Show HN: Build a DIY license plate reader with machine learning

Hey HN,<p>I’m Caleb, and I’m a maintainer of Cortex, an open source model deployment platform. Not long ago, we published this DIY license plate reader project, and I wanted to share it here for anyone who is interested in computer vision or production machine learning in general.<p>The project is a web service that accepts images and, using three trained models, returns extracted license plate text, assuming there is a license plate in the image. Of the models used, two are pre-trained models from keras-ocr, while one is a fine tuned YOLOv3. All models are freely available.<p>You can see a video of the project in action here: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=gsYEZtecXlA" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=gsYEZtecXlA</a><p>And read a write up by Robert Lucian, the maintainer who spearheaded this entire project, about how he built a camera system to interface with the web service using a Raspberry Pi and 5G: <a href="https:&#x2F;&#x2F;towardsdatascience.com&#x2F;i-built-a-diy-license-plate-reader-with-a-raspberry-pi-and-machine-learning-7e428d3c7401" rel="nofollow">https:&#x2F;&#x2F;towardsdatascience.com&#x2F;i-built-a-diy-license-plate-r...</a>

Take care to understand your local law regarding automatic license plate recognition. Use of such software and&#x2F;or data collected with it may be under regulation. IANAL, but see your state statutes here if you live in the US: <a href="https:&#x2F;&#x2F;www.ncsl.org&#x2F;research&#x2F;telecommunications-and-information-technology&#x2F;state-statutes-regulating-the-use-of-automated-license-plate-readers-alpr-or-alpr-data.aspx" rel="nofollow">https:&#x2F;&#x2F;www.ncsl.org&#x2F;research&#x2F;telecommunications-and-informa...</a>

Am I missing something or in this image, is the confidence score 99.23% and has the incorrect license plate? The plates clearly have a 5 but it&#x27;s predicting an S.<p><a href="https:&#x2F;&#x2F;camo.githubusercontent.com&#x2F;5138d6bce5c840d15e533db6f300660c80eb2619&#x2F;68747470733a2f2f692e696d6775722e636f6d2f746731504531452e6a7067" rel="nofollow">https:&#x2F;&#x2F;camo.githubusercontent.com&#x2F;5138d6bce5c840d15e533db6f...</a>

&quot;Do-it-yourself&quot; seems like a not quite appropriate title for an interface that calls a webservice with pre-built vision ML.

I&#x27;ve been watching dashcam footage lately and thinking about this. It would be very helpful for hit and runs if the dashcam had some on-board recognition that saved the last x license plates just like it saves the last x seconds of videos.<p>Often the video footage is not good enough to get a plate because the capture settings are set lower so that more footage can be stored.

I wonder how much more revenue police departments are making with ANPR[1].<p>They can, for example, have a camera in all the cruisers, and automatically alert when a nearby car has an expired registration, inspection, etc. Or, if they have the tie-ins, registered to an owner with an open warrant or unpaid traffic citations.<p>[1] <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Automatic_number-plate_recognition" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Automatic_number-plate_recogni...</a>

I’ve pondered the implications of running an ‘installation art’ project where a bunch of these are deployed around a city, with aggregated data visible over a web UI (assuming that this doesn’t violate any laws.) The aim would be to raise public awareness of pervasive surveillance, and perhaps catalyze changes to the law.

How does this implementation perform compared to classic approaches such as OpenALPR [1] ?<p>At the very least running local inference becomes much more expensive, and possibly provides worse results.<p>[1] - <a href="https:&#x2F;&#x2F;github.com&#x2F;openalpr&#x2F;openalpr" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;openalpr&#x2F;openalpr</a>

The actual computation here shouldn&#x27;t be that great, right? (especially given the RPi has a GPU on) Seems like an inconsiderate design to stream images to a cloud service, rather than process locally and stream the plates themselves.

This is great! I&#x27;ve been meaning to build something like this for a long time. I&#x27;m always excited when I notice I&#x27;ve passed the same car on commutes on different days, or see a car I recognize from home in a different part of town. Automating that process and having my phone tell me e.g. &quot;You just passed a car that you once drove past 1,000 miles away from here&quot; is such a tempting side project. Bit big-brothery, though.

This is great. One of my side project ideas is to build an ANPR based app and &#x27;social network&#x27; starting with basic tracking, I.e. you&#x27;ve passed this car 4 times before. I&#x27;m new to the ML world so still learning vut Do you think this is feasible to run in real time using CoreML on iOS?

I was thinking the other day about a DIY solar panel for apartments that could be part of the decoration like within plant vases or something in that fashion that would be cool and would look good and not have &quot;space being occupied&quot;, that would be interesting.

<a href="https:&#x2F;&#x2F;music.youtube.com&#x2F;watch?v=w1mUAVPaX98&amp;feature=share" rel="nofollow">https:&#x2F;&#x2F;music.youtube.com&#x2F;watch?v=w1mUAVPaX98&amp;feature=share</a>

&quot;You, too, can now actively participate in the surveillance state&quot;<p>Yes, I get the technical appeal. I wish we all wondered less if we could, and more if we should.