ASIFT: A New Framework for Fully Affine Invariant Image Comparison
SIAM Journal on Imaging Sciences
An Efficient Direct Approach to Visual SLAM
IEEE Transactions on Robotics
Hi-index | 0.00 |
Wireless capsule endoscope (WCE) provides a noninvasive method to examine the entire gastrointestinal (GI) tract including small intestine, which other video endoscopic instruments cannot reach. Since the shape of small intestine is extremely complex and the length of small intestine varies from 5 to 9 meters, localization of the WCE inside the small intestine is very challenging. Traditional radio frequency (RF) localization techniques using the received signal strength (RSS) are not able to provide satisfactory location information of the capsule inside the small intestine. In this paper, we present a hybrid localization technique that takes advantage of data fusion from image sequence captured by the WCE's embedded camera and the RSS of the RF signal emitted by the capsule to enhance the positioning accuracy. The proposed method estimates the speed and direction of movement of the capsule by analyzing displacements of feature points between consecutive image frames and this motion information is integrated with RSS measurements by employing a Kalman filter to smooth the RF localization results. Performance of the proposed method is validated under a virtual testbed that emulates the transition of capsule inside small intestine against the traditional RSS-based RF localization.