KLASIFIKASI JENIS MOBIL PADA JALAN RAYA MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORKS (CNN)

dc.contributor.authorHafifah, Febri
dc.date.accessioned2022-12-07
dc.date.available2022-12-07
dc.date.issued2021
dc.identifier.uri https://ejurnal.seminar-id.com/index.php/tin
dc.description.abstract ABSTRACT Traffic congestion is a major problem that occurs in big cities in Indonesia. This can cause various negative impacts such as waste of fuel, waste of time, and air pollution. Therefore, the government divides the types of highways and only allows large cargo trucks to pass on arterial roads. So it is necessary to have a smart city to implement government policies in order to overcome the impacts caused by traffic congestion. Classification of the types of vehicles that pass on the highway needs to be done so that there are no vehicle violations outside the specifications that are allowed to enter certain highways. Classification of vehicle types using the Convolution Neural Network (CNN) method. The architecture used is in the form of an existing CNN architecture or an existing CNN architecture, namely googlenet and shufflenet. We fine tune Googlenet and Shufflenet to get maximum accuracy. The dataset used is data taken from several CCTV camera points in several cities in Indonesia in July 2021. The proposed method can classify vehicle types with an accuracy of 95.88% Googlenet and 96.48% Shufflenet. Thus, it is hoped that it can contribute to researchers to develop a better CNN so that it can be implemented for the benefit of road traffic in Indonesia in the future. Keywords: Vehicle type classification, convolutional neural network (CNN), Googlenet, Shufflenet, fine tuning en_US
dc.language.isoenen_US
dc.publisherUniversitas Harapan Medanen_US
dc.subjectAnalisis Klasifikasi Menggunakan Convolutional Neural Network (CNN)en_US
dc.titleKLASIFIKASI JENIS MOBIL PADA JALAN RAYA MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORKS (CNN)en_US
dc.typeSkripsien_US


File In This Item

No Thumbnail
Name 884e4513f76d53e81ed5f6566020b92f915-Article Text-3054-1-10-20211028.pdf
Size 908419 Mb
Format application/pdf
Description fulltext
Fulltext
No Thumbnail
Name 884e4513f76d53e81ed5f6566020b92f17350120.docx
Size 17457 Mb
Format application/vnd.openxmlformats-officedocument.wordprocessingml.document
Description peer_review
Peer Review

This item appears in the following Collection(s)

Skripsi [1393]

Show Simple Item Record