README

This project consists several parts:

• LAMPIR driven code, based on arduino
• Raspberry Pi code for camera, data storage
• Human detection of photos and analog data analyzing

How do I get set up?

1. LAMPIR and PIR are as slaves of Raspi.
2. Raspi runs Python code, with multithread to collect data (plot) and storage photos. See Python multithreading.
3. Filename is the system time when starting recording.
4. Picamera takes record video continously, the video files are seperated by 1 min.

How to use code to synchronized log video and PIR output

1. The code MUST run under raspberry PI3. PiCam must present as well as LAMPIR node.
2. Final code is in ./videolog/videolog.py or ./videolog/videolog_2pir.py (for logging data from LAMPIR and traditional PIR).
3. Afterrunning, the video and analog output will be stored in videodir='/home/pi/datalog/videolog/', datadir='/home/pi/datalog/lampirdata/' pirdir='/home/pi/datalog/pirdata, with 1 min intervals.

Dataprocess of voltage signals

1. All analog signals are converted through an ADC on Arduino board, range from (0, 1024), indicating (0, 5 V). Raspi logs these data into a .CSV file, in such form: b'526\n'2018-03-08 17:57:40.330576.
2. A preprocessing is required to extract such signals as well as meaningful time stamps.
3. Types of data signals:

High frequency	Signal high Vpp	Single low Vpp
Moving	Stationary	Unoccupied

4. First step is seperate high and low frequency.
5. Second step is classify stationay and unoccupied scenarios in low frequency signals.

copyright by Libo Wu, libo.wu@stonybrook.edu