Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/156707Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Sitanggang, Imas Sukaesih | - |
| dc.contributor.advisor | Hasibuan, Lailan Sahrina | - |
| dc.contributor.author | Dewi, Kartika | - |
| dc.date.accessioned | 2024-08-08T12:24:52Z | - |
| dc.date.available | 2024-08-08T12:24:52Z | - |
| dc.date.issued | 2024 | - |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/156707 | - |
| dc.description.abstract | Penyakit Mulut dan Kuku (PMK) pada sapi merupakan ancaman serius bagi industri peternakan di Indonesia yang dapat mempengaruhi kesehatan dan produktivitas ternak. Penelitian ini bertujuan mengembangkan aplikasi sederhana berbasis website untuk klasifikasi citra sapi sehat dan terinfeksi PMK menggunakan CNN dengan arsitektur VGG-16 dan MobileNet-V2. Dataset yang digunakan terdiri dari 197 citra, yaitu citra kelas sehat dan pmk. Proses penelitian meliputi pengumpulan dataset, pra-proses data, pembuatan model CNN, hyperparameter tuning, evaluasi model, dan implementasi sistem berbasis website. Model terbaik untuk arsitektur VGG-16 menggunakan optimizer Adam, learning rate 0,01, dan epoch 50, mencapai akurasi training 0,941, akurasi validasi 0,937, dan akurasi testing 0,905 dengan precision 0,846, recall 1,000, dan F1-score 0,917. Untuk MobileNet-V2, konfigurasi terbaik menggunakan optimizer Adam, learning rate 0,001, dan epoch 100, memberikan akurasi model pada training 0,988, akurasi model pada validasi 0,947, dan akurasi model pada testing 0,952 dengan precision 1,000, recall 0,910, dan F1-score 0,952. Penelitian ini telah berhasil membangun aplikasi sederhana berbasis web untuk klasifikasi citra sapi sehat dan PMK menggunakan framework Flask. Aplikasi ini bekerja berdasarkan sistem voting dari dua model, yaitu VGG-16 dan MobileNet-V2 yang memberikan hasil identifikasi yang sama jika kedua model menghasilkan output yang sama. | - |
| dc.description.abstract | Foot and Mouth Disease (FMD) in cow is a serious threat to the livestock industry in Indonesia that can affect the health and productivity of livestock. This research aims to develop a simple web-based application for image classification of healthy and FMD-infected cow using CNN with VGG-16 architecture and MobileNet-V2. The dataset used consists of 197 images, namely healthy and FMD class images. The research process includes dataset collection, data pre-processing, CNN model building, hyperparameter tuning, model evaluation, and website-based system implementation. The best model for VGG-16 architecture using Adam optimizer, learning rate 0.01, and epoch 50, achieved training accuracy 0.941, validation accuracy 0.937, and testing accuracy 0.905 with precision 0.846, recall 1.000, and F1-score 0.917. For MobileNet-V2, the best configuration uses Adam optimizer, learning rate 0.001, and epoch 100, giving model accuracy on training 0.988, model accuracy on validation 0.947, and model accuracy on testing 0.952 with precision 1.000, recall 0.910, and F1-score 0.952. This research has successfully built a simple web-based application for the classification of healthy and FMD cow images using the Flask framework. This application works based on a voting system of two models, namely VGG-16 and MobileNet-V2 which gives the same identification result if both models produce the same output. | - |
| dc.description.sponsorship | null | - |
| dc.language.iso | id | - |
| dc.publisher | IPB University | id |
| dc.title | Model Klasifikasi Citra Penyakit Mulut dan Kuku pada Sapi Menggunakan Convolutional Neural Network | id |
| dc.title.alternative | Image Classification Model of Foot and Mouth Disease in Cow Using Convolutional Neural Network | - |
| dc.type | Skripsi | - |
| dc.subject.keyword | klasifikasi citra | id |
| dc.subject.keyword | Convolutional Neural Network | id |
| dc.subject.keyword | MobileNet-V2 | id |
| dc.subject.keyword | Penyakit Mulut dan Kuku | id |
| dc.subject.keyword | VGG-16 | id |
| Appears in Collections: | UT - Computer Science | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| cover_G6401201040_57361bc8148b47d78e88194cb946a1b1.pdf | Cover | 1.24 MB | Adobe PDF | View/Open |
| fulltext_G6401201040_8b5f04904f9e4535b904f21cd931a95c.pdf Restricted Access | Fulltext | 910.15 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.