Huda, Muhammad Fahmil (2015) Rancang Bangun Aplikasi Komputer untuk Manajemen Konten Multimedia Berbasis Webcam sebagai Sensor Gerak Menggunakan Metode Background Subtraction. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1110100704-Undergraduate Thesis.pdf - Published Version Download (2MB) | Preview |
Abstract
Era computer vision telah cukup lama dimulai. Berbagai algoritma telah dikembangkan sehingga memungkinkan komputer untuk melihat layaknya manusia. Penelitian ini mencoba menerapkan computer vision ke dalam aplikasi manajemen konten multimedia yang diintegrasikan dengan webcam sebagai sensor gerak. Metode yang digunakan adalah background subtraction. Penelitian dilakukan dengan sebuah laptop dengan dukungan webcam dan GPU. Aplikasi dibuat dengan Visual Studio 2012, librari OpenCV, dan CUDA Toolkit. Analisis gerak dilakukan melalui nilai standar deviasi dan kontur hasil background subtraction.. Nilai standar deviasi hasil background subtraction adalah antara 0 – 127,5. Dalam analisis kontur, pergerakan yang direspons aplikasi adalah jika dalam rentang jarak 80 piksel objek memiliki 5-6 gambar yang terekam oleh webcam. Dari hasil percobaan, webcam mampu mendeteksi pergerakan dengan kecepatan antara 0,63 m/s sampai 1,17 m/s dalam jarak 69 cm dari objek.
===================================================================================================
The era of Computer Vision has been started for a long time. Many algorithms have been developed for computer to has vision like human. This research was intended to apply computer vision to a multimedia content management application integrated with webcam as a motion sensor. The method used for this research is background subtraction which the background model is updated gradually. The research was conducted with a laptop supported by webcam and GPU. The application is developed by Visual Studio 2012, OpenCV library, and CUDA. The motion is analyzed by the standard deviation and contours of background subtraction results. The standard deviation varies from 0 – 127.5. In the contours analysis, motion will be responden by the application if in the range of 80 pixel there is object with 5-6 pictures captured by webcam. From the experiment, the webcam can detect motion with the velocity between 0.63 m/s to 1.17 m/s with the distance of 69 cm from the object.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSFi 006.42 Hud r |
| Uncontrolled Keywords: | Analisis Gerak, Background Subtraction, Computer Vision, CUDA, OpenCV. |
| Subjects: | Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) |
| Divisions: | Faculty of Mathematics and Science > Physics > 45201-(S1) Undergraduate Thesis |
| Depositing User: | Mr. Tondo Indra Nyata |
| Date Deposited: | 18 May 2018 04:07 |
| Last Modified: | 18 May 2018 04:07 |
| URI: | http://repository.its.ac.id/id/eprint/51906 |
Actions (login required)
 |
View Item |