Now that the CloudWatcher is able to measure more weather data than ever, that is:
- ambient temperature,
- sky reflected IR,
- atmospheric pressure,
we are reactivating a project of old to apply machine learning to cloud and rain detection.
The idea is to collect a huge amount of data, cloudwatcher sensor data and all-sky images, preprocess it, and apply a self-learning algorithm.
At this stage, we want to start collecting data for a period of 1 year.
If you have a CloudWatcher and an all-sky camera, and want to participate, please contact us.
What we ask from you is:
– to let us recover 1 set of data each hour, consisting of the CloudWatcher sensor readings and 1 all-sky image.
– to share your location (GPS coordinates, including altitude) with us – we won’t disclose them.
– we will make public any findings
– we will improve your cloud detection, either with improved K-factors and limits for the current algorithm or with a new and more accurate system, depending on the outcome of the research.
We will also invite participants to help characterize the data, namely, identify the sky in the images depending on the cloud formations.