Enhancing Image Recognition of Damaged Number Plate in the Running Vehicle using Genetic Algorithm Compared with Edge Detection Algorithm

Abstract

Aim : Innovative novel automatic number plate detection of damaged vehicle number plates using machine learning algorithms and improving the accuracy of recognition using genetic algorithms. Materials and methods : Two sample groups using 237 images are used as the sample dataset, which is tested at 80% for G power with t-test analysis.  To improve the accuracy of recognizing the number plates of the vehicle, the genetic algorithm is proposed and compared with the edge detection algorithm. Results and Discussion: Test results prove that the genetic algorithm has an average accuracy of 91.05%, which seems to be better than the edge detection algorithm’s accuracy of 90.02%. Since the significance is around 0.46, there appears to be a statistically significant difference among the study group with (p< 0.05). Conclusion: Recognition of number plates of a vehicle using novel genetic algorithm can have an effect of detection of violation by road side CCTV cameras.