Hybrid Tracking and Matching Algorithm for Mosaicking Multiple Surgical Views

In recent years, laparoscopic surgery has become major surgery due to several advantages for patients. However, it has disadvantages for operators because of the narrow surgical field of view. To solve this problem, our group proposed camera-retractable trocar which can obtain multiple surgical viewpoints while maintaining the minimally invasiveness. The purpose of this study is to obtain a wide visual panoramic view by utilizing image mosaicking of camera-retractable trocar viewpoints videos. We utilize feature points tracking in different videos to generate panoramic video independent of inter-cameras overlap and to increase mosaicking speed and robustness. We evaluate tracking accuracy according to several conditions and mosaicking accuracy according to overlap size. In contrast to the conventional mosaicking approach, the proposed approach can produce panoramic image even in the case of 0% inter-cameras overlap. Additionally, the proposed approach is fast enough for clinical use.