Over counting and feedback

[Smexah]

Hi,

First of all: Thanks for the code! I love the idea of this project.
I have implemented your code on an ESP like 3 month ago. After a little time of reviewing what you did there - all worked fine. Little more documentation would be nice, but it was ok anyway. So, I am "testing" since then.

First, I mounted the TOF in the doorframe (1.95m), facing down the frame. In many cases, it worked fine. Biggest problem is the "over counting" so that I sometime have up to 15 imaginary people around me ;)
Then I remounted the sensor over a door (2m) with the door opening to the other side. But in my opinion it made things worse.
For example: when you leaving the room by opening the door and closing it in the same move, the counter do not decrease. And if you are entering again, you have a new imaginary friend.

After encountering the “door issue” I mounted it over a frame with no door in it. I would say it works better, but after all not perfect. I have tried different angles and positions on the frame, but nothing seems to be optimal.

Ideas as feedback: A little more doc, like ports and maybe a “how to install” with steps. For better understanding, an image or schema of your real world hardware setup/installation (like sensor position) would be a wonderful extension.
Anyway thanks for your work.
Have a nice day :)

[mksmo]

Got the same issue.
It will be convenient to add switch of reseting counter without restart.

[Smexah]

Got the same issue.
It will be convenient to add switch of reseting counter without restart.

Well there is already a implementation for reset the ESP via switch in HA - this will reset the counter in HA too.

[mksmo]

Found some info to calibrate ROI
https://www.st.com/resource/en/application_note/dm00516219-using-the-programmable-region-of-interest-roi-with-the-vl53l1x-stmicroelectronics.pdf