Skip to content

sharmatanesh/VehicleInfoOCR

Repository files navigation

Vehicle Info Camera

An app made using Firebase MLKit Text Recognizer, JSoup and a little bit of opencv. Useful for general purpose OCR as well.

WIP

TODO (Details in Rough.md)

[X] Use Firebase OCR
[X] Add Permission handler
[X] Add OpenCv morph
[X] now fix jsoup to download captcha and show in image preview 
[X] Find CV parameters on sample from pc
[X] getting majorText from the bounding boxes
[X] Putting numberplate data into text on preview btn click
[X] Flash btn from OMR
[X] Add Bottom Drawer 
[X] Floating number plates
[ ] Instruction visuals
[ ] Blurring effect
[ ] Focus guider
[ ] Improve Webview display using js
[ ] Celebrate Konfetti popper on succesful result!

Excess

Textbox style just like numplate

Test Imgproc preprocess onto cameraSource bytebuffer frames

Much Excess

[X] Remove opencv, process only bitmap

Recent searches into cards! - make use of miniRTO here too

Size Analysis(TODO : Add screenshot)

jar libs : 7MB resources : 5MB android-web-scraper : 2 MB FirebaseMLKit : 2 MB

Credits

Vahan portals

https://parivahan.gov.in/rcdlstatus/

https://vahan.nic.in/nrservices/faces/user/searchstatus.xhtml

App workflow inspired from miniRTO app: https://github.com/chandruscm/miniRTO

Firebase MLKit, and This blog : https://medium.com/digital-curry/ firebase-mlkit-textdetection-in-android-using-firebase-ml-vision-apis-with-live-camera-72ef47ad4ebd

Android web scraping wrapper: https://github.com/daandtu/android-web-scraper