Number-Plate-Detection-HumAIn

Introduction:

Given project is divided into two parts

1.Data Extraction and Model Generation :-

In this step we extract dataset from the Indian_Number_plates.json file.

We used VehicleDataDownload.ipynb jupyter notebook for downloading the vehicle images from the json file

In Model generator foder we created a notebook named as model_generator.ipynb, this create a CNN model for the OCR.

2.Prediction

In this step we creted a PlateRecognize.py file that takes input from the user and output the predicted number plate of the vehicle.

Softwares and Technology used :

• OpenCV
• PIL
• Python 3
• Tensor Flow 1.14
• Tesseract OCR
• Anaconda package manager for python 3

How to run program :

• Step 1.Open command line (in windows ) or Terminal
• Step 2.Run the command python PlateRecognize.py
• Step 3.Enter the file name ( image name ) with extension (ENTER Image Name if image in same directory else write dirname/imagename.extension)
• Step 4.Get the result
• OR
• Step 1. Open PlateRecognize.py in idle or any other IDE and run the program
• Step 2. Enter image name (ENTER Image Name if image in same directory else write dirname/imagename.extension)
• Step 3. Get the result