I built a local-first, privacy-focused desktop app to turn regular CCTV cameras into smart text & license plate monitors (Python + OpenCV + Electron)

Hey everyone,

Like a lot of you running local camera setups, I wanted a way to pull useful data out of my RTSP/IP camera streams (like reading vehicle license plates at a gate or tracking text readouts on machinery) without sending my video feeds to a third-party cloud subscription service.

Existing ANPR or CCTV AI software is either heavily locked into enterprise pricing or privacy-invasive. So, I spent the last few weeks building my own lightweight desktop utility: VisuDock Live OCR.

How it works:

100% Local Processing: It runs entirely on your desktop using Python, OpenCV, and RapidOCR. No cloud calls, no external data leakage.

Dual Modes:

ANPR Mode: For monitoring gates, driveways, or parking logs.

General Mode: For tracking general text regions on camera feeds (e.g., container numbers, warning lights, status readouts) with instant alerts.

Lightweight & Fast: Designed not to hog system resources while processing live RTSP or IP camera feeds.

Why I built it:

I wanted something simple that I could just fire up on a local machine alongside my existing cameras to handle logging automatically, replacing messy paper logs or expensive commercial software.

I’d love to hear your thoughts, feedback, or what features you'd find useful in a local-first CCTV companion tool like this. Happy to answer any technical questions about how the OpenCV/OCR pipeline is put together!

Link is in my profile / DM me for the link if you want to check it out.

https://redd.it/1x4wie4
@r_SelfHosted