MSR-YOLO: Method to Enhance Fish Detection and Tracking in Fish Farms

Tasks involving the monitoring of fish farms such as controlling fish ponds is one of the expensive and difficult tasks for fish farmers. Usually, fish farmers are doing these tasks manually which costs them time and money. We propose a system that automates the monitoring of the fish farm. This pap...

Full description

Saved in:
Bibliographic Details
Published in:Procedia computer science 2020, Vol.170, p.539-546
Main Authors: Mohamed, Hussam El-Din, Fadl, Ali, Anas, Omar, Wageeh, Youssef, ElMasry, Noha, Nabil, Ayman, Atia, Ayman
Format: Article
Language:English
Subjects:
Fish farming
Image enhancement
Object detection
Optical flow
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Tasks involving the monitoring of fish farms such as controlling fish ponds is one of the expensive and difficult tasks for fish farmers. Usually, fish farmers are doing these tasks manually which costs them time and money. We propose a system that automates the monitoring of the fish farm. This paper presents a technique to enhance the detection of fish and their trajectories in challenging water conditions. Firstly, we used image enhancement techniques to enhance unclear water images and to better identify fish. Then, we applied an object detection algorithm to detect fish. Finally, the detected objects’ coordinates are then used to extract features like count and trajectories. All experiments were done on our experimental setup. The technique showed promising results in regards to detection and tracking accuracy when applied.
ISSN:1877-0509
1877-0509
DOI:10.1016/j.procs.2020.03.123