Comparison of Vehicle License Plate Detection Algorithms and LP Character Segmentation and Recognition using Image Processing
Geerisha Jain

Geerisha Jain, Department of Computer Science Engineering, Vellore Institute of Technology, Vellore (Tamil Nadu), India.

Manuscript received on 01 November 2022 | Revised Manuscript received on 16 November 2022 | Manuscript Accepted on 15 November 2022 | Manuscript published on 30 November 2022 | PP: 67-75 | Volume-11 Issue-12, November 2022 | Retrieval Number: 100.1/ijitee.L934211111222 | DOI: 10.35940/ijitee.L9342.11111222
Open Access | Ethics and 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: In the last couple of decades, the number of vehicles has increased drastically, consequently, it is becoming difficult to keep track of each vehicle for purpose of law enforcement and traffic management. License Plate Detection is used increasingly nowadays for the same. The system performing the task of License Plate detection is known as the LPR system which generally consists of three steps: Detection of the License plate, Segmentation of License plate characters, and Recognition of the characters of the License Plate (LP). But in real-world scenarios, the various lighting conditions, camera angle, and rotation degrades the accuracy of License Plate region detection, which in turn causes inaccurate segmentation and recognition of the license plate characters hence leading to low accuracy of the LPR systems. Therefore, it is vital to consider the most promising algorithm or technique for LP detection. In this paper, we will be analyzing and comparing five different methods for license plate detection: Morphological reconstruction, Sobel Operator, Top Hat Transform, Histogram processing, and Canny Edge detection. We will be experimentally applying these techniques on real-time captured vehicle images, using the Bounding Box algorithm for character segmentation, performing license plate character recognition using Template matching, and subsequentially evaluating and demonstrating the LPR system that promises the most accurate and efficient results. 
Keywords: License Plate Detection, Image Processing, Bounding Box, Morphological Reconstruction, Top Hat Transform, Histogram
Scope of the Article: Image Processing and Pattern Recognition