Gaussian Blur Filter Effect Analysis on Facial Detection Accuracy Using Viola Jones Method

Saryanto, Saryanto and Andriyani, Widyastuti (2024) Gaussian Blur Filter Effect Analysis on Facial Detection Accuracy Using Viola Jones Method. IJCCS (Indonesian Journal of Computing and Cybernetics Systems), 18 (3). ISSN 2460-7258

[img] Text (Jurnal: Gaussian Blur Filter Effect Analysis on Facial Detection Accuracy Using Viola Jones Method)
COVER TESIS JURNAL 2310090101TSD12 Saryanto.pdf
Download (1MB)

Abstract

Human face detection is one of the most studied topics in computer vision. The purpose of facial detection is to find out whether or not a face is present in an image. Blur can be caused by many things, such as motion that occurs when the camera takes a picture or the use of a camera that is not focused when taking a picture. For facial recognition, blur becomes difficult to get information about an object, get a description about it, or identify a face in the image. The more blur a picture, the more difficult it is to identify it. This research applies the Viola-Jones relative method for facial detection with a high degree of accuracy and fast computation. This study analyzed the influence of a gaussian blur filter by calculating how much radius an object has been given a gausian blur filter so that it can no longer be identified as an object, and also looking for the minimum PSNR value that is still acceptable in the object detection process. The minimum PSNR value for the image is 16.6 dB, and the minimum PSR value before the face can no longer be detected is 17.84 dB. Keywords: Gaussian Blur, Viola Jones, Object Detection, PSNR, Image Quality Measurement.

Item Type: Article
Additional Information: Link Jurnal: https://journal.ugm.ac.id/ijccs/article/view/96017
Uncontrolled Keywords: Gaussian Blur, Viola Jones, Object Detection, PSNR, Image Quality Measurement.
Subjects: A Karya Umum (General) > Ilmu Komputer (Computer Science) > Komputer Grafis (Computer Graphics)
Divisions: Jenjang S2 > Teknologi Informasi (S2)
Depositing User: Titis Pratiwi
Date Deposited: 01 Okt 2024 07:44
Last Modified: 01 Okt 2024 07:44
URI: http://eprints.utdi.ac.id/id/eprint/10522

Actions (login required)

View Item View Item
UTDI Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.