All issues
Predecessor
36.54
Displaying 1-24 of 24 articles from this issue
- |<
- <
- 1
- >
- >|
-
Article type: Cover
Pages Cover1-
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (12K) -
Article type: Index
Pages Toc1-
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (67K) -
Yuki IWAKAMI, Daishi WATABE, Hideyasu SAI, Takanari MINAMIDANI, Katsuh ...Article type: Article
Session ID: ME2012-128
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSWe identify the ear shape that is easy to individualize from the viewpoint of statistics of the morphological classification, which is not easy to identify by our previous results.View full abstractDownload PDF (555K) -
Yan WANG, Dashi WATABE, Wenbo ZHAO, Takanari MINAMIDANI, Hideyasu SAI, ...Article type: Article
Session ID: ME2012-129
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSRobustness improvement was discussed through increasing the number of normal vector modelsView full abstractDownload PDF (485K) -
Takanari MINAMIDANI, Daishi WATABE, Hideyasu SAI, Katsuhiro SAKAIArticle type: Article
Session ID: ME2012-130
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSAn ear feature point detection algorithm is proposed based on the jet space similarity and eigen ear graphView full abstractDownload PDF (567K) -
Daisuke UKAI, Kazuhito MURAKAMIArticle type: Article
Session ID: ME2012-131
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSPersonal authentication with facial thermo-image has a problem that it is weak to the change of the environment temperature. This paper proposes a method for facial authentication using stable parts extracted by a thermo-vision camera. In order to realize a personal authentication system, first, this paper presents a method to extract some characteristic facial parts which are stable against the change of the environment temperature. Then, proposes a new facial authentication method using these characteristic parts.View full abstractDownload PDF (759K) -
Genki MOMOSE, Hiroki TAKAHASHIArticle type: Article
Session ID: ME2012-132
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThis paper aims for analyzing 3D(Dimensional) gait feature for person authentication. Gait is useful even if it is taken from faraway place. A Kinect^<TM> is used to obtain gait data. Subjects walked at 3m apart from the camera and the camera is pulled by hand cart. 3D coordinates of both elbows, both knees and both wrists are measured. Furthermore, errors by moving of the camera is measured as preliminary experiment. Then, the errors of gait data are less than 1.0mm. Obtained gait data are analyzed by principal component analysis. As a result, both knees heights are the significant factors on a human gait feature.View full abstractDownload PDF (786K) -
Norio TAGAWA, Shoei KOIZUMI, Kan OKUBOArticle type: Article
Session ID: ME2012-133
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSWe already proposed a measurement method of a distance between a camera and a target object using image motion blur caused by small camera rotations imitating fixational eye movements. However, in this method, firstly the image motion blur is detected, and then it is analyzed to compute the distance. Therefore, this method does not have a complete optimality for computation. In this study, we propose two new algorithms for measuring directly the distance from image motion blur, and their effectiveness is confirmed by numerical evaluations.View full abstractDownload PDF (572K) -
Kosuke WAKABAYASHI, Norio TAGAWA, Kan OKUBOArticle type: Article
Session ID: ME2012-134
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThere are various and a lot of depth recovery methods have been studied, but a discussion about an unification of individual methods is expected not to be enough yet. In this study, we assert that the importance and the necessity of an image generation consistency. Various clues including stereo, motion, texture and so on can be considered as a good tool for obtaining initial values for unification algorithm based on an image generation consistency. Based on the above strategy, as a first step of our research, through a simple problem with two-views, the unification of stereo and shading without explicit disparity detection is examined based on an image generation consistency, and simple evaluation results are shown by simulations using artificial images.View full abstractDownload PDF (604K) -
Jianting CAOArticle type: Article
Session ID: ME2012-135
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThe brain death is defined as the cessation and irreversibility of all brain and brain stem functions. Based on this definition, the basic clinical criterion has been established in most countries. However, in the standard process of brain death diagnosis, it often involves certain risks and takes a long time. Therefore, in order to reduce the risk and to save the precious time for the medical care in clinical practice, an EEG-based preliminary examination system was proposed in the standard clinical procedure. The purpose of the preliminary EEG examination is to explore advanced signal processing tools to discover whether any brain wave activity occurs in the patient's brain. The EEG preliminary examination system includes a portable EEG data acquisition tool and the EEG-oriented signal processing tools. The signal processing tools include several routines of noise reduction, source separation, feature extraction, as well as statistical tests. This study will demonstrate the performance using the approximate entropy and EEG energy criteria to evaluate the differences between deep comatose patients and brain death.View full abstractDownload PDF (649K) -
Yuri HIRAIWA, Satoru INOUEArticle type: Article
Session ID: ME2012-136
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn this study, we propose the method of splitting the human body into several parts from the captured images into computer via Web camera. This method is intended to be used in virtual fashion coordination, To realize the purpose we mentioned above, we compare the following two methods. One method is to split the body into parts based on the changes of clothes sewing pattern extracted by discrete cosine transform. The other is to specify the body division point based on the changes in the angle of several straight lines seen in the human body extracted by probabilistics Hough transform.View full abstractDownload PDF (665K) -
Msahiko HATAKI, Dong-ying JUArticle type: Article
Session ID: ME2012-137
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSNow, a look move measuring instrument has many expensive things. In this research, in order that future many may obtain as a result of measurement, a short form look move counting system is developed. A USB camera (PlayStationEye) is used, image processing is performed to the acquired image, and look movement is measured. From this measurement result, usefulness of a short form look move counting system is clarified.View full abstractDownload PDF (319K) -
Koyuki HAYAKAWA, Hiroki TAKAHASHIArticle type: Article
Session ID: ME2012-138
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThis paper aims for swapping a face to other ones retaining facial expressions to protect portrait and privacy rights. ASM(Active Shape Models) is employed to detect facial feature points. Face parts, which are a face base, eyes and a mouth, are extracted. These parts are distinguished by AU(Action Units) corresponding to "Happiness" and "Sadness". Facial expression similarity is defined by a slant or a distance between facial feature points. These parts positions are corrected by total error of all feature points belonging to their parts to minimized the error. As a result, video face swapping with retaining facial expression is generated.View full abstractDownload PDF (768K) -
Masakatsu NISHIGAKIArticle type: Article
Session ID: ME2012-139
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (113K) -
Hiroyuki NAKAJIMA, Pengfei LIU, Taiyo MAEDAArticle type: Article
Session ID: ME2012-140
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn this paper, we propose system that provides similar behavior from the web access log when students learn using the interne. In order to analyze these efficiently, we construct a distributed environment using Apache Hadoop, and discuss the evaluation in several cases. As the result, speed up ratio using Pentium4 with 3 nodes to analyze the data of 1GB is 2.65 times. However, in the case of data is less than 1GB, the distributed environment is not efficient. As for behavior patterns, grouping of students with similar behavior are achieved by using collaborative filtering using Euclidean distance.View full abstractDownload PDF (473K) -
Machiko Kobori, Yuji TokiwaArticle type: Article
Session ID: ME2012-141
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIssues in improving teaching skills in language education are discussed on the basis of four-year empirical research study conducted in English teacher training courses in Japan: it examines the effect of the lesson observation by using digital video streaming in teacher development. It also introduces how student teachers went on a process of conducting their presentation and micro-teaching and gives an insight into their challenge in making progress with the related techniques in teaching by experiencing the lesson analysis and self-assessment.View full abstractDownload PDF (693K) -
Daisuke KANEKO, Minoru MITSUIArticle type: Article
Session ID: ME2012-142
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSRecently, PC audio which uses PC as reproduction equipment is popular. Since an analog audio signal would be affected, it would lead to sound quality deterioration if D/A conversion are performed inside PC. Therefore the exterior DAC appeared. In PC which only a digital signal process, signal does not change theoretically. However it is said that the sound quality are changed with the components of PC. Its cause is considered to be a digital signal jitter when performing D/A conversion. Jitter is fluctuation of the time interval of upon digital pulse wave from a signal waveform. If jitter is contained, the analog output signal which should not change theoretically will change. In this study, Relationship between the hardware requirement of PC, the amount of jitter, and also sound quality was clarified.View full abstractDownload PDF (584K) -
Yoshinori KUBOTA, Satoru INOUEArticle type: Article
Session ID: ME2012-143
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn this research, we propose the furniture layout system reflects on the user's preference. Our proposed system determines where the furniture is arranged based on the physical potential surface. This surface is constructed based on the history that furniture has been placed by user previously. The more the user prefers the placement at certain position as a area where the furniture is arranged, the depth of potential become deeper. The position which has the deeper potential depth attracts the object strongly the as a physical dynamics. Adopting the algorithm mentioned above, our system is able to proposed the arrangement of furniture relatively preferred.View full abstractDownload PDF (459K) -
Syoji KOBASHI, Yuri T. KITAMURA, Kei KURAMOTO, Kuriko SHIMONI, Masako ...Article type: Article
Session ID: ME2012-144
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThere are no methods for noninvasively detecting epleptogenic focus, which is the part of the brain from which epileptic discharges originate. This study investigates a method using magnetic resonance (MR) images. The method is composed of (1) granulation, (2) feature extraction, and (3) pattern classification. Granulation is to construct a set of voxels, and feature extraction and classification are done for each granule. Feature extraction is to calculate a set of numbers that express features of granule, and pattern classification is to estimate the degree of epleptogenic focus by evaluating the features.View full abstractDownload PDF (669K) -
Aya HASHIOKA, Syoji KOBASHI, Kei KURAMOTO, Yuki WAKATA, Kumiko ANDO, R ...Article type: Article
Session ID: ME2012-145
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSNeonatal cerebral disorders such as hypoxic-ischemic encephalopathy might deform the brain shape, and reduce patient's cerebral functions. Early detecting and rapid cure of the cerebral disorders prevents the cerebral disorders developing worse. For quantitative diagnosis of the cerebral disorders, measurement of cerebral volume and surface area is effective. In this research, we propose an automated brain segmentation method for newborn brain. The proposed method produces fuzzy object models from learning dataset. Fuzzy object models express brain features by fuzzy membership functions. Using the FOMs, deformable surface model estimates subject's brain region. We applied the proposed method to 12 newborn subjects. From experimental results, the proposed method can segment the brain segmentation with high segmentation accuracy.View full abstractDownload PDF (608K) -
Shohei Suwa, Jiangting CAOArticle type: Article
Session ID: ME2012-146
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSBrain computer interface (BCI) is a system that uses Electroencephalogram (EEG) signal to control various external devices. BCI system using an auditory P300 response is the most popular for gives much freedom to the BCI user. This paper presents an novel BCI system using P300 with P100 responses. The main advantage of our BCI system is that we can obtain two different kinds of responses in a single trial EEG task. And we have EEG experimented for induce auditory P300 and P100. In the EEG data analysis, we first apply the averaging approach to validate our experiments. Next, we employ the Multivariate empirical mode decomposition (MEMD) algorithm to extract P300 and P100 components.View full abstractDownload PDF (8854K) -
Article type: Appendix
Pages App1-
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (76K) -
Article type: Appendix
Pages App2-
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (76K) -
Article type: Appendix
Pages App3-
Published: December 04, 2012
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (76K)
- |<
- <
- 1
- >
- >|