This house I live in does not have much in the way of garden; there are some loosely termed ‘terraces’ with native and Imported plants. Some lawn outside the front fence.

There is also a steep rear ‘terrace’ that is little more than a leftover from the original house construction.

Most of it has some grass, some plant-life and some texture to it. Except the bit next to the garage.

That is a sloping, 3 metre wide x 9 metre long desolate, clay & dirt space that grows nothing but weeds.

I keep it weed-free as much as possible, and so it is now the neighbourhood’s Cat Toilet.

Of course it is …

I’ve tried soft-drinks bottles with water, empty cans on pieces of string, pieces of hose, small piles of dead weeds, external sensor-lights … all to no avail.

Unfortunately as an animal-lover I cannot consider lethal means, which leaves me Technology.

Having a rummage around this afternoon in my Raspberry Pi box, I found this:

  • 2 x Raspberry Pi-connectible Cameras
  • 5 x Raspberry Pi v3B+
  • 1 x Raspberry Pi v4
  • Networking gear
  • Spare car battery from the Garage
  • Python & OpenCV surveillance code that I wrote a few years back

Project Cat Meet Python (CMP):

  • Mission: Put together an adhoc but functional surveillence system
  • Components: Hardware listed above; Python; OpenCV; Chewing Gum
  • Task: Identify the miscreant Kitty(s) toileting in my yard
  • Data: Capture Images, Date/Time and Deposit details
  • Search: Build a web-fronted Database (PostGRES/SQL) for queries
  • Visualisation: Event Type; Frequency; Location; Subject (Bad Kitty)
  • Expansion: Motion-activated Hose Blast
  • Enhancements: Sound or Light Generation

That will do for a start, and since Tomorrow is the weekend, I can source, test and implement some simple frame-grabbing routines and go from there.

A great place to start with planning the camera control will be Image and video recording with the Raspberry PI or specifically PiCamera.

In general there is so much quality RPi information on the WWW, that all it takes is some time, a RPi and some effort, and you can create almost anything. A great place to start reading on what to do with an RPi and Python is The Raspberry Pi Guide for scientists and anyone else.

I can’t wait.

Project CMP is a go.

References: