Deteksi penyakit pada ikan menggunakan Algoritma Convolutional Neural Network (CNN)

Salsabila, Hilmi Ali (2025) Deteksi penyakit pada ikan menggunakan Algoritma Convolutional Neural Network (CNN). Sarjana thesis, UIN Sunan Gunung Djati Bandung.

[img]
Preview
Text (COVER)
1_cover.pdf

Download (187kB) | Preview
[img]
Preview
Text (ABSTRAK)
2_abstrak.pdf

Download (203kB) | Preview
[img]
Preview
Text (SK BEBAS PLAGIARISM)
3_skbebasplagiarism.pdf

Download (3MB) | Preview
[img]
Preview
Text (DAFTAR ISI)
4_daftarisi.pdf

Download (187kB) | Preview
[img]
Preview
Text (BAB I)
5_bab1.pdf

Download (246kB) | Preview
[img] Text (BAB II)
6_bab2.pdf
Restricted to Registered users only

Download (527kB) | Request a copy
[img] Text (BAB III)
7_bab3.pdf
Restricted to Registered users only

Download (340kB) | Request a copy
[img] Text (BAB IV)
8_bab4.pdf
Restricted to Registered users only

Download (716kB) | Request a copy
[img] Text (BAB V)
9_bab5.pdf
Restricted to Registered users only

Download (190kB) | Request a copy
[img] Text (DAFTAR PUSTAKA)
10_daftarpustaka.pdf
Restricted to Registered users only

Download (196kB) | Request a copy

Abstract

INDONESIA : Penyakit pada ikan merupakan salah satu penyebab utama menurunnya produktivitas dalam budidaya perikanan. Deteksi dini terhadap penyakit ikan sangat penting untuk mencegah penyebaran dan mengurangi kerugian ekonomi. Penelitian ini bertujuan untuk mengembangkan sistem deteksi penyakit ikan berbasis citra digital menggunakan algoritma Convolutional Neural Network (CNN) dengan pendekatan transfer learning pada arsitektur InceptionV3, menggunakan lapisan Mixed7 sebagai ekstraktor fitur utama. Dataset yang digunakan terdiri dari gambar ikan sehat dan gambar ikan yang terinfeksi enam jenis penyakit: White Tail Disease, Saprolegniasis, Aeromoniasis, Bacterial Gill Disease, Bacterial Red Disease, dan Parasitic Disease. Gambar dikumpulkan dari sumber terbuka dan diproses melalui tahapan pra-pemrosesan serta pembagian data untuk pelatihan dan pengujian. Model dikembangkan dengan memanfaatkan arsitektur InceptionV3 tanpa lapisan klasifikasi akhir dan dilengkapi dengan lapisan klasifikasi kustom. Hasil pelatihan menunjukkan bahwa model berhasil mencapai akurasi validasi hingga 96%, dan mampu mengklasifikasikan penyakit ikan dengan baik pada data asli dengan tingkat akurasi yang tinggi. Sistem ini memiliki potensi untuk diterapkan dalam deteksi dini penyakit ikan secara otomatis dan efisien di lingkungan nyata. ENGLISH : Fish diseases are one of the major factors contributing to reduced productivity in aquaculture. Early detection of fish diseases is essential to prevent outbreaks and minimize economic losses. This study aims to develop a fish disease detection system based on digital images using the Convolutional Neural Network (CNN) algorithm with a transfer learning approach, employing the InceptionV3 architecture and utilizing the Mixed7 layer as the main feature extractor. The dataset comprises images of healthy fish and fish infected with six types of diseases: White Tail Disease, Saprolegniasis, Aeromoniasis, Bacterial Gill Disease, Bacterial Red Disease, and Parasitic Disease. The images were sourced from open platforms and processed through preprocessing and data splitting for training and testing. The model was built by utilizing InceptionV3 without its top classification layers and enhanced with a custom classification head. The training results indicate that the model achieved up to 96% validation accuracy and was able to accurately classify fish diseases in real-world image data. This system shows strong potential for practical and efficient early disease detection in aquaculture environments.

Item Type: Thesis (Sarjana)
Uncontrolled Keywords: Deteksi penyakit ikan; Convolutional Neural Network (CNN); Transfer Learning; InceptionV3; Mixed7; Citra digital Keywords: Fish disease detection, Convolutional Neural Network (CNN), Transfer Learning, InceptionV3, Mixed7, Digital image.
Subjects: Data Processing, Computer Science > Computer Science Education
Special Computer Methods
Special Computer Methods > Artificial Intelligence
Technology, Applied Sciences
Divisions: Fakultas Sains dan Teknologi > Program Studi Teknik Informatika
Depositing User: Hilmi Ali Salsabila
Date Deposited: 08 Sep 2025 03:34
Last Modified: 08 Sep 2025 07:46
URI: https://digilib.uinsgd.ac.id/id/eprint/117788

Actions (login required)

View Item View Item