Work

Performance Analysis of Convolutional Neural Network in Pempek Food Image Classification with MobileNetV2 and GoogLeNet Architecture

Year: 2025

Type: article

Abstract: This research develops a pempek food image classification system using two Deep Learning architectures, namely MobileNetV2 and GoogLeNet. The dataset consists of five types of pempek with a total of 446 images, which are divided for training (70%), validation (15%), and testing (15%). The model was evaluated based on accuracy, precision, recall, and F1-score. The results showed that GoogLeNet achieved a validation accuracy of 96.21%, higher than MobileNetV2 which was only 70.58%. GoogLeNet is also more stable in convergence and more accurate in recognizing different types of pempek. This research shows that GoogLeNet is more optimal for pempek classification. In the future, this research can be extended by adding more datasets, exploring more sophisticated models, and developing mobile or web-based applications.

Source: Journal of Information Systems and Informatics

DOI: 10.51519/journalisi.v7i1.1026

Authors: Yudha Pratomo, Muhammad Taufik Roseno, Hadi Syaputra, Darius Antoni

Institutions: Universitas Sumatera Selatan, Universitas Indo Global Mandiri

Language: English

Topic: Advanced Neural Network Applications

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

SDG: Zero hunger

Open Access status: gold

Estimated APC paid: $32