Raharja, Rahmad (2026) Rancang Bangun Aplikasi Mobile Untuk Analisis Kualitas Las Pada Midship Kapal Menggunakan Mikroskop Portabel dan Computer Vision. Bachelor thesis, Institut Teknologi Kalimantan.
This is the latest version of this item.
![[img]](http://repository.itk.ac.id/style/images/fileicons/text.png) |
Text
04221047_cover.pdf Restricted to Repository staff only until 4 October 2028. Download (404kB) | Request a copy |
|
Text
04221047_statement_of_authenticity.pdf Restricted to Repository staff only until 4 October 2028. Download (323kB) | Request a copy |
|
Text
04221047_publishing_agreement.pdf Restricted to Repository staff only until 4 October 2028. Download (315kB) | Request a copy |
|
Text
04221047_approval_sheet.pdf Restricted to Repository staff only until 4 October 2028. Download (332kB) | Request a copy |
|
Text
04221047_preface.pdf Restricted to Repository staff only until 4 October 2028. Download (703kB) | Request a copy |
|
Text
04221047_abstract_id.pdf Restricted to Repository staff only until 4 October 2028. Download (464kB) | Request a copy |
|
Text
04221047_abstract_en.pdf Restricted to Repository staff only until 4 October 2028. Download (464kB) | Request a copy |
|
Text
04221047_table_of_content.pdf Restricted to Repository staff only until 4 October 2028. Download (785kB) | Request a copy |
|
Text
04221047_illustrations.pdf Restricted to Repository staff only until 4 October 2028. Download (579kB) | Request a copy |
|
Text
04221047_tables.pdf Restricted to Repository staff only until 4 October 2028. Download (388kB) | Request a copy |
|
Text
04221047_notations.pdf Restricted to Repository staff only until 4 October 2028. Download (238kB) | Request a copy |
|
Text
04221047_chapter_1.pdf Restricted to Repository staff only until 4 October 2028. Download (2MB) | Request a copy |
|
Text
04221047_chapter_2.pdf Restricted to Repository staff only until 4 October 2028. Download (8MB) | Request a copy |
|
Text
04221047_chapter_3.pdf Restricted to Repository staff only until 4 October 2028. Download (8MB) | Request a copy |
|
Text
04221047_conclusions.pdf Restricted to Repository staff only until 4 October 2028. Download (1MB) | Request a copy |
|
Text
04221047_bibliography.pdf Restricted to Repository staff only until 4 October 2028. Download (1MB) | Request a copy |
|
Text
04221047_enclosure.pdf Restricted to Repository staff only until 4 October 2028. Download (8MB) | Request a copy |
|
Text
04221047_paper.pdf Restricted to Repository staff only until 4 October 2028. Download (598kB) | Request a copy |
|
Text
04221047_presentation.pdf Restricted to Repository staff only until 4 October 2028. Download (670kB) | Request a copy |
|
Text
04221069_Form. TA-020.pdf Restricted to Repository staff only until 4 October 2028. Download (232kB) | Request a copy |
|
Text
04221047_chapter_4.pdf Restricted to Repository staff only until 4 October 2028. Download (4MB) | Request a copy |
Abstract
Pemastian mutu (quality control) pengelasan midship kapal sangat krusial untuk mencegah risiko kecelakaan maritim. Namun, metode Non-Destructive Testing (NDT) konvensional seperti Liquid Penetrant Testing (PT) membutuhkan logistik kimia berbahaya, berisiko K3, dan memakan waktu lama (15–30 menit). Penelitian ini bertujuan membangun aplikasi mobile WeldInspector AI sebagai instrumen inspeksi visual permukaan otomatis berbasis computer vision yang terintegrasi dengan mikroskop portabel. Sistem dikembangkan menggunakan framework Flutter, modul OpenCV pada handphone, serta model Hybrid MobileNetV2-SVM yang dioptimasi dengan Squared Hinge Loss. Pelatihan menggunakan dataset seimbang 150 citra per kelas untuk Crack, Porosity, dan Good Welding. Hasil pengujian menunjukkan model Hybrid MobileNetV2-SVM mengungguli baseline MobileNetV2 Standar (Softmax) dengan Akurasi Global 86,33% (naik +9,66% dari model standar yang hanya 76,67%), rata-rata makro Precision 0,88, dan Recall makro 0,87. Model hibrida ini sukses mengunci Recall 1,00 pada kelas Crack (zero false negative) dan Precision 1,00 pada kelas Porosity (zero false positive). Pengujian lapangan membuktikan integrasi mikroskop portabel berbasis On-The�Go (OTG) berhasil mengeliminasi distorsi bayangan dan fluktuasi cahaya kamera biasa, dengan tingkat kepercayaan hingga 99,6%. Validasi skala Likert oleh Welding Inspector menghasilkan indeks keterterimaan 100% (Sangat Layak) dengan durasi inferensi lokal di bawah 2 detik. Disimpulkan bahwa WeldInspector AI sangat valid, efisien, dan andal untuk melengkapi ekosistem kontrol kualitas galangan kapal modern.
| Item Type: | Thesis (Bachelor) |
|---|---|
| Subjects: | A General Works > AI Indexes (General) |
| Divisions: | Jurusan Teknologi Industri dan Proses > Teknik Elektro |
| Depositing User: | Rahmad Rahmad Raharja |
| Date Deposited: | 16 Jul 2026 03:25 |
| Last Modified: | 16 Jul 2026 03:25 |
| URI: | http://repository.itk.ac.id/id/eprint/27295 |
Available Versions of this Item
-
Rancang Bangun Aplikasi Mobile Untuk Analisis Kualitas Las Pada Midship Kapal Menggunakan Mikroskop Portabel dan Computer Vision. (deposited UNSPECIFIED)
- Rancang Bangun Aplikasi Mobile Untuk Analisis Kualitas Las Pada Midship Kapal Menggunakan Mikroskop Portabel dan Computer Vision. (deposited 16 Jul 2026 03:25) [Currently Displayed]
Actions (login required)
 |
View Item |