In deepstream LPR crop detected license plate and then provide it for license plate recognition

I’m working with the deepstrem LPR app. In India we have two lines of license plate but the license plate recognition model works only with one line.

To overcome this, I want to divide the detected license plate into two equal half’s and merge those half’s so that i can get the license plate in one single line which can be further provided for recognition part.

Question is how can i possibly do this because I do not know how the detected vehicle is getting cropped and is been provided for license plate detection and also how can i possibly get that detected image and do the processing to convert the 2 line license plate into 1 line and do the recognition on it.

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)

Hardware Platform - Jetson NX developer kit.
DeepStream Version - Deepstream 5.1
JetPack Version - 4.5
TensorRT Version - 7.1.3
Issue Type - new requirements

Why don’t you consider to train your LPR model to handle the two lines images? So there is no need to processing the image.

may you guide how can i train the LPR for two lines?

@Morganh Is there a way to train LPR model to recognize two lines license?

Currently it is not supported. Workaround is mentioned in Is the LPRnet trainable for license plates with two separate lines instead of single one as of US? - #10 by Morganh