A Hybrid Camera System for High-Resolutionization of Target Objects in Omnidirectional Images

Abstract

Recent improvements in cameras and image processing techniques have advanced research on object detection and classification. The cameras capable of capturing omnidirectional images are advantageous for detecting surrounding objects, we focus on their usage method. Omnidirectional camera images generally have lower resolution than available cameras, leading to difficulty in identifying objects farther from the camera. To address this problem, we propose a hybrid camera system that first detects indistinct target regions during omnidirectional viewing, then uses a pan-tilt camera to capture a high-resolution image of the target. We also propose an algorithm for reducing the number of required complementary shots to realize efficient shooting.