Car Number Plate Recognition Scheme Using Morphology and Backpropagation Neural Network

Car Number Plate Recognition Scheme Using Morphology and Backpropagation Neural Network

Blog Article

Car Number Plate Recognition (CNPR) is a term for automatic recognition of vehicle license plates.It is compatible with future projected transportation systems and has acquired considerable notice for its broad implementation domain in traffic departments, law execution, and security systems.The main target of CNPR ORG BRAZIL NUTS systems is to extract characters from vehicle license plate images precisely.However, the type of number plate, the font style, color, and font size of the plate, as well as the location of the number plate and environmental elements like brightness, and weather, are key challenges to detecting and identifying license plates.

This paper discusses a novel CNPR scheme evolution that combines a segmentation method using mathematical morphology with a backpropagation neural network scheme as an accurate recognition solution to overcome these limitations.Furthermore, this paper contributes to segmenting the car number plate stage by using a 2D entropy function for the binarization image of the pre-processing method, which is proposed for their role in improving the overall performance of the CNPR scheme, in addition to image resizing and noise reduction.In addition, the effect of dataset size on the training and testing phases of the CNPR scheme is discussed.The proposed scheme displays the best accuracy score of 97.

5% when utilising the entire dataset and 98.8% N-Acetyl D-Glucosamine when using only half of the dataset.