CNN-based YOLOv3 comparison for underwater object detection / Mohamed Syazwan Asyraf …[et al.]

Asyraf, Mohamed Syazwan and Isa, Iza Sazanita and Marzuki, Mohd Ikhmal Fitri and Sulaiman, Siti Noraini and Hung, Chin Chang (2021) CNN-based YOLOv3 comparison for underwater object detection / Mohamed Syazwan Asyraf …[et al.]. Journal of Electrical and Electronic Systems Research (JEESR), 18. pp. 30-37. ISSN 1985-5389

Download

[thumbnail of 47324.pdf] Text
47324.pdf

Download (431kB)

Abstract

Object detection that deals with identifying and locating object is one of area that integrate from the advance- ment in machine learning and computer vision. Modern object detection which carried out supervised learning utilizes Convo- lutional Neural Network (CNN) as the backbone of the detection architecture which is significant for underwater object detection as the underwater images are usually low in quality and blurry. Single stage detection such as You Only Look Once (YOLO) is one the famous object detection model that is prominent among researchers due to high performance in accuracy and processing speed. However, YOLO has many versions where the current incremental improvement model of YOLOv3 has been widely used by researchers to solve different types of problem relatedto object detection. Therefore, there is a need to explore thetrade-off relationship between the processing speed and precisionof each YOLO model. In the study, two different open source underwater datasets were used in four different YOLOv3 modelsnamely as YOLOv3-SPP, YOLOv3-Tiny, YOLOv3-Tiny-PRN and the original YOLOv3 in order to study their performance based on metrics evaluation of precision and processing speed (FPS). The result shows that YOLOv3-SPP proved to be the best in terms of precision while YOLOv3-Tiny-PRN lead in terms of execution speed. So, this study shows that YOLOv3 model is highly significant to be implemented and able to accurately detect underwater objects with haze and low-light environment. This study can help researchers and industry in determining the best YOLOv3 model specifically for detection of the underwater images and its application.

Metadata

Item Type: Article
Creators:
Creators
Email
Asyraf, Mohamed Syazwan
UNSPECIFIED
Isa, Iza Sazanita
UNSPECIFIED
Marzuki, Mohd Ikhmal Fitri
UNSPECIFIED
Sulaiman, Siti Noraini
UNSPECIFIED
Hung, Chin Chang
UNSPECIFIED
Subjects: T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Electronics > Detectors. Sensors. Sensor networks
T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Electronics > Apparatus and materials > Detectors. Sensors. Sensor networks
Divisions: Universiti Teknologi MARA, Shah Alam > Faculty of Electrical Engineering
Journal or Publication Title: Journal of Electrical and Electronic Systems Research (JEESR)
UiTM Journal Collections: UiTM Journal > Journal of Electrical and Electronic Systems Research (JEESR)
ISSN: 1985-5389
Volume: 18
Page Range: pp. 30-37
Official URL: https://jeesr.uitm.edu.my
Item ID: 47324
Uncontrolled Keywords: CNN, YOLOv3, YOLOv3-SPP, YOLOv3-Tiny
URI: https://ir.uitm.edu.my/id/eprint/47324

ID Number

47324

Indexing


View in Google Scholar

Edit Item
Edit Item