Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/26176
Title: | Klasifikasi Kerusakan pada Ban Menggunakan Cnn dengan Arsitektur Mobilenet |
Other Titles: | Tire Damage Classification Using CNN with Mobilenet Architecture |
Authors: | Pane, Nurul Almadinah |
metadata.dc.contributor.advisor: | Muhathir |
Keywords: | klasifikasi;kerusakan;ban;cnn;arsitektur;mobilenet;classification;damage;tire |
Issue Date: | Aug-2024 |
Publisher: | UNIVERSITAS MEDAN AREA |
Series/Report no.: | NPM;198160007 |
Abstract: | Ban merupakan komponen yang berbentuk wadah yang terbuat dari karet, kawat, banang, dan beberapa zat kimia lainya yang akan diisi dengan udara. Ban berfungsi sebagai peran penopang utama seluruh beban kendaraan, meneruskan arah streering dan menjaga kestabilan kemudi, menahan dan meneruskan tenaga mesin, dan peredam getaran pada kendaraan. Maka diperkukan pendekatan digital agar dapat mengenali ban tersebut layak atau tidak secara cepat, tepat, efesien, dan akurat. Sehingga penelitian ini membuat suatu penelitian kerusakan pada ban agar dapat menegenali ban layak atau tidaknya secara cepat, tepat, efesien dan akurat. Hyperparameter Epoch 50, Batch 64, Optimizer (Adam,NAdam,RMSprop, SGD,Adadelta) dan Learning Rate 0.001 dengan dasaet 1000 citra Ban. Klasifikasi yang dilakukan dalam penelitian ini menggunakan arsitektur Mobilenet dari CNN. Berdasarkan hasil Training model yang diuji diperoleh model terbaik pada model dengan tingkat akurasi 91%, model menggunakan Hyperparameter NAdam. Dengan Epoch 50, batch 64, Akurasi 91%, Presisi 91%, Recall, 91%, dan F1-score 91%. A tire is a component in the form of a container made of rubber, wire, tires and several other chemicals which will be filled with air. Tires function as the main supporting role for the entire vehicle load, continuing the steering direction and maintaining steering stability, holding and transmitting engine power, and reducing vibrations in the vehicle. So a digital approach is needed to be able to identify whether a tire is suitable or not quickly, precisely, efficiently and accurately. So this research makes a study of tire damage in order to identify whether a tire is suitable or not quickly, precisely, efficiently and accurately. Hyperparameter Epoch 50, Batch 64, Optimizer (Adam, NAdam, RMSprop, SGD, Adadelta) and Learning Rate 0.001 with dasaet 1000 Tire images. The classification carried out in this research uses the Mobilenet architecture from CNN. Based on the training results of the tested model, the best model was obtained in a model with an accuracy level of 91%, the model used NAdam Hyperparameters. With Epoch 50, batch 64, Accuracy 91%, Precision 91%, Recall, 91%, and F1-score 91%. |
Description: | 58 Halaman |
URI: | https://repositori.uma.ac.id/handle/123456789/26176 |
Appears in Collections: | SP - Informatic Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
198160007 - Nurul Almadinah Pane - Fulltext.pdf | Cover, Abstract, Chapter I, II, III, V, Bibliography | 1.81 MB | Adobe PDF | View/Open |
198160007 - Nurul Almadinah Pane - Chapter IV.pdf Restricted Access | Chapter IV | 644.38 kB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.