:: ICCAS 2015 ::

TB04		[OS] Machine Vision Applications
Time : 13:00~14:30 Room : 104		Chair : Sungho Kim, Yeungnam University / (Young-Su Moon, Samsung Electronics, )



13:00~13:15	TB04-1

Image-based Guidance System for Intravascular Microrobot: Fiducial Marker-based Registration using Biplanar Fluoroscopy Images & CTA Images
Jinsik Oh, Jong-Oh Park, Sukho Park, Seong Young Ko(Chonnam National University, Korea, Republic of )

The ultimate goal of this work is to develop an image-based guidance system for a microrobot that treats vascular disease. In the preoperative stage, a CTA image is taken after fiducial markers are implanted. To obtain the position information of the microrobot, a biplanar fluoroscopic image is used. Fiducial marker-based registration is carried out to measure the relative position between CTA images and X-ray images. Result error of registration procedure was about 2 mm, and the image-based guidance system for the microrobot will be developed using the proposed algorithm.



13:15~13:30	TB04-2

A Sensor Fusion Method to Solve the Scale Ambiguity of Single Image by Combining IMU
onecue Kim(Pusan National University, Korea, Republic of ), Dong-Joong Kang(Pusan National Univ, Korea, Republic of )

This paper proposes a method for fusing IMU with a single camera to overcome the scale uncertainty problem of the SFM-based SLAM. The key notion of the proposed approach is to use distance (translational) information for a robot estimated from an IMU sensor to correct (or compensate for) the translational information, which has universal scale ambiguity from the SFM and to complete the projection matrices that will be used to estimate the scale. To enhance the accuracy of the scale estimation, the back-projection and verification method is used.



13:30~13:45	TB04-3

Improvement of traffic sign recognition by accurate ROI refinement
Sungho Kim(Yeungnam University, Korea, Republic of ), Soon Kwon(DGIST, Korea, Republic of )

This paper presents the accuracy effect of the detected traffic sign region to the traffic sign recognition (TSR) and an improved TSR method is proposed by the accurate traffic sign region extraction. The conventional HOG-based traffic sign detection (TSD) shows limited localization accuracy. Inaccurate traffic sign region affect to the TSR performance directly. For the specific TSR, speed limit signs are used to analyze the effect of localization error. Enhanced color channel based traffic sign extraction in TSR learning and testing stage validates the upgraded TSR performance.



13:45~14:00	TB04-4

Region-based Pringting Quality enhancement
Hyun-Seung Lee, Ji-young Yi, YoungSu Moon(Samsung Electronics (SEC), Korea, Republic of )

This paper presents a new region-adaptive printing quality enhancement algorithm for mixed scanning documents. To adapt a variety of line-per-inch (LPI) printing documents, novel pixel-wise LPI estimation is done via hadamard transform-based local frequency components analysis. After, LPI region adaptive-halftone-noise smoother and least square optimal restoration filter enhance the quality of scanned documents. Experimental results indicate that the proposed one provides outperforms other state of the art noise reduction methods on a variety of LPI documents.



14:00~14:15	TB04-5

Indoor Localization of Quadropter using Vision based Sensor fusion
Wangheon Lee(Hansei University, Korea, Republic of ), Byoung kyun Kim, Hyun gyu Kang(Hansei University of Korea, Korea, Republic of )

Usually, the DRONEs (Quadroter) are equipped by GPS sensor and Rate Gyro so as to localize their localization in executing their mission. However, the GPS has an error more than 10 meters, furthermore, in case of the DRONE used in indoor , the GPS is not useless. In this paper , we propose localization of the DRONE used in indoor by fusing the vision information from the WEB camera with Gyro data compensated by KALMAN Filter. We will confirm the usefulness of the algorithm by real experimentation as well as theoretical analysis.