Abstract
Abstract- The study of face recognition is one of the areas of computer vision that requires significant research at the moment. Numerous researchers have conducted studies on facial image recognition using a variety of techniques or methods to achieve the highest level of accuracy possible when recognizing a person's face from existing images. However, recognizing the image of a human face is not easy for a computer. As a result, several approaches were taken to resolve this issue. This study compares two (two) machine learning algorithms for facial image recognition to determine which algorithm has the highest level of accuracy, precision, recall, and AUC. The comparison is carried out in the following steps: image acquisition, preprocessing, feature extraction, face classification, training, and testing. Based on the stages and experiments conducted on public image datasets, it is concluded that the SVM algorithm, on average, has a higher level of accuracy, precision, and recall than the k-NN algorithm when the dataset proportion is 90:10. While the k-NN algorithm has the highest similarity in terms of accuracy, precision, and recall at 80%: 20% and 70%: 30% of 99.20. However, for the highest AUC percentage level, the k-NN algorithm outperforms SVM at a dataset proportion of 80%: 20% at 100%.
Highlights
The study of face recognition is one of the areas of computer vision that requires significant research at the moment
conducted studies on facial image recognition using a variety of techniques or methods to achieve the highest level of accuracy
machine learning algorithms for facial image recognition to determine which algorithm has the highest level of accuracy
Summary
Pengolahan citra ialah salah satu metode yang mentransformasikan citra input menjadi citra output untuk mempunyai mutu lebih bagus dari citra yang diinputkan [1]. Salah satu pendekatan yang dapat digunakan untuk mengidentifikasi wajah adalah Feature Extraction [3]. Penelitian ini bertujuan untuk mengklasifikasi pengenalan wajah seseorang berdasarkan perbandingan 2 (dua) buah algoirtma machine learning sehingga diketahui seberapa besar tingkat akurasi, presisi, recall serta AUC nya. Klasifikasi merupakan teknik yang akan digunakan untuk menentukan item dari suatu dataset ke dalam kategori atau kelas tertentu. Algoritma klasifikasi yang akan digunakan adalah k-Nearest Neighboard dan Support Vector Machine, alasan dipilihnya 2 (dua) algoritma tersebut adalah untuk menghasilkan perbandian performance terbaik dalam memprediksi target secara akurat dari setiap kasus yang terjadi dalam dataset [4]. Banyak penelitian yang membahas tentang face recognition ini, namun masih sedikit yang membahas tentang feature extraction untuk mengklasifikasi melalui teknik perbandingan performance terbaik yang dihasilkan dari 2 (dua) algoritma atau lebihv[5][6].
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Jurnal ICT : Information Communication & Technology
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
