EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-SA 4.0

Machine Learning Approach for Ibing Penca Stance Recognition Using Landmark Detection and Angle-Based Classification

Ratnadewi Ratnadewi, Agus Prijono, Aan Darmawan Hangkawidjaja, Sri Rustiyanti, Deri Al Badri

Journal of Applied Science, Engineering, Technology, and Education · 30.04.2026 · 8 (1) · 102-118

DOI: 10.35877/454ri.asci4539 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Applied Science, Engineering, Technology, and Education, online erschienen am 30.04.2026. Verfasst von Ratnadewi Ratnadewi und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The problem is how students can learn independently when no assistant teachers are present. The main objective of this research is to build an application system that can recognize the stances in Ibing Penca martial art. This research aims to help facilitate independent learning for martial arts practitioners, especially Ibing Penca, by developing a system that is able to recognize and classify the movements in 62 Ibing Penca stances. To achieve these goals, the research method used is to collect input data in the form of images or videos taken using an Orbbec camera. After the images are obtained, the next stage is data processing to detect important points on the body using landmark detection techniques. The next process is the identification of 33 keypoints on the body using the MediaPipe algorithm. From these keypoints, six important angles were calculated which included the right arm, left arm, right leg, left leg, right foot and left foot. This angle calculation is done using the angle method of the three relevant key points. The system is able to recognize the movements in Ibing Penca with a high degree of accuracy, which is very useful for learners who want to practice independently. The results of this study show that the system is able to classify 62 Ibing Penca moves with a success rate of 95.2% (58 moves), while the error rate is only 4.8% (4 moves). For future research, it is expected to develop this system by adding variations of movements and improving detection accuracy in more diverse environmental conditions.

Zitieren

APA

Ratnadewi, R., Prijono, A., Hangkawidjaja, A. D., Rustiyanti, S., & Badri, D. A. (2026). Machine Learning Approach for Ibing Penca Stance Recognition Using Landmark Detection and Angle-Based Classification. Journal of Applied Science, Engineering, Technology, and Education, 8(1), 102-118. https://doi.org/10.35877/454ri.asci4539

Vancouver

Ratnadewi R, Prijono A, Hangkawidjaja AD, Rustiyanti S, Badri DA. Machine Learning Approach for Ibing Penca Stance Recognition Using Landmark Detection and Angle-Based Classification. Journal of Applied Science, Engineering, Technology, and Education. 2026;8(1):102-118. doi:10.35877/454ri.asci4539

BibTeX
@article{Ratnadewi2026,
  author = {Ratnadewi, Ratnadewi and Prijono, Agus and Hangkawidjaja, Aan Darmawan and Rustiyanti, Sri and Badri, Deri Al},
  title = {Machine Learning Approach for Ibing Penca Stance Recognition Using Landmark Detection and Angle-Based Classification},
  journal = {Journal of Applied Science, Engineering, Technology, and Education},
  year = {2026},
  volume = {8},
  number = {1},
  pages = {102--118},
  doi = {10.35877/454ri.asci4539}
}
Kostenlos starten →