Klasifikasi Citra Pada Jenis Ikan Air Tawar Menggunakan Metode Convolutional Neural Network (CNN) dengan Arsitektur ResNet

Henbangesen, Abigael Allayd (2024) Klasifikasi Citra Pada Jenis Ikan Air Tawar Menggunakan Metode Convolutional Neural Network (CNN) dengan Arsitektur ResNet. Undergraduate thesis, Universitas Muhammadiyah Malang.

[thumbnail of PENDAHULUAN.pdf] Text
PENDAHULUAN.pdf
Restricted to Registered users only

Download (1MB) | Request a copy
[thumbnail of BAB I.pdf] Text
BAB I.pdf
Restricted to Registered users only

Download (286kB) | Request a copy
[thumbnail of BAB II.pdf] Text
BAB II.pdf
Restricted to Registered users only

Download (501kB) | Request a copy
[thumbnail of BAB III.pdf] Text
BAB III.pdf
Restricted to Registered users only

Download (675kB) | Request a copy
[thumbnail of BAB IV.pdf] Text
BAB IV.pdf
Restricted to Registered users only

Download (803kB) | Request a copy
[thumbnail of BAB V.pdf] Text
BAB V.pdf
Restricted to Registered users only

Download (154kB) | Request a copy
[thumbnail of POSTER.pdf] Text
POSTER.pdf
Restricted to Registered users only

Download (792kB) | Request a copy

Abstract

The classification of freshwater fish images is a challenging task in the field of pattern recognition and image processing. This research utilizes the Convolutional Neural Network (CNN) method with the ResNet50 architecture to distinguish various freshwater fish species. The ResNet50 architecture was chosen for its ability to extract complex features through 50 deep network layers, as well as its capacity to address accuracy degradation issues using shortcut connections that allow information to skip directly to deeper layers. The dataset used consists of 2,235 freshwater fish images, grouped into 6 categories: Bangus, Catfish, Goldfish, Gourami, Mullet, and Tilapia. The ResNet50 model was trained on this dataset and optimized through data augmentation techniques and hyperparameter tuning to improve performance. The model evaluation showed a high accuracy of 95.89%, with additional metrics such as precision, recall, and F1-score each reaching 96%. This CNN implementation with ResNet50 is expected to provide an effective tool for fish identification and classification, with potential applications in areas such as conservation, fisheries management, and biodiversity research.

Item Type: Thesis (Undergraduate)
Student ID: 201910370311089
Keywords: CNN, ResNet-50, Freshwater Fish, Image
Subjects: A General Works > AI Indexes (General)
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
T Technology > T Technology (General)
Divisions: Faculty of Engineering > Department of Informatics (55201)
Depositing User: 201910370311089 abi_al20
Date Deposited: 25 Oct 2024 09:09
Last Modified: 25 Oct 2024 09:09
URI: https://eprints.umm.ac.id/id/eprint/11759

Actions (login required)

View Item
View Item