:: ICCAS 2015 ::

WC07		[OS] Statistical Inference and Data Mining
Time : 16:30~18:00 Room : 202		Chair : Yohei Saika (National Institute of Technology, Gunma College, )



16:30~16:45	WC07-1

Monte Carlo Simulation for Phase Unwrapping Using Super-resolution for Remote Sensing
Yohei Saika(National Institute of Technology, Gunma College, Japan)

We investigate the problem of phase unwrapping using multiple interferograms on the basis of Bayesian inference using the maximizer of the posterior marginal (MPM) estimate by making use of Monte Carlo simulation for several artificial wave-fronts in remote sensing via the synthetic aperture radar (SAR) interferometry. The simulations find that there is a phase where phase unwrapping is realized under the constraint of the surface-consistency in the hyper-parameter space, and that the upper phase boundary of the phase is steady along the parameter tuning fluctuations around the MAP solution. A



16:45~17:00	WC07-2

Evaluation of Impressions and Movements Related to Negative Emotional Expressions in Dance
Nao Shikanai(Japan Women's University, Japan), Kozaburo Hachimura(Ritsumeikan University, Japan)

This experiment investigated the impressions and movement characteristics that observers use to perceive expressions of negative emotions in dance. Expressions using movements by dancers were recorded by video cameras and a motion capture system. Observers rated words related to impressions and movements. The results of factor analyses indicated that the expressions of negative emotions in dance have particular features.



17:00~17:15	WC07-3

Sparse Representation Approach to Inverse Halftoning by Means of K-SVD Dictionary
Masahiro Hirao, Toshiaki Aida(Okayama University, Japan)

We approach to the problem of inverse halftoning within the frameworks of Bayesian inference and compressed sensing. In this paper, we adopt the K-SVD dictionary for the sparse representation of an original image to be inferred, and develop our previous work with the DCT dictionary restricted to a small number of the slowest basis vectors. Numerical analysis confirms the effectiveness of our approach with the K-SVD dictionary, and makes clear the difference between the characteristics of the K-SVD dictionary and those of the restricted DCT one.



17:15~17:30	WC07-4

Development of Ranging Method for Inter-Vehicle Distance using Visible Light Communication and Image Processing
Nobuo Sasaki, Naoyuki Iijima, Daiki Uchiyama(National Institute of Technology, Gunma College, Japan)

Simple and low-cost method for the measurement of inter-vehicle distance has been developed. Proposed method is based on geometric similarity between an object on the rear surface of a vehicle and its image on image sensor. Size of the object should be well-known or be informed using visible light communication. This time, a license plate was used as the object during the day, and a distance between tail lamps was used during the night. Distance measurements were performed from 2 to 15 m at day and night. Using this method, less than 5% error ratios have been obtained.



17:30~17:45	WC07-5

Wireless communication in mobile robot system
Jang Myung Lee, Minh Nhat Hoang, Dongeon Kim, Seongik Han, Hyunuk Ha(Pusan National University, Korea, Republic of )

This paper proposes an approach to control the communication between 3 mobile robots by using wireless communication.