-
Article type: Cover
Pages
Cover1-
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Index
Pages
Toc1-
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Shin YAMAMOTO, Yoshimitsu AOKI
Article type: Article
Session ID: ME2009-52
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
These days in an aged society the problem in medical practice is the dysphagia. It causes choke, pneumonia and apoplexy. Therefore the evaluation of swallowing function is important for the prevention and the rehabilitation. Now in clinical practice they use manipulation test and Video Fluorography test. But one is invasive, the other is not safe and easy. For the issue, we present non-invasive, safe and easy evaluation system of swallowing function by the Image analysis on the neck using dual windows. This system calculates the larynx elevation time by analyzing the shape variation on neck taking by FG vision sensor. In an experiment, we verify increase of the time caused by aging. It is similar to an analysis by Video Fluorography.
View full abstract
-
Akiko OSAKA, Yew KURABAYASHI, Toshihiro FURUKAWA, Tae IWASAWA, Tomio I ...
Article type: Article
Session ID: ME2009-53
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In order to estimate performance of respiratory motion, we analyzed diaphragm motion by 3rd order spectrum. We evaluated bispectrum for a point on each diaphragm, while obtained cross bispectrum for three different points on it, then compared them between normal subjects and patients. As a result, there existed some flat areas in the phase spectra of normal subjects, but these phases tended to fluctuate randomly with patiants. In addition, in the case that we couldn't clarify differences between normal subjects and patients by their bispectra, cross bispectra denoted certain difference. Consequently, 3rd order spectrum suggests its possibility to extract a feature of respiratory motion and estimate its performance.
View full abstract
-
Naoto UMEMORI, Toshiyuki GOTOH, Seiichiro KAGEI, Tae IWASAWA
Article type: Article
Session ID: ME2009-54
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Geometric parameters and threshold values are useful for 3D region segmentation in images. This paper proposes a segmentation method based on Distance-Threshold Maps (DTM). In this method, distance and threshold maps are generated using 3D Voronoi diagram and pseudo Euclidean distance transformation from sparse points with local information on geometric parameters and threshold values, and 3D segmentation is performed referring the maps. Experimental results for evaluating the method are shown using 3D chest CT images.
View full abstract
-
Yosuke MUKAI, Saori NOZAWA, Mie SATO, Masao KASUGA
Article type: Article
Session ID: ME2009-55
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In recent years, by the situation that computer networks reached the general popularization, the advance information and communication technology society has been reaching maturity and a sensuous effect of the sensitivity to the addressee by the communication is in the spotlight. Therefore the purpose of this study is comparing the electronic media with the paper media, and examining the difference between paper media and electronic media for sensitivity, by analysing of the evaluation of the sensitivity. The evaluation index of the sensitivity of both media is suggested and the characteristic extraction of both media by the evaluation experiment has been carried out.
View full abstract
-
Tsuyoshi SATO
Article type: Article
Session ID: ME2009-56
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
There are many signboards to provide visual information in urban space, which provide various information to us, such as the store name and direction. However, when the sign is inappropriate, there would be a possibility that we couldn't attract enough customers. In this study, the relationship between signboards and the characteristic of eye fixation of driver's is revealed, and location to establish signboards and appropriate format of signboards is also considered.
View full abstract
-
Akihisa ITO, Jun-ichi IMAI, Masahide KANEKO
Article type: Article
Session ID: ME2009-57
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Mobile robots are required to recognize dynamically changing environments to avoid collisions with walkers. Distance information obtained by stereo vision is useful for recognition of surrounding environments. However, there are some problems such that a stereo camera cannot obtain distance information for textureless regions and that objects and their surrounding cannot be distinguished by distance information if they are very close to each other. This paper proposes a method for generating environment models more robustly by integrating distance and color information, and for finding robot's path autonomously using the environment models. The effectiveness of the proposed method is shown through experiments with a real robot.
View full abstract
-
Yusuke In, Kento Hirano, Kenta Asakawa, Tetsuya Tsukada, Hitomi Muraka ...
Article type: Article
Session ID: ME2009-58
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper prompted us to conduct a trial to confirm the margins of error in the current GPS services available on mobile phones. It was carried out in several sites in Japan in order to ascertain the nature of the errors. To identity more specifically the by GPS function, image processing technique, using autocorrelation method between mobile phone still image and accumulated motion images was examined.
View full abstract
-
Tetsuya Tsukada, Kento Hirano, Kenta Asakawa, Yusuke In, Hitomi Muraka ...
Article type: Article
Session ID: ME2009-59
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper prompted us to conduct a trial to confirm the margins of error in the current GPS services available on mobile phones. It was carried out in several sites in Japan in order to ascertain the nature of the errors. These results suggested that the GPS errors were estimated to be within 50m〜100m for the most of sites measured.
View full abstract
-
Michio ISHII, Jun-ichi IMAI, Masahide KANEKO
Article type: Article
Session ID: ME2009-60
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
To make robots work in daily living spaces for humans, it is important that a robot can estimate an arrival direction of an interrupting sound and recognize its class. Conventional robots can only recognize a single sound source and simultaneous utterance of multiple speakers supposing the number of speakers in advance. However, they cannot deal with the interrupting sounds and recognize their classes without any prior knowledge. This paper proposes a method for estimating arrival directions of interrupting sounds without any supposition on the number of sound sources, and for recognizing their classes. The effectiveness of the proposed method is shown through experiments with a real robot.
View full abstract
-
Kazuhiro MAEDA, Yasuyuki SAITO
Article type: Article
Session ID: ME2009-61
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper presents a generation method of the music piece by using the characteristics of the image. In the picture exhibition, music is played around a displayed picture. The music emphasizes the impression of the displayed picture if these impressions are same. Using K-mean method, extract domain information from an image and generate the melody line on the domain information. A musical piece is generated by converting the average color of the domain on the melody line into notes. As a result, the musical piece which had the characteristics of the image is generated.
View full abstract
-
Keisuke SOEDA, Yasuyuki SAITO
Article type: Article
Session ID: ME2009-62
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper describes a control of electronic musical instruments by gesture. Recently, electronic musical instruments have got cheap and useful, the quality of sound has improved, and therefore everybody has had a nice musical atmosphere. However, it is difficult to handle electronic musical instruments which have a large number of functions. So, I developed a highly novel MIDI interface which uses gesture to control electronic musical instruments easily.
View full abstract
-
Daisuke TSUGANE, Toshiyuki GOTOH, Naoyoshi TAMURA, Hiroshi SHIMADA
Article type: Article
Session ID: ME2009-63
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper describes an software environment for constructing multi-agent models of Orchestral Musicians which play instruments under the interaction of musical information between a conductor and musicians. The experimental results using prototype software are also shown.
View full abstract
-
Genki UMEIZUMI, Yoshiki MIZUSHIMA, Syuhei NOMURA, Noriko NAGATA
Article type: Article
Session ID: ME2009-64
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
The need for rendering woven cloth arises frequently in computer graphics. Cloth has a specific appearance, sheen, and transparency. A BRDF (bidirectional reflectance distribution function) model is well known as the basic technology for expressing the appearance of woven cloth; this model has been the subject of several researches till date. In this paper, we propose technique to estimate woven structure and yarn structure of cloth by image processing of the images gained by a BRDF equipment. It becomes possible to express the textures involving woven structure and yarn structure of clothes, without carrying out creation of a complicated shading model and texture mapping.
View full abstract
-
Hitoshi HAYAKAZE, Masashi OKUDAIRA
Article type: Article
Session ID: ME2009-65
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper proposes a visualization method of web sites relations with a primary web site so as for users to easily access interested web sites based on their relevance ratio. The ratio is calculated based on weighted digraph of links. The related Web sites are automatically searched and registered. According to the relevance ratio, the web sites are laid out on a browser so that those Web sites are intuitively identified. Experimental results show that related Web sites with a specified site are easily accessed.
View full abstract
-
Masakazu OGI, Makiko SEMBON, Noriyuki HORIE, Masashi NAKATSU, Mie SATO ...
Article type: Article
Session ID: ME2009-66
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In recent years, although it is easier to establish new company because of legal changes, but the number of companies going bankrupt is increased continuously for 3 years. At this point, as the general evaluation approach of business management, there is quantitative evaluation using financial statement, and is evaluated with financial indicator. In this study, the purpose is making the quantitative evaluation simpler by finding financial indicator which captures total image of company in more easier way. In addition, business management evaluation and picking out financial indicator is performed by obtained data using 27 items of index used by Self Management Diagnostic System of the Small and Medium Enterprise Agency.
View full abstract
-
Yoshihiko SAKURABA, Kota AOKI, Hiroshi NAGAHASHI
Article type: Article
Session ID: ME2009-67
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper proposes a method for simplification of 3D shapes based on spatial approximation using sets of spheres whose radii are discretely different. The sets of spheres with different discrete radii enable us to generate a hierarchical expression easily and to reduce computational costs. Experimental results show that our representation of range data is stable for various 3D models. The proposed model representation is applied to the hierarchical registration by ICP algorithm. The results show that the hierarchical representation is very effective for model registration and matching and its computational cost much lower than that of using the original 3D model.
View full abstract
-
Ryosuke KAWANISHI, Atsushi YAMASHITA, Toru KANEKO
Article type: Article
Session ID: ME2009-68
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Map information is important for path planning and self-localization when mobile robots accomplish autonomous tasks. Then, in unknown environments, they should generate an environment map by themselves. An omni-directional camera is effective for environmental measurement, because it has a wide field of view. We propose an estimation method of omni-directional camera motions with feature flow model. Our method fits feature point movements to feature flow model for estimation of camera motions. Experimental results showed the effectiveness of the proposed method.
View full abstract
-
Daigo TAKEUCHI, Yu ITO, Atsushi YAMASHITA, Toru KANEKO
Article type: Article
Session ID: ME2009-69
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Person tracking in images is an important function to an automatic surveillance system using a vision sensor. However, it is difficult to identify a person exactly in an image due to the variety of poses. We previously proposed a person tracking method by Haar-like features and mean-shift algorithm. However, there is a problem that a person is not tracked until his/her frontal face is detected. This problem is solved in this study by face detection of arbitrary poses using trained cascades.
View full abstract
-
Hirokatsu KATAOKA, Kazuo OHZEKI, Yoshimitsu AOKI, Yoshihiro YAMADA, At ...
Article type: Article
Session ID: ME2009-70
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
The attempt to understand players' movement from sports scene is seen a lot. To implement it, obtaining the players' positional information is essential: however, several cameras are needed. Here, we propose a method to track football players using a single camera set up at the stand of stadiums. The camera pans and brings the players and the ball into the view, enabling to record the whole pitch. The players' trajectories in frames and camera movement are obtained to estimate the trajectories on the pitch. Moreover, the effectiveness of the proposed football players tracking method is evaluated.
View full abstract
-
Masakazu TSUCHIHASHI, Jun-ichi IMAI, Masahide KANEKO
Article type: Article
Session ID: ME2009-71
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper proposes the template-based method to generate expressive caricatures automatically. This method consists of two steps, that is, the first step to extract and analyze facial parts in an input facial image, and the second step to select an appropriate template for each facial part. Facial parts such as eyes, eyebrows, nose and mouth are extracted from an input face image by using a shape-based pattern matching method. Feature parameters are calculated using principal component analysis on luminance distribution of facial parts extracted from the facial image database of 300 people. Next a hierarchical clustering method is applied to classify facial parts into multiple classes with similar shapes. Multi-class classifier is designed by SVM based on feature parameters and classified results. A class of each facial part in an input face is determined by the multi-class classifier and a caricature can be generated by assigning an appropriate template to each of facial parts. Each template represents the typical or exaggerated shape of selected class. The proposed method reduces the operator's load to interactively select and operate templates. Subjective evaluation of automatically generated caricatures shows the usefulness of the proposed method.
View full abstract
-
Jie LI, Jun-ichi IMAI, Masahide KANEKO
Article type: Article
Session ID: ME2009-72
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Caricatures sometimes impress people more than original photographs. This paper aims to improve the automatic caricature system in [1] by generating more expressive caricatures based on analyzing the features of face images with impressions of ethnicity, gender, and expressions. An eigen space model is set up by the principal component analysis of 2D edge information of faces. Qualitative and quantitative description of differences and importance of principal components of each impression are given by the linear discriminant analysis method. By using Support Vector Machine method, the mixed eigen space is divided into four, that is, Japanese male/female, Caucasian male/female. Towards Japanese male input images, expressions transformations by learning transformation rules between six basic expressions and exaggerated caricatures by the caricaturist are carried out. More expressive caricatures can be drawn and animated by the proposed method.
View full abstract
-
Yuki OSATO, Toshiyuki GOTOH, Takeshi TAKEGAMI
Article type: Article
Session ID: ME2009-73
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper describes the letter input interface for the severely disabled using one's blinks. The method to discriminate a voluntary from an involuntary blinks by image processing and to adaptively adjust the letter presentation speed has been discussed. Experimental results using developed prototype system have also been shown.
View full abstract
-
T. TAKIMOTO, M. OHARA, J. YAMAGUCHI, H. ISONO, A. KUROSAWA
Article type: Article
Session ID: ME2009-74
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Hearing ability decreases with age. When elderly people watch TV, the volume must be raised higher than that for young people, often making it difficult for elderly people to watch TV together with young people. In this research, we studied the use of bone-conducting headphones allowing elderly people to view TV comfortably without straining their ears
View full abstract
-
T. KOISO, T. SENSUI, T. KANBE, J. YAMAGUCHI, H. ISONO
Article type: Article
Session ID: ME2009-75
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
LED traffic lights are replacing conventional electric-bulbs; however, LED traffic lights are often difficult for people with color blindness. In this research, we arranged three white LED lamps directly below three blue, yellow, and red LED lamps. Individuals' methods of recognizing the light status from the position of the white LED lamps were then observed by interlinking the white LED lamps with the colored lamps.
View full abstract
-
Daiki SAKASHITA, Mie SATO, Masao KASUGA, Syuji KUBOTA, Seigou YASUDA, ...
Article type: Article
Session ID: ME2009-76
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Recently, the burden of assistance is increased due to the rapidly falling birhrate and the aging population, as the number of elders rose while those who help them are insufficient. In this research, the aim is to build a care system in order to alert the assistance to the abnormal situation, through the forecast of the potential threat when the condition of the one who needs help is understood. First, the condition of the person who needs help is figure out using a still picture, and the information is accumulated in the data base, and the based on that, the action forecast by the system is constructed.
View full abstract
-
Takuya HIRAO, Yutaka NAKANO, Toshiyuki YOSHIDA
Article type: Article
Session ID: ME2009-77
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper proposes a simple tablet digitizer using a cellular phone camera, which outputs the coordinate of a point on a paper indicated, for example, by a finger. First, a perspective image including the whole paper is obtained by the camera, and then the paper area as well as the indicated point are extracted from the image. The coordinate of the point is calculated by applying a bisection algorithm on the paper area. Experimental results show that the proposed technique is promising in an actual environment.
View full abstract
-
Ryu TAKAYAMA, Masashi OKUDAIRA
Article type: Article
Session ID: ME2009-78
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper proposes a method of generating fireworks CG interactively by shaking an acceleration sensor with OpenGL. Generation modes consist of a burst mode and a launch mode. The modes as well as positions and sizes of fireworks are determined with the values received from the acceleration sensor. Unreal fireworks are also realized, for example, a firework that is visible in daytime or that has special "star" shapes. These fireworks demonstrate their attractiveness. Through examinations, it is found out that shaking strength depends on each person feeling about the normal size of fireworks to be generated. Thus, the threshold is set adjustable in two levels for intuitively acceptable operation.
View full abstract
-
Mitsutoshi ITO, Kenji HONDA, Mie SATO, Naoki HASHIMOTO, Masao KASUGA
Article type: Article
Session ID: ME2009-79
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Immersive display system is an image projection device that provides wide-angle images by a screen surrounding a user. Since immersive display is an expensive and huge device, moving and using this system in a standard room is difficult. In this study, we provide a wide-angle image projection device using flat mirrors that can be used in a standard room. Then, we compare this device with normal image projection and consider utility of a provided device.
View full abstract
-
Satoru SUGAWARA, Kenji HONDA, Mie SATO, Naoki HASHIMOTO, Masao KASUGA
Article type: Article
Session ID: ME2009-80
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In Recent years, study on producing wide-angle images are popular. In these studies, the way that reconstructs wide view angle images by extracting peripheral images from time-series image frames in real-time was proposed, that is called "Pseudo Wide-anglize". So, purpose of my study is researching effects of pseudo wide-angle images on immersive feeling by evaluation experiment using various image contents including taken on spot, CG and game playing.
View full abstract
-
Rie MIYAKI, Atsushi YAMASHITA, Toru KANEKO
Article type: Article
Session ID: ME2009-81
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
There are individual differences in color vision. It is difficult for a person with defective cones in the retina to recognize the difference of specific colors. We propose a presentation method of color information by using a projector-camera system. The system projects border lines or color names on real objects when they have specific color combinations. Registration of projected images and real objects is made by using projected markers.
View full abstract
-
Shuhei ITABASHI, Mie SATO, Naoki HASHIMOTO, Masao KASUGA
Article type: Article
Session ID: ME2009-82
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Recently, the image projection on any optional place, not only on a white screen, without any incompatibility, is necessary. Therefore, in this research, reducing the pattern effect by luminance correction is studied. In this research, based on human perception of brightness of moving subject, we tested a correction process for improving the darkness point of image when luminance correction is being carried out. In this paper, we report the experiment result about human perception of brightness of moving subject in projected image in order to obtain basic data of that correction process.
View full abstract
-
Kenta Asakawa, Kento Hirano, Tetsuya Tsukada, Yasuhiko Hayashi, Yusuke ...
Article type: Article
Session ID: ME2009-83
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In the era of ICT, mobile phone will have position of own daily life. Especially GPS function of mobile phone will become essential tool. This paper introduces collection and presentation system applying such a GPS function against disaster.
View full abstract
-
Takeshi KOSUGI, Minoru MITSUI, Tomoharu ISHIKAWA, Jianwu DANG, Makoto ...
Article type: Article
Session ID: ME2009-84
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
A high fidelity sound reproduction is important for evoking human's deep KANSEI. To reproduce sound with high fidelity, in this study, we proposed a new type amplifier to improve efficiency of a loudspeaker and dynamic property of the speakers. A circuit was designed to increase the BL product of speaker equivalently. The characteristics of the proposed amplifier was evaluated objectively and subjectively. As a result, proposed amplifier preformed better than conventional one.
View full abstract
-
Masahide KANEKO
Article type: Article
Session ID: ME2009-85
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
A "face" is very familiar with us and that is the object which we see most frequently in our daily lives. Everyone has some interests in faces. In the academic fields, the face is treated in various fields such as anthropology, psychology, engineering, and so on. This paper first describes various information which a face conveys and summarizes various research fields relating to a face. Then the progress of studies on processing of facial image information is overviewed. The typical research results on computerized facial caricatures are also introduced.
View full abstract
-
Article type: Appendix
Pages
App1-
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Appendix
Pages
App2-
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Appendix
Pages
App3-
Published: February 28, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
