ATIKAH AULIA PUTRI, . (2021) DETEKSI HELM KESELAMATAN BERBASIS CITRA MENGGUNAKAN AUTOENCODER DAN SUPPORT VECTOR MACHINE DIBANDINGKAN DENGAN GENERALIZED HOUGH TRANSFORM. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.
![[img]](http://repository.unj.ac.id/style/images/fileicons/text.png) |
Text
BAB 4.pdf Restricted to Registered users only Download (1MB) |
|
Text
BAB 2.pdf Restricted to Registered users only Download (1MB) |
|
Text
BAB 5.pdf Restricted to Registered users only Download (1MB) |
|
Text
BAB 3.pdf Restricted to Registered users only Download (2MB) |
|
Text
DAFTAR PUSTAKA.pdf Download (1MB) |
|
Text
Surat Pernyataan Publikasi.pdf Download (454kB) |
|
Text
COVER.pdf Download (1MB) |
|
Text
BAB 1.pdf Download (1MB) |
|
Text
LAMPIRAN.pdf Restricted to Registered users only Download (1MB) |
Abstract
Terjadi hampir 393 luka mematikan yang disebabkan oleh peralatan yang jatuh di tempat kerja, namun deteksi dan monitoring manual memiliki beberapa kelemahan karena tidak dapat dilakukan secara real-time. Tujuan dari penelitian ini adalah untuk mengidentifikasi apakah pekerja memakai helm keselamatan kerja (hardhat) dengan computer vision. Deteksi penggunaan helm keselamatan kerja atau hardhat terdiri dari dua langkah: ekstraksi fitur dan klasifikasi helm keselamatan kerja atau hardhat. Ekstraksi fitur diimplementasikan menggunakan metode berbasis pengolahan citra Canny edge detection dan metode autoencoder yang berbasis neural network. Klasifikasi hardhat diterapkan dengan generalized hough transform (GHT) dan support vector machine (SVM). Hasil percobaan menunjukkan autoencoder efektif untuk pengurangan dimensi dan keluaran autoencoder tidak cocok jika digunakan sebagai masukan generalized hough transform. Almost 393 deadly injuries led by exposure to falling equipment at workplaces, however manual monitoring has several downsides as it could not be real-time. The purpose of this study is to identify whether the workers are wearing safety hardhat by computer vision. The hardhat wearing detection consists of two steps: feature extraction and hardhat classification. Feature extraction is implemented using both image processing-based method Canny edge detection and neural network-based method autoencoder. Hardhat classification is applied by generalized hough transform (GHT) and support vector machine (SVM). Experimental results illustrated the robustness and effectiveness of autoencoder for dimensionality reduction and autoencoder output is not compatible with generalized hough transform
| Item Type: | Thesis (Sarjana) |
|---|---|
| Additional Information: | 1). Med Irzal, M.Kom ; 2). Muhammad Eka Suryana, M.Kom |
| Subjects: | Sains > Matematika > Ilmu Komputer |
| Divisions: | FMIPA > S1 Ilmu Komputer |
| Depositing User: | Users 14629 not found. |
| Date Deposited: | 15 Jul 2022 00:59 |
| Last Modified: | 15 Jul 2022 00:59 |
| URI: | http://repository.unj.ac.id/id/eprint/31794 |
Actions (login required)
 |
View Item |
