Image Processing Based Intelligent Parking System Using Number Plate Recognition

In this work, an automatic license plate detector is proposed using digital image processing methods in which video camera can be implemented at the entry point of the vehicles which is further processed by breaking the video into frames such that license plate comes into focus range of the camera. We have developed a system where all these processes are automated using an intelligent algorithm on MATLAB. The systems use different color spaces which are converted first from RGB channels which are further used by image Binearization. Binarized image is applied to connected component analysis in which properties of the regions i.e. Area, Bounding box containing height and width parameters and centroid are used to exclude the unwanted portions in the image and to segment the characters of license plate in the frames. Further template matching is implemented to extract the license plate number in text form. The proposed system is experimented on a private parking spot in the locality in which video clips are taken for the vehicles in the entry point and proposed method is applied on the frames of the clips where license plates are focused from the viewer’s point of view. Experimental results show 90% accuracy of detection of number plate regions and characters in the collected dataset. Results can be improved further for other language number plates as well as this work is carried out on Indian license plates written in English Language.