with relative ease.

Troubleshooting and errors

  • Following the basic example, an error like the following: TypeError: Invalid shape (8, 960, 1280, 3) for image data may be caused by all packages being installed, but not having the conda virtual environment activated. Type: conda activate python36 or source activate python36 and re-run the command.