Solicha, Rachma Aji (2019) Perancangan Sistem Stabilisasi Video sebagai Fitur Pendukung pada Kamera UAV. Diploma thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
10311600000006_Undergraduate_Theses.pdf Download (2MB) | Preview |
Abstract
Video yang direkam pada UAV seringkali mengalami efek getaran karena gerakan kamera yang tidak diinginkan. Kondisi video yang tidak stabil dapat mempersulit pengguna dalam mengamati target tertentu. Oleh karena itu, pada penelitian ini dirancang sebuah sistem stabilisasi video berbasis digital image processing untuk meredam efek getaran dalam video. Sistem stabilisasi video terdiri dari empat proses utama. Pertama, proses deteksi dan pelacakan feature point antara dua frame video yang berurutan untuk mendapatkan jalur pergerakan kamera. Proses ini menerapkan metode Shi-Tomasi corner detector dan Optical flow matching. Kedua, proses estimasi gerak dengan pendekatan affine transform. Ketiga, konsep simple low pass filter diterapkan untuk menentukan nilai kompensasi pergerakan objek dalam video. Terakhir, data frame video ditransformasi berdasarkan nilai kompensasi untuk meredam efek gerak getar. Hasil pengujian terhadap penerapan sistem stabilisasi video menunjukkan peningkatan kualitas video dengan spesifikasi yakni penurunan level SAD sebesar 1,59 dan MSE sebesar 176,11, serta peningkatan level SSIM sebesar 0,019. Kecepatan proses pengolahan gambar mencapai 21 FPS (frame per second).
================================================================================================================================
Videos recorded on UAVs often suffer vibration effects due to unwanted camera movements. Unstable videos make it difficult for users to observe certain targets. Therefore, this research designed a video stabilization system using image processing to reduce vibration effects in a video. Video stabilization system consists of four main processes. Firstly, a feature point detection and tracking process between two consecutive frames to get the camera movement path. This process applied Shi-Tomasi corner detector and Optical flow matching. Secondly, motion estimation is performed based on affine transform. Third, the simple low pass filter is applied to determine a motion compensation for the object movement in the video. Lastly, video frames is transformed based on compensation value to reduce the vibration effect. The test results on the implementation of the video stabilization system show video quality improvement based on a decrease in SAD level of 1,59 and MSE of 176,11, and an increase in the SSIM level of 0,019. Speed of the image processing reach 21 FPS (frame persecond).
| Item Type: | Thesis (Diploma) |
|---|---|
| Additional Information: | RSEO 621.367 Sol p-1 2019 |
| Uncontrolled Keywords: | Unmanned Aerial Vehicle (UAV), Feature point, Shi-Tomasi corner detector, Optical flow matching, Simple low pass filter |
| Subjects: | T Technology > T Technology (General) > T57.5 Data Processing T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TR Photography > TR260.7 Autofocus cameras T Technology > TR Photography > TR845 Cinematography. Video recordings. |
| Divisions: | Faculty of Vocational > 36304-Automation Electronic Engineering |
| Depositing User: | Rachma Aji Solicha |
| Date Deposited: | 20 Jun 2023 01:38 |
| Last Modified: | 20 Jun 2023 01:38 |
| URI: | http://repository.its.ac.id/id/eprint/70655 |
Actions (login required)
 |
View Item |