Negara, Pandito Hudiarso Abdul Rahim Setia (2020) Deteksi Kendaraan Berbasis Video Menggunakan Video Deep Learning. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
05111540000060-Undergraduate_Thesis.pdf Download (2MB) | Preview |
Abstract
Pendeteksian kendaraan sangat identic dengan analisis sebuah rekaman. Banyak perusahaan, badan riset dan universitas yang terus mengembangkan machine learning agar mendapat hasil yang lebih akurat dan cepat. Dari situlah lahir algoritma deep learning. Mask Region-Convolutional Neural Network (CNN) adalah salah satu deep neural network yang cocok digunakan untuk mengolah data yang berbentuk 2 dimensi, seperti rekaman. Data gambar pada rekaman diproses menjadi features map. Selanjutnya, features map diproses sehingga menghasilkan label dan bounding box. Terakhir, masking dilakukan terhadap daerah objek. Pada tugas akhir ini, dilakukan implementasi Mask R-CNN untuk melakukan pendeteksian kendaraan pada data video. Data uji merupakan data rekaman yang diambil manual. Data latih diambil dari model pre-trained MS-COCO yang memiliki gambar lebih dari 80.000 gambar. Hasil uji coba optimal didapatkan dari arsitektur Mask R-CNN nya dengan nilai akurasi 79,79%.
=================================================================================================================================
Vehicle detection is very identic with video analysis. Many companies, researchers, and universities keep developing this method to get fastest and most accurate result. From that demand, deep learning was created. Mask Region-Convolutional Neural Network (Mask R-CNN) is one of deep neural network that satisfy 2 dimentional data type processing, such as video. Frames from video will be processed into features map. After that, features map is processed to produce labels and bounding boxes. Last step, masking is done on the object area In this final project, Mask R-CNN implementation performed vehicle detection on video data. Testing Data video is taken manually. Training Data is taken from MS-COCO pre-trained model, having more than 80.000 pictures in it. Optimal testing Result is taken from modified parameter of the method with accuracy 79,79%.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSIf 006.42 Neg d-1 2020 |
| Uncontrolled Keywords: | Kata kunci: deteksi kendaraan, machine learning, deep learning, mask region-convolutional neural network. |
| Subjects: | T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing T Technology > TA Engineering (General). Civil engineering (General) > TA1573 Detectors. Sensors T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques. Image analysis--Data processing. |
| Divisions: | Faculty of Information Technology > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Pandito Hudiarso Abdul Rahim Setia Negara |
| Date Deposited: | 11 Mar 2025 02:09 |
| Last Modified: | 11 Mar 2025 02:09 |
| URI: | http://repository.its.ac.id/id/eprint/74025 |
Actions (login required)
 |
View Item |