PERFORMANCE ANALYSIS OF SSD ALGORITHM FOR SOCIAL DISTANCE VIOLATION

HENDRAWAN, VIVIAN DAVINA (2023) PERFORMANCE ANALYSIS OF SSD ALGORITHM FOR SOCIAL DISTANCE VIOLATION. Other thesis, Universitas Katholik Soegijapranata Semarang.

[img] Text
19.K1.0011-VIVIAN DAVINA HENDRAWAN-COVER_a.pdf
Download (573kB)
[img] Text
19.K1.0011-VIVIAN DAVINA HENDRAWAN-BAB I_a.pdf
Download (117kB)
[img] Text
19.K1.0011-VIVIAN DAVINA HENDRAWAN-BAB II_a.pdf
Restricted to Registered users only
Download (202kB)
[img] Text
19.K1.0011-VIVIAN DAVINA HENDRAWAN-BAB III_a.pdf
Download (190kB)
[img] Text
19.K1.0011-VIVIAN DAVINA HENDRAWAN-BAB IV_a.pdf
Download (430kB)
[img] Text
19.K1.0011-VIVIAN DAVINA HENDRAWAN-BAB V_a.pdf
Download (662kB)
[img] Text
19.K1.0011-VIVIAN DAVINA HENDRAWAN-BAB VI_a.pdf
Download (177kB)
[img] Text
19.K1.0011-VIVIAN DAVINA HENDRAWAN-DAPUS_a.pdf
Download (197kB)
[img] Text
19.K1.0011-VIVIAN DAVINA HENDRAWAN-LAMP_a.pdf
Download (229kB)

Abstract

The increase in the number of cases of transmission of Covid-19 has been a problem for many sectors. Many solutions have been attempted to prevent transmission. One of the solutions is to utilize computer vision technology to detect violations of health protocols such as social distancing violations. The use of computer vision to detect social distancing violations in this study was carried out by utilizing the NVIDIA Jetson Nano tool with the pre-trained SSD MobileNet V2 model obtained from the Tensorflow model zoo. The testing was carried out with the camera being parallel to objects within a room lighting and to evaluate the performance of the proposed system, the researcher used confusion matrix calculations. Based on research results, the utilization of the NVIDIA Jetson Nano with SSD MobileNet V2 model for the detection system gives an accuracy value of 88%, a precision value of 100%, and a recall value of 87%.

Item Type: Thesis (Other)
Subjects: 000 Computer Science, Information and General Works > 004 Data processing & computer science
Divisions: Faculty of Computer Science > Department of Informatics Engineering
Depositing User: mr AM. Pudja Adjie Sudoso
Date Deposited: 05 Apr 2023 01:25
Last Modified: 05 Apr 2023 01:25
URI: http://repository.unika.ac.id/id/eprint/31408

Actions (login required)

View Item View Item
Unika 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.