SMART OPTIMIZER SELECTION TECHNIQUE: A COMPARATIVE STUDY OF MODIFIED DENSNET201 WITH OTHER DEEP LEARNING MODELS

Kamaran Manguri

kamaran@uor.edu.krd
1Erbil Polytechnic University, Erbil Technical Engineering College, Department of Technical Information System Engineering, 2University of Raparin, Department of Software and Informatics Engineering (Iraq)
https://orcid.org/0000-0001-8567-3367

Aree A. Mohammed


University of Sulaimani, College of Science, Computer Science Department (Iraq)
https://orcid.org/0000-0001-9710-4559

Abstract

The rapid growth and development of AI-based applications introduce a wide range of deep and transfer learning model architectures. Selecting an optimal optimizer is still challenging to improve any classification type's performance efficiency and accuracy. This paper proposes an intelligent optimizer selection technique using a new search algorithm to overcome this difficulty. A dataset used in this work was collected and customized for controlling and monitoring roads, especially when emergency vehicles are approaching. In this regard, several deep and transfer learning models have been compared for accurate detection and classification. Furthermore, DenseNet201 layers are frizzed to choose the perfect optimizer. The main goal is to improve the performance accuracy of emergency car classification by performing the test of various optimization methods, including (Adam, Adamax, Nadam, and RMSprob). The evaluation metrics utilized for the model’s comparison with other deep learning techniques are based on classification accuracy, precision, recall, and F1-Score. Test results show that the proposed selection-based optimizer increased classification accuracy and reached 98.84%.


Keywords:

deep learning, optimization technique, transfer learning, customized dataset, modified DenseNet201

Ahmed T. et al.: A Deep Learning based Bangladeshi Vehicle Classification using Fine-Tuned Multi-class Vehicle Image Network (MVINet) Model. 2023 International Conference on Next-Generation Computing, IoT and Machine Learning – NCIM, 2023, 1–6.
DOI: https://doi.org/10.1109/NCIM59001.2023.10212619   Google Scholar

Ahmed U. et al.: Multi-aspect detection and classification with multi-feed dynamic frame skipping in vehicle of internet things. Wireless Netw, 2022, 1–12.
DOI: https://doi.org/10.1007/s11276-022-03076-9   Google Scholar

Ashir S. M. et al.: A Transfer-Learning-Based Approach for Emergency Vehicle Detection. Eurasian Journal of Science and Engineering 8(1), 2022.
DOI: https://doi.org/10.23918/eajse.v8i1p75   Google Scholar

Biswas D. et al.: An automatic car counting system using OverFeat framework. Sensors 17(7), 2017, 1535.
DOI: https://doi.org/10.3390/s17071535   Google Scholar

Dong S. et al.: A survey on deep learning and its applications, Computer Science Review 40, 2021, 100379.
DOI: https://doi.org/10.1016/j.cosrev.2021.100379   Google Scholar

Fouad M. M. et al.: Automated vehicle inspection model using a deep learning approach. J Ambient Intell Human Comput 14, 2023, 13971–13979.
DOI: https://doi.org/10.1007/s12652-022-04105-3   Google Scholar

Ghazal B. et al.: Smart traffic light control system. Third international conference on electrical, electronics, computer engineering and their applications – EECEA, 2016, 140–145.
DOI: https://doi.org/10.1109/EECEA.2016.7470780   Google Scholar

Hassan E. et al.: The effect of choosing optimizer algorithms to improve computer vision tasks: a comparative study. Elmougy and Applications 82(11), 2023, 16591–16633.
DOI: https://doi.org/10.1007/s11042-022-13820-0   Google Scholar

Impedovo D. et al.: Vehicular traffic congestion classification by visual features and deep learning approaches: a comparison. Sensors 19(23), 2019, 5213.
DOI: https://doi.org/10.3390/s19235213   Google Scholar

Jain N. K. et al.: A review on traffic monitoring system techniques. SoCTA 2019, 569–577.
DOI: https://doi.org/10.1007/978-981-13-0589-4_53   Google Scholar

Joo H. et al.: Traffic signal control for smart cities using reinforcement learning. Computer Communications 154, 2020, 324–330.
DOI: https://doi.org/10.1016/j.comcom.2020.03.005   Google Scholar

Jung H. et al.: ResNet-based vehicle classification and localization in traffic surveillance systems. IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2017, 61–67.
DOI: https://doi.org/10.1109/CVPRW.2017.129   Google Scholar

Ke X. et al.: Multi-dimensional traffic congestion detection based on fusion of visual features and convolutional neural network. IEEE Transactions on Intelligent Transportation Systems 20(6), 2018, 2157–2170.
DOI: https://doi.org/10.1109/TITS.2018.2864612   Google Scholar

Khan A. et al.: A survey of the recent architectures of deep convolutional neural networks. Artif Intell Rev 53, 2020, 5455–5516.
DOI: https://doi.org/10.1007/s10462-020-09825-6   Google Scholar

Leitner D. et al.: Recent advances in traffic signal performance evaluation. Journal of Traffic and Transportation Engineering 9(4), 2022, 507–531.
DOI: https://doi.org/10.1016/j.jtte.2022.06.002   Google Scholar

Manguri K. H. K. et al.: A Review of Computer Vision–Based Traffic Controlling and Monitoring. UHD Journal of Science and Technology 7(2), 2023, 6–15.
DOI: https://doi.org/10.21928/uhdjst.v7n2y2023.pp6-15   Google Scholar

Manguri K. H. K., Mohammed A. A: Emergency vehicles classification for traffic signal system using optimized transfer DenseNet201 model. Indonesian Journal of Electrical Engineering and Computer Science 32(2), 2023, 1058–1068.
DOI: https://doi.org/10.11591/ijeecs.v32.i2.pp1058-1069   Google Scholar

Mohammad M. A. et al.: New Ontology structure for intelligent controlling of traffic signals. Procedia Computer Science 207, 2022, 1201–1211.
DOI: https://doi.org/10.1016/j.procs.2022.09.176   Google Scholar

Qadri S. S. S. M. et al.: State-of-art review of traffic signal control methods: challenges and opportunities. Eur. Transp. Res. Rev. 12(55), 2020, 1–23.
DOI: https://doi.org/10.1186/s12544-020-00439-1   Google Scholar

Razali N. A. M. et al.: Gap, techniques and evaluation: traffic flow prediction using machine learning and deep learning. J Big Data 8(1), 2021, 1–25.
DOI: https://doi.org/10.1186/s40537-021-00542-7   Google Scholar

Roy S., Rahman M. S.: Emergency vehicle detection on heavy traffic road from cctv footage using deep convolutional neural network. International Conference on Electrical, Computer and Communication Engineering – ECCE, 2019, 1–6.
DOI: https://doi.org/10.1109/ECACE.2019.8679295   Google Scholar

Tomar I. et al.: State-of-Art review of traffic light synchronization for intelligent vehicles: current status, challenges, and emerging trends. Electronics 11(3), 2022, 465.
DOI: https://doi.org/10.3390/electronics11030465   Google Scholar

Download


Published
2023-12-20

Cited by

Manguri, K., & Mohammed, A. A. (2023). SMART OPTIMIZER SELECTION TECHNIQUE: A COMPARATIVE STUDY OF MODIFIED DENSNET201 WITH OTHER DEEP LEARNING MODELS. Informatyka, Automatyka, Pomiary W Gospodarce I Ochronie Środowiska, 13(4), 39–43. https://doi.org/10.35784/iapgos.5332

Authors

Kamaran Manguri 
kamaran@uor.edu.krd
1Erbil Polytechnic University, Erbil Technical Engineering College, Department of Technical Information System Engineering, 2University of Raparin, Department of Software and Informatics Engineering Iraq
https://orcid.org/0000-0001-8567-3367

Authors

Aree A. Mohammed 

University of Sulaimani, College of Science, Computer Science Department Iraq
https://orcid.org/0000-0001-9710-4559

Statistics

Abstract views: 192
PDF downloads: 173