The Open Telemetry Kit (OTK) is an open source package for extracting and parsing telemetry associated with video streams and converting to common formats. It comes out of a need for a singular API that can be used for multiple different video telemetry formats.

Telemetry that can be parsed includes: GPS, time, camera information, speed

Features:

• Automatically detect telemetry format
• Manipulate telemetry with ease
• Write telemetry to a new format

Python version: >=3.6

ffmpeg and ffprobe.

On Debian systems these can be installed with:

$ sudo apt install ffmpeg

dateutil

On Debian systems this can be installed with:

$ pip3 install python-dateutil

$ pip3 install open-telemetry-kit

The OTK package can be imported into your python3 project with:

import open_telemetry_kit as otk

Download the OTK quickstart package (~90 MB). (Mac users can install wget using these instructions)

$ wget https://hivemapper-sample-videos.s3-us-west-2.amazonaws.com/OTK/OTK_quickstart.tgz

This includes a sample .csv, .srt, and .mov with embedded telemetry. It also contains quickstart.py which you can use to extract the telemetry from the sample files.

Extract the package:

$ tar xzvf OTK_quickstart.tgz

This will extract everything into a new directory called OTK_quickstart/

The quickstart.py script accepts a standalone .csv or .srt or a video file with an embedded .srt. It will read in the data, convert it to JSON, and write it to the provided destination.

In your terminal go to the new OTK_quickstart directory.

Extract telemetry from the sample video:

$ python3 quickstart.py embedded_srt_example.mov embedded_srt_example.json

Extract telemetry from the sample srt or csv files respectively:

$ python3 quickstart.py srt_example.srt srt_example.json

$ python3 quickstart.py csv_example.csv csv_example.json

Note: This process will create a new JSON file containing the telemetry extracted from the sample file. The data is organized into an array of objects (or, in python terminology, a list of dictionaries)

For an example of simple data manipulation, open quickstart.py and uncomment the lines:

# gps = Telemetry()
# for packet in telemetry:
#   gps.append({ k:v for k, v in packet.items() if k in ['latitude', 'longitude', 'altitude']})

# write.telemetryToJson(gps, dest)

Rerun the script with one of the provided commands above.

The OTK currently supports the following forms of telemetry:

• .csv files
• .srt files
• Most videos file with embedded telemetry encoded as a .srt (e.g. video taken with some DJI drone models)
• .gpx files
• .kml files
• KLV/MISB embedded data
• Open Camera .srt

• JSON
• CSV

Planned expansions and updates for the OTK include:

• geoJSON

• Unit Tests