Damaged Number Plate Recognition
P. Ezhilarasi1, S. Rajeshkannan2, Kovendan. A K P3, T. Sasilatha4, K. R. Kayalvizhi5
1P. Ezhilarasi, Department of ECE, St. Joseph’s College of Engineering, OMR, Chennai (Tamil Nadu), India.
2S. Rajeshkannan, Department of ECE, St. Joseph’s College of Engineering, OMR, Chennai (Tamil Nadu), India.
3Kovendan A K P, Department of ECE, St. Joseph’s College of Engineering, OMR, Chennai (Tamil Nadu), India.
4T. Sasilatha, Professor and Dean, Department of EEE, AMET Deemed to be University, Chennai (Tamil Nadu), India.
5K. R. Kayalvizhi, Department of ECE, St. Joseph’s College of Engineering, OMR, Chennai (Tamil Nadu), India.
Manuscript received on 16 December 2019 | Revised Manuscript received on 24 December 2019 | Manuscript Published on 31 December 2019 | PP: 22-25 | Volume-9 Issue-1S2 December 2019 | Retrieval Number: A10181291S219/19©BEIESP | DOI: 10.35940/ijeat.A1018.1291S219
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Around the world every vehicle are identified by its number plate. Number plate detection is one of the existing automated video surveillance systems that are used to detect the number plate. This system fails if the number plates are damaged, no proper illumination, blurry images. Thus here we will be able to recognizeze such damaged number plate. The technique involves four main stages viz. pre-processing, localization, recognition and segmentation. The entire process includes capturing the image, erasing the background details and removing the noise, cropping the number plate and then recognizing the characters followed by segmenting in order to recognize the plate. All this is done in Python because it had better results compared to MATLAB. When done in MATLAB, additional error and noise gets added to the input image and can causes inclusion of a new characters in the number plate and leads to misinterpretation of the number plate. About 100 images were gathered and 98 images of them were detected correctly. The efficiency in recognizing the damaged number plate using our system is about 98%.
Keywords: Pre-Processing, Character Segmentation, Character Recognition, Localization.
Scope of the Article: Pattern Recognition and Analysis