Image Content Authentic Detection System using Convolutional Neural Network Method

Komang Indah, Ida Manuaba, Putu Prihatini

2021

Abstract

Digital content is often manipulated for a specific purpose, where image data is often falsified from the original data to provide information that is different from the original. In this study, an image detection system will be built through image classification techniques to detect image patterns that are manipulated using the Convolutional Neural Network (CNN) method. CNN is a development of Multilayer Perceptron (MLP) which is designed to process two-dimensional data. Each relationship between layers is carried out by linear operations with the existing weight values using linear convolution operations. This application serves to detect the authenticity of image content with the backpropagation process for accuracy and comparison with numbers from the training data set. The analysis of the research results produces an accuracy curve and loss validation, which states the classification of whether the image is original or has been modified. The application uses the Python programming language with Tensorflow objects to classify CNN images using two convolutional layers, one Max Pooling layer, one fully connected layer, and one output layer with softmax achieving 91.83% accuracy. Suggestions for system development, namely the use of metadata extraction with deep learning CNN can increase efficiency and reduce computational costs of the training dataset process.

Paper Citation

in Harvard Style

Indah K., Manuaba I. and Prihatini P. (2021). Image Content Authentic Detection System using Convolutional Neural Network Method. In Proceedings of the 4th International Conference on Applied Science and Technology on Engineering Science - Volume 1: iCAST-ES, ISBN 978-989-758-615-6, pages 969-975. DOI: 10.5220/0010957400003260

in Bibtex Style

@conference{icast-es21,
author={Komang Indah and Ida Manuaba and Putu Prihatini},
title={Image Content Authentic Detection System using Convolutional Neural Network Method},
booktitle={Proceedings of the 4th International Conference on Applied Science and Technology on Engineering Science - Volume 1: iCAST-ES,},
year={2021},
pages={969-975},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010957400003260},
isbn={978-989-758-615-6},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 4th International Conference on Applied Science and Technology on Engineering Science - Volume 1: iCAST-ES,
TI - Image Content Authentic Detection System using Convolutional Neural Network Method
SN - 978-989-758-615-6
AU - Indah K.
AU - Manuaba I.
AU - Prihatini P.
PY - 2021
SP - 969
EP - 975
DO - 10.5220/0010957400003260