Because of the development of highways as well as the increased number of vehicles usage, much attention is required on to develop an efficient and safe intelligent transportation system. The aspect of identifying specific objects present in an image is an important criteria in areas like digital image processing and computer vision. Because of the different formats, colours, shapes, viewpoints and non-uniform illumination environment of license plates, recognising the same proves to be a tasking issue. In this paper, we present a vehicle license plate recognition model using convolutional neural network (CNN) and K-means clustering based segmentation. This methodology works on three major steps such as detection and segmentation using K-means clustering and recognition of the number in the license plate using CNN model. We have also used location and detection algorithms to improve the accuracy of detection. The experimental investigation is carried out using datasets and the observed simulation results prove that the proposed mode is more effective than the other methodologies introduced so far.
A COMPARISON PRE-TRAINED MODELS FOR AUTOMATIC INDONESIAN LICENSE PLATE RECOGNITION
