Wildlife Detection using Motion History Information Captured by Camera Trap in the Dark - LINEヤフーの研究開発

Publications

ワークショップ (国際) Wildlife Detection using Motion History Information Captured by Camera Trap in the Dark

Shogo Yoshida (Nara Institute of Science and Technology), Yuki Matsuda (Okayama University/Nara Institute of Science and Technology), Kota Tsubouchi, Hirohiko Suwa (Nara Institute of Science and Technology), Keiichi Yasumoto (Nara Institute of Science and Technology)

27th International Conference on Distributed Computing and Networking (ICDCN 26)

2026.1.8

This paper proposes a method to analyze infrared camera trap images captured at night for wildlife monitoring. Infrared images are typically noisy, making daytime image analysis ineffective. Our approach extracts motion information from continuous frames to enhance classification accuracy. Experiments using real nocturnal data demonstrate that the proposed method outperforms existing models that analyze single images, achieving higher efficiency and accuracy in wildlife detection under low-light and noisy conditions.

Paper : Wildlife Detection using Motion History Information Captured by Camera Trap in the Dark 新しいタブまたはウィンドウで開く（外部サイト）