Al-Haq, Ahmad Khoir (2024) Analisis Forensik Kecelakaan Mobil Menggunakan MobileNet dan Optical Flow. Masters thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 6025211013-Master_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
6025211013-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 July 2026. Download (3MB) | Request a copy |
Abstract
Investigasi kecelakaan merupakan aspek yang efektif dalam merekonstruksi kasus kecelakaan mobil dengan memanfaatkan video yang direkam oleh pihak ketiga menggunakan closed-circuit television (CCTV). Video tersebut diproses menggunakan computer vision dan dianalisis untuk menentukan pelaku atau objek yang pergerakannya tidak normal atau anomali. Satu per satu frame dari CCTV diperiksa untuk mengidentifikasi anomali pergerakan objek mobil di jalan raya yang tertangkap kamera CCTV. Hal ini dilakukan untuk memastikan penyebab kecelakaan mobil tersebut. Dalam penelitian ini, diusulkan penggunaan model MobileNet untuk mengklasifikasikan gambar kecelakaan mobil yang diambil dari CCTV. Selain itu, teknik optical flow diterapkan untuk mendeteksi aliran dari pergerakan arah objek. Penelitian ini menggabungkan MobileNet dan metode optical flow untuk mendapatkan kinerja tinggi dalam klasifikasi kecelakaan mobil. Model MobileNetV2 pada kelas collision mencapai precission 97,27%, recall 98,94%, dan F1 score 98,10%. Selain itu, model optical flow menunjukkan nilai MAPE sebesar 1,82%.
=======================================================================================================================================
Accident investigation is an effective aspect of reconstructing car accident cases by utilizing video recorded by a third party using closed-circuit television (CCTV). The video is processed using computer vision and analyzed to determine the perpetrator or object whose movement is abnormal or anomalous. One by one, frames from CCTV are examined to identify anomalies in the movement of car objects on the highway that were caught on CCTV cameras. This is performed to ascertain the cause of the car accident. In this study, it is proposed to use the MobileNet model to classify car accident images extracted from CCTV. In addition, optical flow techniques are applied to detect flow in object direction movement. This study combines MobileNet and optical flow methods to obtain high performance in car accident classification. The MobileNetV2 model in the collision class achieves precision of 97.27%, recall of 98.94%, and F1 score of 98.10%. In addition, the optical flow model shows that the MAPE value is 1.82%.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | analisis forensik, kecelakaan mobil, MobileNet, optical flow; forensic analysis, car accidents, MobileNet, optical flow. |
| Subjects: | Q Science Q Science > Q Science (General) Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA76.9.A25 Computer security. Digital forensic. Data encryption (Computer science) Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55101-(S2) Master Thesis |
| Depositing User: | Ahmad Khoir Al-Haq |
| Date Deposited: | 06 Feb 2024 02:38 |
| Last Modified: | 06 Feb 2024 04:18 |
| URI: | http://repository.its.ac.id/id/eprint/106160 |
Actions (login required)
 |
View Item |