Similarity-AF-Face Recognition Using Machine Learning Algorithm Based on Raspberry Pi 4b

Sunardi, Sunardi and Fadlil, Abdul and Prayogi, Denis Similarity-AF-Face Recognition Using Machine Learning Algorithm Based on Raspberry Pi 4b.

Text Similarity-AF-Face Recognition Using Machine Learning Algorithm.pdf Download (3MB)

Abstract

Machine learning is one of artificial intelligence that is used to solve various problems, one of which is classification. Classification can separate a set of objects based on certain characteristics. The face is a part of the body with unique characteristics that can distinguish one person from another. For humans, recognizing someone from the face is easy, but for computers, it requires complex algorithms to solve this
problem. In this paper, we propose a face image classification using a machine learning algorithm Support Vector Machine (SVM) with Principal Component Analysis (PCA) feature extraction which is implemented on a room security device with all processing using raspberry pi 4b. The dataset used is facial images collected from 15 employee respondents with 2100 training data and 525 test data. Image data is taken from the face with various pose variants with both eyes, nose, and ears visible. Training using raspberry pi 4b resulted in a model with the best score of 99% accuracy in 0.10 seconds, while
testing of 525 data resulted in a model with a 99% precision score, 99% recall, and 99% f1 score. Testing the facial recognition device using the raspberry pi 4b with the SVM model can recognize facial images in real-time on the webcam and the sensors installed on the raspberry pi work according to their functions. The test results show that SVM can be applied properly to facial recognition devices as long as the facial features are still clearly visible.

Item Type:	Artikel Umum
Subjects:	T Technology > T Technology (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisi / Prodi:	Master (Magister) > Master of Technology Informatica (Magister Teknologi Informatika)
Depositing User:	Drs. Abdul Fadlil, M.T., Ph.D.
Date Deposited:	22 Aug 2022 01:49
Last Modified:	22 Aug 2022 01:50
URI:	http://eprints.uad.ac.id/id/eprint/36391

Actions (login required)

View Item