Every year, traffic problems such as traffic violations, accidents, and others require strict supervision from law enforcers on the road as such problems can reduce the living quality of civilians in urban environments. As the number of vehicles keeps growing, the police department needs an optimal approach to ease their work. Thus, with the development of a computer aid-based system that allows optimal support, a vehicle plate recognition system becomes one of the solutions that can help the security department to reduce the number of traffic violations. A vehicle plate recognition system simply identifies the number of the captured vehicle plate and provides any necessary details for the police. In this case, many studies have been done using Machine Learning (ML) and Deep Learning (DL) to aid this system for higher accuracy and efficiency. However, ML and DL require further optimization to keep learning new features from the ever- changing environment. As the result, several studies stated the importance of adapting to these changes by using computational intelligence algorithms to adapt to these changes such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), etc (Quiros et al., 2016; Tang et al., 2021).

Related Works

Regarding the aspect of computational intelligence, there are several studies that have been done related to the implementation of the aspect for vehicle plate recognition systems. Most of the studies used GA for the system while there were others that implemented PSO and Artificial Bee Colony (ABC) algorithm. The table below shows the details like the following:

Table 1. Related Works

Authors Study Purpose Method Dataset Result
(Muhammad & Altun, 2016) Increasing               the evaluation score of plate detection using GA repeatedly to get the highest score possible Histogram of Oriented Gradient feature extraction with Genetic Algorithm A dataset of Turkish plate numbers which comprised 1537 images 98.75%


(Vijayalakshmi & Sumathi, 2012) Identifying vehicles by detecting the registered            plate number and character Genetic Algorithm (GA) LPR database comprised of 108 images for plate number detection and 310 images for character recognition 92.5%

accuracy for vehicle detection and 91% accuracy for character recognition

(Thakur et al., 2015) Fixing the existing techniques by using fixed parameter to recognize error during plate area scaling by designing new GA with flexible fitness function                 and connected component analysis technique Artificial Neural Network combined with GA A database comprised of vehicle plate numbers 97% accuracy
(Yu et al., 2016) Proposing framework of vehicle plate recognition that can handle constrains such as illumination states, noise, shadowing and uniform size of plate text Artificial Bees Colony (ABC) Algorithm in Back- Propagation Network A dataset of vehicle images with plate number The author claimed that their method was effective.
(Alavi & Varmazabadi, 2015) Creating a more precise approach for car plate detection to minimize any existing errors Neural Network Based on PSO A dataset of 400 samples of Iranian and foreign license plates 96.4%



From this simple study review of computational intelligence algorithm for vehicle plate recognition system, we can conclude that genetic algorithm is the standard used in the aspect of computational intelligence for the vehicle plate recognition system. Among the studies, the method proposed by Muhammad & Altun in their study has the highest accuracy. However, the dataset used is different from the other works so it is quite ambiguous for the comparison. Furthermore, they require more details on dataset and results of number as proof of study.


Alavi, S. M., & Varmazabadi, M. A. (2015). Car plate detection with neural network based on particle swarm optimization (PSO). Advances in Environmental Biology, 9(3), 774–779.

Muhammad, J., & Altun, H. (2016). HOG-temelli Öznitelikler ve Genetik Algoritma Kullanarak Iyilestirilmis Plaka Bölgesi Belirleme. 2016 24th Signal Processing and Communication Application Conference, SIU 2016 – Proceedings, 2014, 1269–1272.

Quiros, A. R. F., Abad, A., Bedruz, R. A., Uy, A. C., & Dadios, E. P. (2016). A genetic algorithm and artificial neural network-based approach for the machine vision of plate segmentation and character recognition. 8th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management, HNICEM 2015, December.

Tang, J., Zeng, J., Wang, Y., Yuan, H., Liu, F., & Huang, H. (2021). Traffic flow prediction on urban road network based on License Plate Recognition data: combining attention-LSTM with Genetic Algorithm. Transportmetrica A: Transport Science, 17(4), 1217–1243.

Thakur, M., Raj, I., & Ganesan, P. (2015). The cooperative approach of genetic algorithm and neural network for the identification of vehicle License Plate number. ICIIECS 2015 – 2015 IEEE International Conference on Innovations in Information, Embedded and Communication Systems, 1–6.

Vijayalakshmi, P., & Sumathi, M. (2012). Design of algorithm for vehicle identification by number plate recognition. 4th International Conference on Advanced Computing, ICoAC 2012.

Yu, W., Hu, D., & Li, C. (2016). Using Artificial Bees Colony Algorithm for License Plate Recognition. 71(Icmmita 2016), 51–55.

Alvin Linardi 2201795684
Aeron Mitchell 2201752691
Evan Edbert 2201750105

Antoni Wibowo