OpenCV License Plate Recognition with Python | #107 (OpenHAB #5)
The short version: A hobbyist's rough OpenCV/pytesseract plate-reader experiment shows just how easy DIY ALPR has become, even before commercial systems like Flock enter the picture.
Our Take
Brandon Jacobson's video isn't a surveillance expose — it's a home automation tinkerer trying to get OpenHAB and a MotionEyeOS camera to log plates for "home security." The OCR doesn't cooperate, so he's eyeing a TensorFlow Lite model instead. Credit to him for being transparent about the trial and error. But that transparency is exactly why this video matters to us: it's a real-time demonstration of how low the barrier to entry has dropped for building your own plate-reading camera network.
If a hobbyist can stitch together open-source computer vision tools and a Raspberry Pi-adjacent setup in an afternoon, imagine what a company with venture funding and law enforcement contracts can deploy at scale. Flock Safety isn't doing anything technically exotic — it's productizing the same basic pipeline this video walks through, then wiring it into a nationwide database that police, HOAs, and private clients can query. The gap between "guy in his garage" and "corporate ALPR dragnet" is mostly money and integration, not some unreachable technical leap.
That's worth sitting with. Every driveway camera, every DIY project like this one, and every Flock installation adds another node to a tracking system nobody voted for. Check our camera map to see what's already watching your routes, and hit up take-action if you want to push back on the infrastructure before it becomes background noise.
This is DeFlock The USA’s original commentary. The video above is the work of Brandon Jacobson (Jacobson Enterprises), published on YouTube — full credit to the creator.