IMPLEMENTASI BACKGROUND SUBTRACTION DAN CONNECTED COMPONENT LABELING DALAM MEMPREDIKSI JUMLAH IKAN LELE

This thesis aims to implement Background Subtraction and Connected Component Labeling techniques in predicting the number of seeds in an aquarium. The Background Subtraction method is used to remove the background on the image, while Connected Component Labeling is used for identifying and calculati...

Full description

Saved in:
Bibliographic Details
Main Author: Mochammad Farqan Ramadhan, (Author)
Format: Book
Published: Universitas Pembangunan Nasional Veteran Jakarta, 2023-07-06.
Subjects:
Online Access:Link Metadata
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This thesis aims to implement Background Subtraction and Connected Component Labeling techniques in predicting the number of seeds in an aquarium. The Background Subtraction method is used to remove the background on the image, while Connected Component Labeling is used for identifying and calculating the number of lele seeds detected. The research involved taking digital images from aquariums using cameras and applying pre-processing techniques to remove backgrounds and improve the quality of digital images. Next, by applying the Connected Component Labeling method, the number of seeds detected will be calculated and predicted. The test results showed that the accuracy of detection of the number of seeds by using the Connected Component Labeling (CCL) technique varied depending on the amount of seed available. For the number of lily seeds less than or equal to 10, an average accuracy of 100% is obtained. However, on the number of lily seeds of 25, the average accuracy was below 76%. As for the number of seeds of lele fish of 50, the average accuracy ranges between 46% - 50%. This research has contributed to the development of monitoring and management systems for fish populations in aquariums. Although there are variations in detection accuracy depending on the number of seeds, the implementation of Background Subtraction and Connected Component Labeling techniques opens up opportunities to further improve the predictions of seed numbers by improving the methods and techniques used.
Item Description:http://repository.upnvj.ac.id/25661/9/ABSTRAK.pdf
http://repository.upnvj.ac.id/25661/19/AWAL.pdf
http://repository.upnvj.ac.id/25661/18/BAB%20I.pdf
http://repository.upnvj.ac.id/25661/20/BAB%20II%20.pdf
http://repository.upnvj.ac.id/25661/21/BAB%20III.pdf
http://repository.upnvj.ac.id/25661/22/BAB%20IV.pdf
http://repository.upnvj.ac.id/25661/23/BAB%20V%20SIMPULAN%20DAN%20SARAN.pdf
http://repository.upnvj.ac.id/25661/24/DAFTAR%20PUSTAKA.pdf
http://repository.upnvj.ac.id/25661/14/RIWAYAT%20HIDUP.pdf
http://repository.upnvj.ac.id/25661/15/LAMPIRAN.pdf
http://repository.upnvj.ac.id/25661/10/Hasil%20Plagiarisme.pdf
http://repository.upnvj.ac.id/25661/13/ARTIKEL%20KI.pdf