Hello everyone, it's a lot of time I don't write in public on the internet. When something difficult/weird comes to my mind, it happens that I go to DonationCoder.com :)

Here's my difficulty: I'd like to have a device, for personal use at my home, but it doesn't exist - so I can't just purchase it.

It's something that probably technically can be done these days, and I have an overall idea of how to, but can't go from start to finish myself without spending an inappropriate amount of time.

So I thought: it's 2019, the Internet and everything, of course there are communities who "design and build together" things, like software, hardware, etc.
But I'm not sure which could be, for hardware+software devices for personal use.

Thus my question: do you have recommendations? I'd basically would like to post the necessity, and try to contribute as anyone else over time, but hoping that someone smarter than me, together with someone else smarter than him/her, etc. could help with significant progress so to not wait for a lifetime for having it operative on my home's balconies.

The device that I would like to build is something that detects pigeons on balcony with artificial vision and ML, using several cameras and raspberry pi's or whatever, and moves one or more flashlights to point towards the pigeons and try to persuade them to go away. And/Or, moves a robot owl or a robot falcon, that goes towards them, scares them until they go away, and then goes back to the charging station.

The following would be useful for lucid dreaming purposes (by using Webcam Video Diary to record dreams):

- when recording or playing a video, it's possible to pause and define a "in-video note", that is made by a "title" (input line field) and a "free text" (textarea field);

- the in-video notes are stored into a sqlite database, or whatever (text file?): a record (or text-file line) is an identifier (name?) of the video they belong to, the timestamp in that video, the note's title, and the note's free text;

- possibly, when you insert a in-video note, typing the title it autocompletes using the other in-video notes' titles of every video with the same identifier label (I mean, "%IDENTIFIER%"); and also show the entries somewhere (a popup?) in a sorted-by-occurrence list, where you can click one of those titles and it gets copied into the title input field, together with the associated free-text into the textarea field (but you can still edit both, of course).

Besides, in some tab of the app or wherever you want:

- for a selected %IDENTIFIER%, show the in-video notes' titles sorted by occurrence, and through filters on the free text;

- by clicking them, jump to the corresponding video and timestamp (and play the video from there).

So, for lucid dreaming purposes, for instance one could use the titles for simple descriptions of the hooks (e.g. "pink elephant"), and the free text for some conventional string, e.g. "__IMPOSSIBLE__".
The software could allow to filter notes containing (or not) the string "__IMPOSSIBLE__" into the free text; and eventually also more complex queries (e.g. notes not containing string "__IMPOSSIBLE__" and containing string "stairs").