https://scholar.dgist.ac.kr/handle/20.500.11750/203 2026-04-04T14:15:25Z https://scholar.dgist.ac.kr/handle/20.500.11750/11403 Title: An Adaptive Gain Dynamics for Time Delay Control Improves Accuracy and Robustness to Significant Payload Changes for Robots Author(s): Lee Junyoung; Chang, Pyung Hun; Jin Maolin Abstract: Time delay control (TDC) is a promising technique for robot manipulators because it is model-free, efficient, and yet, robust. Nevertheless, when a robot operates under significant payload changes, it is difficult to achieve satisfactory performance with a constant gain. To cope with this problem, several adaptive rules have already been proposed thus far, but they are less effective to significant payload changes, and parameter tuning procedures are too complicated. In this paper, we propose an adaptive gain dynamics that is more effective in payload changes and yet simpler to implement. Through simulations using a one-link arm and experiments using a whole arm manipulator with payload changes, the proposed dynamics was compared with the conventional TDC and two other existing methods. Simulation results show that the proposed algorithm can adapt to significant payload changes, achieving better accuracy than the conventional TDC. Experimental results show that the proposed method has consistently better adaptation capability than other methods, achieving significantly better accuracy. In addition, the proposed method is simpler to implement, having only one tuning parameter, whereas the existing methods require four or five such tuning parameters. 2020-03-31T15:00:00Z https://scholar.dgist.ac.kr/handle/20.500.11750/9459 Title: Secure Image-authentication Schemes with Hidden Double Random-phase Encoding Author(s): Yi, Faliu; Kim, Youhyun; Moon, Inkyu Abstract: We present a new image-authentication algorithm based on binary-quantified double random-phase encoding (DRPE) and a discrete cosine transformation (DCT) domain watermarking scheme. The image is encrypted using a DRPE scheme, in which only the phase part of the encoded image is preserved. Then, this phase image is quantified to become a binary image by giving 0 to these phase values that are less than 0 and setting others to 1. Then, the quantified binary image is secretly inserted into a host image with a DCT-domain watermarking algorithm. During image authentication, the receiver gets the binary image from the watermarked image using an inverse DCT operation and codes 0 values as -pi and values of 1 as pi to create a phase image. Finally, the input image is decoded from the retrieved phase image based on a double random phase decryption technique and is further authenticated using a nonlinear cross-correlation method. The present image-authentication algorithm offers an additional layer of system security because the hidden binary image reveals no information that is from the original image. Moreover, the image decrypted from the retrieved phase image cannot be easily recognized with naked eyes. However, it can be successfully authenticated by nonlinear cross-correlation, even in the face of attacks including noise attacks, filtering attacks, partial occlusion attacks, or geometric transformation attacks to the watermarked image. Our simulation results demonstrated the capability of the proposed image-authentication technique. 2018-10-31T15:00:00Z https://scholar.dgist.ac.kr/handle/20.500.11750/9379 Title: Perspective pinhole model with planar source for augmented reality surgical navigation based on C-arm imaging Author(s): Ha, Ho-Gun; Jeon, Sangseo; Lee, Seongpung; Choi, Hyunseok; Hong, Jaesung Abstract: Purpose: For augmented reality surgical navigation based on C-arm imaging, accuracy of the overlaid augmented reality onto the X-ray image is imperative. However, overlay displacement is generated when a conventional pinhole model describing a geometric relationship of a normal camera is adopted for C-arm calibration. Thus, a modified model for C-arm calibration is proposed to reduce this displacement, which is essential for accurate surgical navigation. Method: Based on the analysis of displacement pattern generated for three-dimensional objects, we assumed that displacement originated by moving the X-ray source position according to the depth. In the proposed method, X-ray source movement was modeled as variable intrinsic parameters and represented in the pinhole model by replacing the point source with a planar source. Results: The improvement which represents a reduced displacement was verified by comparing overlay accuracy for augmented reality surgical navigation between the conventional and proposed methods. The proposed method achieved more accurate overlay on the X-ray image in spatial position as well as depth of the object volume. Conclusion: We validated that intrinsic parameters that describe the source position were dependent on depth for a three-dimensional object and showed that displacement can be reduced and become independent of depth by using the proposed planar source model. © 2018, CARS. 2018-09-30T15:00:00Z https://scholar.dgist.ac.kr/handle/20.500.11750/9225 Title: Comparative study of hand-eye calibration methods for augmented reality using an endoscope Author(s): Lee, Seongpung; Lee, HyunKi; Choi, Hyunseok; Jeon, Sangseo; Ha, Hogun; Hong, Jaesung Abstract: We performed a comprehensive study of hand-eye calibration approaches for augmented reality (AR) using endoscopes, aiming to find the approaches that yield the best performance and to reveal the mechanism that makes these approaches successful. The two unknown values in this calibration problem are the hand-eye transformation between an endoscope and the endoscope-attached optotracked marker and the transformation between a calibration board with a checker pattern and the board-attached marker. We classify possible approaches to solving hand-eye transformation as direct, simultaneous, and sequential. The effect of the translation components of transformations on an approach's accuracy is theoretically analyzed using error equations derived from the approaches and demonstrated using both synthetic and real data. We found that sequential approaches performed the best when the magnitude of the translation of hand-eye transformation was larger than that between the board and its marker, which is the general case in implementing AR using endoscopes. In addition, this approach is less sensitive to noise and the number of calibration poses than others. Our results and analyses provide guidance for choosing an optimal hand-eye calibration solution for AR using endoscopes. © 2018 SPIE and IS and T. 2018-06-30T15:00:00Z