thermocam tools

This set of tools serves for post processing thermal imaging raw data. Currently supported are:

• calibrating and converting an entire set (one measurement campaign) of raw data files to temperature files
• exploring and plotting temperature files
• post processing using gmic

File formats and filename convention

raw data

Raw data is currently expected to be in DIY-Thermocam *.DAT format. Only the raw image data is actually read and all metadata discarded. It should be easy to support other raw data formats in the future.

temperature data

Temperature data is currently stored as 32-bit float values in native byte order without any header or meta data. Surely there exists already an established file format for this kind of data - please tell me if you know something about this! Temperature data files use the filename extension .raw, if they are just converted and calibrated data without further enhancements like noise removal.

filenames

are expected to be in the format YYYYMMDDhhmmss_object.ext where _object is optional and describes the primary object of the scene. This is mostly used to automatically detect images of reference bodies for calibration.

Tools available

Programs with the .py filename extension are intended to be used from the shell, but are also loadable as module from other python programs or the python interpreter.

dat2temp.py - convert and calibrate

Call it with --help to get an overview of options available. However since the program currently only performs offset calibration, you need to at least specify the --slope parameter.

datinfo.py - show min/max raw values of each file

This is meant to provide the info necessary to manually tweak offset/slope calibration. Currently it only works on raw values without any filtering or processing.

tempplot.py - explore and plot a temperature files

Plots the specified temperature files with matplotlib.pyplot for interactive exploration. You can save the plot as *.png file from the plot window. If you specify more then one file, all will be shown in the same plot with one common color map.

Command line options:

• --cmap: specify a color map (whatever matplotlib supports)
• --columns: the number of files to show in one row
• --title: a plot title.

tempviewer - explore and post process temperature files with gmic

Any command line argument is expected to be a temperature file. All temperature files are shown as thumbnails. Selecting (clicking) one allows to explore the temperature values interactively.

After closing the window you are prompted for gmic commands to operate on the set of open images. This allows for further post processing like noise filtering or saving to different image formats.

To quit the interpreter loop, press CTRL+D (end of file).

Suggested workflow

I have the thermocam configured to show the current time on the display. Whenever I take an image of a reference body (usually wet ice), I note down the time after that on paper. After moving the images from the thermocam to my computer, I use the list on paper to locate all images of a reference body in the file manager and rename them from YYYYMMDDhhmmss.DAT to YYYYMMDDhhmmss_ice.DAT. Then I run

dat2temp.py -s 0.047 --ref ice -t 0.0 *.DAT