Convolutional Neural Network Pada Klasifikasi Daging Sapi Dan Daging Babi Menggunakan Resnet-50 Berbasis Website

Fajar, Maulana Thaariq Alburuuj (2024) Convolutional Neural Network Pada Klasifikasi Daging Sapi Dan Daging Babi Menggunakan Resnet-50 Berbasis Website. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

[img] Text
Cover.pdf

Download (1MB)
[img] Text
Abstrak.pdf

Download (785kB)
[img] Text
Abstract.pdf

Download (783kB)
[img] Text
BAB 1.pdf

Download (870kB)
[img] Text
BAB 2.pdf

Download (1MB)
[img] Text
BAB 3.pdf

Download (882kB)
[img] Text
BAB 4.pdf
Restricted to Registered users only

Download (1MB)
[img] Text
BAB 5.pdf

Download (785kB)
[img] Text
Daftar Pustaka.pdf

Download (912kB)
[img] Text
Lampiran.pdf
Restricted to Registered users only

Download (1MB)

Abstract

The identification of beef and pork is an important challenge in the food industry and the provision of halal meat products. This research aims to develop a classification system that can distinguish between beef and pork using Convolutional Neural Network (CNN) technology and ResNet-50 architecture. The need for this system arises due to the general difficulty faced by consumers in distinguishing between these two types of meat based on their physical appearance, especially in terms of color and texture. The use of CNN with ResNet-50 architecture was chosen based on its effectiveness in previous studies, which showed high capability in image classification. Through complex feature analysis and deep neural network modeling, this method can obtain an effective visual representation for distinguishing between the two types of meat. . The results showed that the model with a data composition of 80:10:10 produced the most optimal performance of the 3 data composition division scenarios, by achieving a Validation Accuracy of 100%, a Validation Loss value of 0.0057 and a model evaluation test value using Confusion Matrix of 85%. The implementation of the model in the form of a website using the Flask framework facilitates access for general users. This method also has the potential to be implemented in the food industry and ensure compliance with people's preferences and beliefs related to the type of meat consumed. Keywords : Convolutional Neural Network, Flask, ResNet-50

Item Type: Thesis (Undergraduate Thesis)
Subjects: T Technology > T Technology (General)
Divisions: Faculty of Informatics > Software Engineering
Depositing User: repository staff
Date Deposited: 18 Sep 2024 08:24
Last Modified: 18 Sep 2024 08:24
URI: http://repository.ittelkom-pwt.ac.id/id/eprint/11341

Actions (login required)

View Item View Item