Menentukan Jumlah Kendaraan pada Persimpangan Jalan Melalui Image Processing Menggunakan Metode Jaringan Fungsi Basis Radial (JFBR) pada Traffic Light Cerdas

Authors

  • Yoyok Isnomo Universitas Negeri Malang, Jl. Semarang No. 5 Malang, Jawa Timur, Indonesia
  • Aji Wibawa Universitas Negeri Malang, Jl. Semarang No. 5 Malang, Jawa Timur, Indonesia

DOI:

https://doi.org/10.17977/um068v1i92021p692-706

Keywords:

jfbr, traffic light, k-means, image processing

Abstract

The number of large, light, and heavy vehicles continues to grow, while the addition of road expansion has not been adequate, resulting in traffic jams, especially in big cities. Conventional, automatic, and adaptive control systems that existed at that time were still unable to overcome the problem of traffic jams. This article proposes to calculate the number of vehicles on each signaled road section, which is applied at each phase of the red, yellow, and green light periods, using a combination of a Radial Basis Function Network (RBFN) and K-Means methods. The results of the calculation of the vehicles   are used to set the duration of the red, yellow, and green lights. The experimental results show that the system can calculate and recognize large, small, super large and non-engine vehicles with an error rate of 0.05397.

Jumlah kendaraan tipe besar, ringan, dan berat terus bertambah, sedangkan penambahan perluaasan jalan belum berimbang, sehingga memberi dampak kemacetan lalu lintas, terutama pada kota-kota besar. Sistem pengontrolan konvensional, otomatis dan adaptif yang ada pada saat itu masih belum bisa mengatasi permasalahan kemacetan arus kendaraan. Artikel ini mengusulkan penghitungan jumlah kendaraan pada masing-masing ruas jalan bersinyal, yang dilakukan pada setiap fase periode lampu nyala merah, kuning dan hijau, dengan menggunakan gabungan metode Jaringan Fungsi Basis Radial (JFBR) dan K-Means. Hasil perhitungan kendaraan digunakan untuk menseting lama nyala lampu merah, kuning, dan hijau. Hasil eksperimen menunjukkan bahwa sistem mampu menghitung dan mengenali jenis kendaraan besar, ringan, berat, dan tanpa mesin dengan tingkat kesalahan rata-rata sebesar 0,05397.

Author Biography

Yoyok Isnomo, Universitas Negeri Malang, Jl. Semarang No. 5 Malang, Jawa Timur, Indonesia

The number of large, light, and heavy vehicles continues to grow, while the addition of road expansion has not been adequate, resulting in traffic jams, especially in big cities. Conventional, automatic, and adaptive control systems that existed at that time were still unable to overcome the problem of traffic jams. This article proposes to calculate the number of vehicles on each signaled road section, which is applied at each phase of the red, yellow, and green light periods, using a combination of a Radial Basis Function Network (RBFN) and K-Means methods. The results of the calculation of the vehicles   are used to set the duration of the red, yellow, and green lights. The experimental results show that the system can calculate and recognize large, small, super large and non-engine vehicles with an error rate of 0.05397.

Jumlah kendaraan tipe besar, ringan, dan berat terus bertambah, sedangkan penambahan perluaasan jalan belum berimbang, sehingga memberi dampak kemacetan lalu lintas, terutama pada kota-kota besar. Sistem pengontrolan konvensional, otomatis dan adaptif yang ada pada saat itu masih belum bisa mengatasi permasalahan kemacetan arus kendaraan. Artikel ini mengusulkan penghitungan jumlah kendaraan pada masing-masing ruas jalan bersinyal, yang dilakukan pada setiap fase periode lampu nyala merah, kuning dan hijau, dengan menggunakan gabungan metode Jaringan Fungsi Basis Radial (JFBR) dan K-Means. Hasil perhitungan kendaraan digunakan untuk menseting lama nyala lampu merah, kuning, dan hijau. Hasil eksperimen menunjukkan bahwa sistem mampu menghitung dan mengenali jenis kendaraan besar, ringan, berat, dan tanpa mesin dengan tingkat kesalahan rata-rata sebesar 0,05397.

References

https://dataindonesia.id/sektor-riil/detail/jumlah-kendaraan-di-indonesia-capai-14199-juta-unit-pada-2021, diakses 29 Januari 2023

Marta L. S.,Lina E. A., Paolo A. C., Luis C. M., Bairon V,. Guillermo G., Ramiro A. A., Mauricio C., Fernando G. M., Alejandro V., Cesar G., Martin J., & Sara H.(2018), Dynamic allocation of traffic light plans as a traffic reduction strategy, IET, https://ieeexplore.ieee.org/search/searchresult.¬jsp?newsear¬ch=-true&query-Text=Dynamic%20allocation%20of%20traffic%20light%20plans%20as%20a%20traffic%20reduction%20strategy

Billie P., Josua C., Muhammad T. H., & Abdul M., (2018), Adaptive Traffic Lights through Traffic Density Calculation on Road Pattern, ICAST, https://ieeexplore.ieee.org/search/searchresult.jsp?newsearch-=true&queryText=Adaptive%20Traffic%20Lights%20through%20Traffic%20Density%20Calculation%20on%20Road%20Pattern

Anam F., Indu, Vandana N., (2020), Smart Density Based Traffic Light System, ICRITO, https://ieeexplore.ieee.org/search/searchresult.jsp?newsearch=true&queryText=Smart%20Density%20Based%20Traffic%20Light%20System

Anuj S., Sachin S., Kamlesh C. P., K. C. Nithin K. (2021), Artificial Intelligence based Framework for Effective Performance of Traffic Light Control System, ICSES, https://ieeexplore.ieee.org/document/9633913

Y.H.P. Isnomo, (2014), Predict of Lost Time at Traffic Lights Intersection Road Using Image Processing, IC-ITECHS, http://jurnal.stiki.ac.id/IC-ITECHS/article/view/353

Rocky H., (2020), Penerapan Metode Laplacian of Gaussian Dalam Mendeteksi Tepi CitraPada Penyakit Meningitis, KLIK, https://djournals.com/klik/article/download/21/5/65

Downloads

Published

26-09-2021

Issue

Section

Articles