ALAMSYAH, TEDI (2023) KLASIFIKASI PEMAKAIAN SAFETY VEST MELALUI CITRA DIGITAL MENGGUNAKAN ALGORITMA CONVOLUTIONAL NEURAL NETWORK. S1 thesis, Universitas Mercu Buana - Menteng.
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (Cover)
41519110115-TEDI ALAMSYAH-01 Cover - tedi alamsyah.pdf Download (458kB) |
|
Text (Abstrak)
41519110115-TEDI ALAMSYAH-02 Abstrak - tedi alamsyah.pdf Download (140kB) |
|
Text (Bab 1)
41519110115-TEDI ALAMSYAH-03 Bab 1 - tedi alamsyah.pdf Restricted to Registered users only Download (151kB) |
|
Text (Bab 2)
41519110115-TEDI ALAMSYAH-04 Bab 2 - tedi alamsyah.pdf Restricted to Registered users only Download (216kB) |
|
Text (Bab 3)
41519110115-TEDI ALAMSYAH-05 Bab 3 - tedi alamsyah.pdf Restricted to Registered users only Download (187kB) |
|
Text (Bab 4)
41519110115-TEDI ALAMSYAH-06 Bab 4 - tedi alamsyah.pdf Restricted to Registered users only Download (962kB) |
|
Text (Bab 5)
41519110115-TEDI ALAMSYAH-07 Bab 5 - tedi alamsyah.pdf Restricted to Registered users only Download (128kB) |
|
Text (Daftar Pustaka)
41519110115-TEDI ALAMSYAH-08 Daftar Pustaka - tedi alamsyah.pdf Restricted to Registered users only Download (136kB) |
|
Text (Lampiran)
41519110115-TEDI ALAMSYAH-09 Lampiran - tedi alamsyah.pdf Restricted to Registered users only Download (4MB) |
|
Text (Pernyataan Keabsahahan dan Persetujuan Publikasi)
41519110115-TEDI ALAMSYAH-PUBLIKASI - tedi alamsyah.pdf Restricted to Repository staff only Download (220kB) |
Abstract
Convolutional neural network (CNN) merupakan metode yang terdapat pada deep learning yang diklaim sebagai metode terbaik atas permasalahan yang berkaitan dengan image classification. Perkembangan metode CNN saat ini telah menghasilkan banyak model atau arsitektur CNN terlatih yang dapat digunakan pada berbagai macam data, salah satunya data citra safety test. Penelitian ini menggunakan beberapa model arsitektur CNN yaitu Baseline CNN, VGG16, ResNet50, dan DenseNet201. Pertanyaan penelitian dalam skripsi ini adalah bagaimana mengimplementasikan model arsitektur Baseline CNN, VGG16, ResNet50, dan DenseNet201 dalam klasifikasi safety vest menggunakan citra digital serta mengetahui model mana yang memiliki tingkat keakuratan terbaik. Berdasarkan pembahasan dan hasil analisis yang telah dilakukan dalam penelitian ini, maka diperoleh accuracy dari setiap model arsitektur CNN. Baseline CNN memiliki accuracy 0.9764, VGG16 0.9843, ResNet50 0.8976 dan DenseNet201 0.9921. Dari penelitian ini disimpulkan bahwa model arsitektur yang memiliki tingkat accuracy terbaik adalah DenseNet201 0.9921, disusul VGG16 0.9843, Baseline CNN 0.9764, dan yang terakhir ResNet50 dengan nilai accuracy 0.8976. Convolutional neural network (CNN) is a method found in deep learning which is claimed to be the best method for problems related to image classification. The development of the CNN method today has produced many trained CNN models or architectures that can be used on a variety of data, one of which is safety test image data. This research uses several CNN architecture models, namely Baseline CNN, VGG16, ResNet50, and DenseNet201. The research question in this thesis is how to implement the Baseline CNN, VGG16, ResNet50, and DenseNet201 architecture models in safety vest classification using digital images and find out which model has the best accuracy. Based on the discussion and analysis results that have been carried out in this study, the accuracy of each CNN architecture model is obtained. Baseline CNN has an accuracy of 0.9764, VGG16 0.9843, ResNet50 0.8976 and DenseNet201 0.9921. From this research it is concluded that the architecture model that has the best accuracy level is DenseNet201 0.9921, followed by VGG16 0.9843, Baseline CNN 0.9764, and finally ResNet50 with an accuracy value of 0.8976.
| Item Type: | Thesis (S1) |
|---|---|
| NIM/NIDN Creators: | 41519110115 |
| Uncontrolled Keywords: | Klasifikasi, Safety Vest, algoritma CNN, Baseline CNN, VGG16, ResNet50, dan DenseNet201. Classification, Safety Vest, CNN algorithm, Baseline CNN, VGG16, ResNet50, and DenseNet201. |
| Subjects: | 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika |
| Divisions: | Fakultas Ilmu Komputer > Informatika |
| Depositing User: | PANCA LEGA SILABAN |
| Date Deposited: | 11 Sep 2023 03:55 |
| Last Modified: | 11 Sep 2023 03:55 |
| URI: | http://repository.mercubuana.ac.id/id/eprint/80648 |
Actions (login required)
 |
View Item |
