[16a-P07-1~8] 3.3 Information photonics and image engineering

[16a-P07-3] Reducing the Number of Mask Patterns in Single-Pixel Imaging Using Deep Learning

〇(B)Naoya Mukojima¹, Masaki Yasugi^1,2, Yasuhiro Mizutani³, Takeshi Yasui⁴, Hirotsugu Yamamoto^1,2 (1.Utsunomiya Univ., 2.JST, ACCEL, 3.Osaka Univ., 4.Tokushima Univ.)

Keywords:Single-pixel imaging, Deep-learning, Gesture recognition

Single-pixel imaging can be used for protecting privacy imaging in public spaces. However, it requires a large number of measurements to obtain a clear image. In order to reduce the number of measurements, deep learning has been proposed. In our research, we use this method to perform gesture recognition while considering privacy. In this study, we examined how much the number of mask patterns required for reconstruction can be reduced without affecting the accuracy of deep learning for real-time gesture detection.