Sayuti Rahman, Marwan Ramli, Fitri Arnia, Rusdha Muharar, Haida Dafitri, Dharmawati, Mahyus Ihsan
Motorists driving around looking for parking spaces has negative impacts such as increasing traffic jams, noise, air pollution, wasting fuel, and panic. A smart parking system providing information on the availability of parking spaces is needed. This study aims to develop a parking system by providing information on the availability of car parking spaces. Classification of car parking spaces using Sufflenet V2 is by trimming it. Based on the tests carried out, pruning was on the number of Sufflenet V2 output channels. This trimming produces the smallest Mini Shufflenet size with almost half the parameter number of mAlexnet and less than 1/10 times the number of FLOPs of mAlexnet. Mini Shufflenet has better accuracy than mAlexnet in processing time per image that is almost the same as mAlexnet. Mini Shufflenet implemented is on low-end computing devices such as the Raspberry Pi for real-Time data classification. So this architecture can be implemented in a low-cost parking system. © 2023 Author(s).
Doctoral Program, School of Engineering, Syiah Kuala University, Banda Aceh, Indonesia; Department of Mathematics, Syiah Kuala University, Banda Aceh, Indonesia; Department of Electrical and Computer Engineering, Syiah Kuala University, Banda Aceh, Indonesia; Department of Informatics Technology, Universitas Harapan Medan, Medan, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock