Perancangan sistem klasifikasi kendaraan berbasis machine learning = Vehicle Classification System Design Based On Machine Learning

Lauwrent, Fedrien (2022) Perancangan sistem klasifikasi kendaraan berbasis machine learning = Vehicle Classification System Design Based On Machine Learning. Bachelor thesis, Universitas Pelita Harapan.

[img]
Preview
Text (Title)
Title.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (324kB) | Preview
[img]
Preview
Text (Abstract)
Abstract.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (870kB) | Preview
[img]
Preview
Text (ToC)
ToC.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (556kB) | Preview
[img]
Preview
Text (Chapter1)
Chapter1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (705kB) | Preview
[img] Text (Chapter2)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (1MB)
[img] Text (Chapter3)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (774kB)
[img] Text (Chapter4)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (2MB)
[img] Text (Chapter5)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (1MB)
[img] Text (Chapter6)
Chapter6.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (703kB)
[img]
Preview
Text (Bibliography)
Bibliography.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (466kB) | Preview
[img] Text (Appendices)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (5MB)

Abstract

Meningkatnya jumlah kendaraan mengakibatkan kebutuhan teknologi dalam mendeteksi kendaraan yang semakin cepat, terutama dalam mendukung peraturan pemerintah dalam melakukan pendeteksian seperti ETLE (Electronic Traffic Law Enforcement). Tujuan penerapan ETLE adalah suatu kamera tilang elektronik yang akan merekam pelanggaran yang terjadi. Melihat kondisi tersebut, maka dibuat perancangan sistem untuk mendeteksi kendaraan yang dapat dilakukan baik dari kamera yang ditempatkan pada jalan raya, maupun direkam dari kendaraan yang sedang bergerak. Salah satu cara untuk mendeteksi kendaraan adalah dengan membuat sebuah sistem machine learning yang dapat membedakan jenis objek yang terdeteksi. Sistem machine learning pada penelitian ini dirancang dengan menggunakan aplikasi tensorflow yang memanfaatkan prinsip dari Convolutional Neural Network dan menggunakan aplikasi LabelImg yang digunakan untuk menandai objek sebagai pengenalan dataset. Dengan input dari kamera yang diletakkan pada bagian depan kendaraan yang akan mengambil gambar pada kendaraan lain yang terdeteksi, sistem machine learning yang akan melakukan pengolahan data dari datasets yang sudah dilatih untuk mengetahui jenis objek yang terdeteksi. Dengan hasil perancangan yang mengenalkan kepada sistem perbedaan jenis pada objek, melakukan uji coba, maka perancangan ini mendapatkan hasil berupa data yang menunjukkan tingkat presisi, akurasi pendeteksian, nilai recall dan confusion matrix. Hasil uji coba dilakukan terhadap dua buah video yang berbeda, di mana pada video pertama hasil akurasi yang didapatkan sebesar 45% dan pada video kedua 68%, tingkat presisi pada video pertama sebesar 42% dan video kedua 83% , dan nilai recall pada video pertama sebesar 33% dan pada video kedua sebesar 66%./The increasing number of vehicles results in the need for technology to detect vehicles faster, especially to supporting government regulations for carrying out detections such as ETLE (Electronic Traffic Law Enforcement). One application of ETLE is an electronic ticketing camera that will record violations that occur. Seeing these conditions, a system is designed to detect vehicles that can be done either from cameras placed on the highway or recorded from moving vehicles. One way to detect vehicles is to create a machine learning system that can distinguish between vehicles and non-vehicles. Machine learning system on this observation using tensorflow which take advantage of Convolutional Neural Network metode, and Labelimg to make a labels in every objects. With input from a camera placed on the front of the vehicle that will take pictures of other detected vehicles, a machine learning system that will process data from datasets that have been trained to determine the class of the vehicle. This vehicle class classification system can be an alternative to the current method of determining vehicle class. With the results of the design that will introduce the system of class differences to objects, conduct trials, so that this design gets results in the form of data that shows the level of precision, detection accuracy, recall value and confusion matrix. The result test was conducted on two different videos, where in the first video the accuracy results were 45% for the first video and 68% for the second video, the precision level was 42% for the first video and 83% for second video, and the recall value was 33% for the first video and 66% for the second video.

Item Type: Thesis (Bachelor)
Creators:
CreatorsNIMEmail
Lauwrent, FedrienNIM01032180003f2riant@gmail.com
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorJunita, JunitaNIDN0302068401UNSPECIFIED
Uncontrolled Keywords: Machine learning; Image processing; Pendeteksi kendaraan; Deteksi objek
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: University Subject > Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Electrical Engineering
Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Electrical Engineering
Depositing User: Users 9148 not found.
Date Deposited: 25 Feb 2022 02:45
Last Modified: 25 Feb 2022 02:45
URI: http://repository.uph.edu/id/eprint/46769

Actions (login required)

View Item View Item