Integrasi Teknologi Deep Learning dalam Pengukuran Kebugaran Fisik Siswa Sekolah Menengah Atas

Butsiarah

Scripta Technica: Journal of Engineering and Applied Technology
Vol 1 No 2 (2025): December: Scripta Technica: Journal of Engineering and Applied Technology

This study aims to integrate deep learning technology to measure the physical fitness of high school students more accurately and efficiently than traditional manual methods. A total of 240 students from three schools participated in the assessment of five fitness components: cardiovascular endurance, muscular strength, flexibility, speed, and body mass index (BMI). A Convolutional Neural Network (CNN)-based system was employed to analyze students’ movement video data and evaluate their fitness levels. The results show that the deep learning model achieved an accuracy of 94.6% compared to manual assessments by professional trainers, while reducing evaluation time by 62% (from 25 minutes to 9.5 minutes per student) and improving inter-rater consistency from 0.71 to 0.93. Additionally, 87% of physical education teachers reported that the system was highly beneficial for assessment and documentation. These findings indicate that the integration of deep learning enhances the accuracy, efficiency, and objectivity of physical fitness evaluation and holds significant potential for broader application in technology-based physical education.