I need to read licence plates in a video file and write their numbers into a text file with a timestamp and confidence index. Something like this, but I don't need the user interface: [login to view URL]
The videos are taken from the side of the street and the cars are moving away. The licence plates are from Canada and US. Sample videos will be provided.
The tool must:
- Have a good accuracy and detection rate
- Perform the image processing locally
- Use non proprietary software (maybe OpenCV, YOLO, Tensorflow?)
- Adapt to the angle of cars in different video files with low quality image
- Be written in python 3 (if possible)
I also need you to explain the tool to me. You must:
- Comment thoroughly the code
- Provide a simple flow chart or something to explain the big picture of the tool
- Possibly have a 30 minute Skype call with me to give explanations / to answer my questions
The tool can be a unique or collection of scripts with a minimal or no user interface at all. It can be run from the command line for example.
There could be a bonus if satisfaction is exceeded and there will probably be a follow up to this project.
34 pekerja bebas membida secara purata $308 untuk pekerjaan ini
Hello Very interested in your project I can show you and deliver source code that detects license plate and save it into DB. Review my work on my profile Let's have a talk for more detail Thanks. :)
---------OCR, ALPR Expert && Best Result In Time!--------- Dear Sir... I'm Python programmer and ML expert with OpenCV. If you hire me, I'll do my best for you. Thanks.
Hi I can build you the project. However I need to discuss the timeline in which you want the project to be completed. please message to finalize details thanks