I needed a microwave radar sensor that can detect presence of the object through the glass, but I didn't find it easily in surfing on internet.
Previously I have used RCWL-0516, but it was only for motion detect.
What you proposed?
I'd use some type of imaging device and Machine Learning to detect objects through the looking glass.
Linear Regression might work.
KNN, K Nearest Known Neighbour, would work.
TensorFlow Lite, would work.
I use a thermal camera to detect the presence of heated bodies. The heated bodies are processed using KNN, to differentiate between humans and animals. Simple machine learning with Arduino KNN | Arduino Blog
I use TensorFlow, on a Raspberry Pi, to do Object ID with regular color camera.
Thermal through glass can be tricky
But a regular color camera would do better?
With regards to radar devices, some RFbeam devices such as the K-LC2, K-LC7, K-LD2, or the K-LD7 could be applicable.
Fitting a straight line to some data? Fascinating!
Worked quite well and fast for handwriting analysis. I was able to create a LR model of a bird, worked poorly. Predicting the air pressure for the next 5 days using the last 5 days of collected air pressure as a trained model, now there LR worked great. All tired using an ESP32.
To reveal pressure trends, linear regression (fitting a straight line) to recent data is indeed the method of choice!
The other important trend: to state that "the weather tomorrow will be the same as today" is correct about 74% of the time.
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.