Detection of Violent Elements in Digital Games Using Deep Learning

Yalçın N., Çapanoğlu A. E.

SN Computer Science, cilt.4, sa.5, 2023 (Scopus) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 4 Sayı: 5
  • Basım Tarihi: 2023
  • Doi Numarası: 10.1007/s42979-023-02064-w
  • Dergi Adı: SN Computer Science
  • Derginin Tarandığı İndeksler: Scopus
  • Anahtar Kelimeler: CNN, Deep learning, Violation in digital games, YOLOv4 Tiny
  • Gazi Üniversitesi Adresli: Evet

Özet

Detection of violence in digital games can be achieved by detecting elements of violence in digital game frames. For young, the detection of these elements can play an important role in the detection and prevention of violence by taking early actions. Violence detection from within the image is based on the analysis of images in periods. This means a large amount of computation based on image processing. In this study, an image detection algorithms based on CNN models are proposed using YOLOV4 Tiny advanced object detection algorithms. It has been shown that the proposed method can be used for real time games with detection speed and 93% accuracy performance.