ABSTRAKPengenalan gerakan tangan dianggap sebagai bagian penting dari interaksi manusia komputer, memungkinkan komputer untuk mengenali dan menafsirkan gerakan tangan dan menjalankan perintah. Penggunaan machine learning dimanfaatkan untuk mencari tren dan pola yang berbeda. Namun, tantangan untuk menerapkan machine learning menjadi bagaimana memilih di antara berbagai model berbeda digunakan untuk kumpulan data atau kasus berbeda. Tujuan dari penelitian ini adalah mengukur kinerja model machine learning yang diusulkan dengan pemilihan hyperparameter yang sesuai dalam mengenali 10 pola angka berdasarkan gerakan tangan di udara. Dalam makalah ini, model KNN, SVM, dan ANN-PSO diusulkan. Eksperimen dilakukan dengan mengumpulkan data gerakan yang berasal dari MPU-6050. Kinerja metode yang diusulkan dievaluasi menggunakan metrik standar seperti akurasi klasifikasi, presisi, recall, f1-score, dan AUC-ROC. Hasilnya menunjukkan bahwa akurasi rata-rata mencapai 87%.Kata kunci: HCI, hand gesture recognition, machine learning, MPU-6050, pola ABSTRACTHand gesture recognition is considered an essential part of human-computer interaction (HCI), enabling computers to recognize and interpret hand gesturesand execute commands. The use of machine learning is utilized to look for different trends and patterns. However, the challenge for implementing machine learning becomes how to choose between different models used for different datasets or cases. This research aims to measure the performance of the proposed machine learning model by selecting the appropriate hyperparameters in recognizing 10 number patterns based on hand gestures in the air. In this paper, KNN, SVM, and ANN-PSO models are proposed. Experiments were carried by collecting gesture data from MPU-6050. The performance of the proposed method was evaluated using standard metrics such as classification accuracy, precision, recall, f1-score, and AUC-ROC. The results show that the average accuracy reaches 87%.Keywords: HCI, hand gesture recognition, machine learning, MPU-6050, pattern
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.