Abstract:
The system in this paper is designed and implemented for the vehicle license plate
detection. Automatic number plate recognition has three major components:
vehicle number plate detection, character segmentation and Character Recognition.
After taking the image, the quality of the picture should be enhanced. With this
enhance image, the first license plate region is located an then K-NN algorithm is
used to character recognition after segmentation of the characters. License plate
recognition system is useful in proceeding the tickets for car park systems, finding
lost vehicles, managing the car parks, identifying the vehicles exceeding the speed
limits on highways, etc. License Plate Recognition (LPR) is an image-processing
technology, also known as Automatic Number Plate Recognition (ANPR). There
are several types of existing number plate recognition systems with different type of
methods and algorithms. This paper presents a more efficient and accurate system
for recognizing the license plate number. The number plate detection is done by
image processing techniques. Number plate extraction is that stage where the
vehicle number plate is detected. The detected number plate is pre-processed to
remove the noise and then the result is passed to the segmentation part to segment
the individual characters from the extracted number plate. Here the KNN
algorithm gives a better approach to the character recognition. k-Nearest Neighbor
algorithm (or k-NN for short) is a non-parametric method used for classification
and regression. In both cases, the input consists of the k closest training examples in
the feature space. The proposed system can be used in license plate recognition
applications which is efficient and accurate than the existing systems.