ITE Technical Report
Online ISSN : 2424-1970
Print ISSN : 1342-6893
ISSN-L : 1342-6893
Search
Advanced search
OR
Browse
Search
22.50
Displaying 1-14 of 14 articles from this issue
Show all abstracts
    • |<
    • <
    • 1
    • >
    • >|
  • [title in Japanese]
    Article type: Cover
    Pages Cover1-
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_Cover1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (15K)
  • [title in Japanese]
    Article type: Index
    Pages Toc1-
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_Toc1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (56K)
  • Condition for Range Measurement from Motion by a Factorization Method with an ITV-based Camera
    Ryuichi Ishino
    Article type: Article
    Pages 1-6
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    The Factorization Method (FM) can reconstruct a 3D shape easily and it requests a lot of views. We investigated conditions, whether or not FM could get 3D information for an object with a monitoring camera like an ITV-based camera. Consequently, if the variance of factors in the matrices is more than I and the rate of summation from the first order singular value to the third one deploy more than 98% for all of them, we can acquire 3 dimensional structure information.
    View full abstract
    Download PDF (778K)
  • A Use of Mean Field Decomposition in MRF-Based Unsupervised Textured Image Segmentation
    Nobuteru TAKAO, Hideki NODA, Eiji KAWAGUCHI
    Article type: Article
    Pages 7-12
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_7
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    This paper proposes a Markov random field(MRF) model-based method for unsupervised segmentation of images consisting of multiple textures. To model a textured image, a hierarchical MRF is used with two layers, the first layer representing an unobservable region image and the second layer representing multiple textures which cover each region. This method uses the Expectation and Maximization(EM) method for model parameter estimation, where in order to overcome the well-noticed computational problem in the expectation step, the Baum function is decomposed using the mean field approximation. Given provisionally estimated parameters at each iteration in the EM method, a provisional segmentation is carried out using the local a posteriori probability of each pixel's region label, which is also derived by mean-field-based decomposition of a posteriori probability of the whole region image.
    View full abstract
    Download PDF (863K)
  • Video Browsing System with Pictorial Indices
    Shin Yamada, Akira Mimori, Akiyoshi Tanaka, Masaki Nakamura
    Article type: Article
    Pages 13-18
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_13
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    We developed an algorithm of automatically detecting scene change points and captioned frames, based on the spatio-temporal distribution of chroma components. This paper reports the trial implementation of this algorithm to a video browing system which reinforces the videotape management such as contents recognition or image searching. This system shows the list of detected indices of a video sequence and offers users the editability of the list. The system has also user friendliness of adapting popular web browser as the image searching screen, and the file structure focused on a re-use of the data. After the trial use with 30 sequences. the effectivity and availability of the system is confirmed. In particular, the indexed still pictures visually urged the comprehensibility of the another video attributes
    View full abstract
    Download PDF (794K)
  • Interactive Presumption of Feature Space Distance and Applying to Image Retrieval
    Aki KOBAYASHI, Toshiyuki YOSHIDA, Yoshinori SAKAI
    Article type: Article
    Pages 19-24
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_19
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    The retrieval demand in the image database has risen because treating the image on the computer generalized. We have proposed the method to retrieve the image through the interaction which are selection and similarity height input by a presentation of the comparison image can easy operation of the user. In this text, we assume the Maharanobis distance to the distance distribution (similarity distribution) in the feature space with the target image, and describe the technique by which these parameters are presumed. The image retrieval system by this method has the characteristic can present the retrieval result by order, and express a correlation between features. First of all, after explaining the feature and the feature space, we describe the outline and the presumption principle in this method. Afterwards, to apply the method to an actual retrieval system, we propose a condition of the comparison image and the presumption procedure of the parameters. Finally, we show the simulation result and confirm the function of the procedure.
    View full abstract
    Download PDF (730K)
  • Edge Direction Features for Image Retrieval
    Yoichi MIZUNO, Aki KOBAYASHI, Toshiyuki YOSHIDA, Yoshinori SAKAI
    Article type: Article
    Pages 25-30
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_25
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    This paper proposes a shape feature for image retrieval from a large-scaled image database. The proposed feature is based on edge direction and its continuty. By using this feature. we can extract shape features from an image without special models or knowledge. Another advantage is that the feature value can be calculated automatically without requiring an extraction of target objects, which enable us to apply the proposed feature to large-scaled databases. First, we explain the edge direction feature. Then, we show the effectiveness of the proposed feature through an image retrieval experiment.
    View full abstract
    Download PDF (880K)
  • Moving detection suitable for image retreaving
    Akira KAMEGAYA, Hirotsugu KINOSHITA
    Article type: Article
    Pages 31-36
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_31
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    In this paper, we study a method of automatic recognition of the motion in moving images for the purpose of the image retreaving. The motion of camera and many objects are mixed in the moving image, and complicated motions are presented. We present a method for the recognition of the motion of camera and objects. In this method, a segmentation method using the motion vector is applied to recognize the motion of each objects. Furthermore, we study characteristics of the motion of camera such as zooming.
    View full abstract
    Download PDF (795K)
  • Image Flow Analysis Using a Speed-Tuning Wavelet Transformation
    Biswas Satyen, Yoshio Ebina
    Article type: Article
    Pages 37-42
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_37
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    Image flow pattern is the display of the local image motion or velocity field distribution. A Wavelet transformation (WT) method has been searched out and the present paper proposes the method to make velocity field or optical flow pattern. The validity of the method is verified to be correct in several mathematical models and microscope images taken in our laboratory, and produces more fined velocity vector pictures than those by other groups, for example, motionless corner blocks have no vector distribution area in a rotating Rubik cube picture. The boundary of a pulsing area is clearly observed as discontinuous optical flow in our experimental microscope Nematode images
    View full abstract
    Download PDF (524K)
  • A New Clustering Method for Nonlinear Fractal Image Coding
    Masataka TAKADA, Mitsuhiko FUJIO
    Article type: Article
    Pages 43-46
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_43
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    Spread of computers and home-page contains images have increased recently. Therefore, it has need of the image coding has high compression rate. In this paper, we study nonlinear fractal image conding proposed by Popescu et al. We explain and implement that algorithm. We propose the new clustering algorithm for be fast and implement that. And we get the data by comparing that with the nonlinear fractal image coding. The data shows our algorithm is fast.
    View full abstract
    Download PDF (452K)
  • A Consideration on Video Information Architectures for Multimedia Scalability Packages Communication Services
    Mei KODAMA, Hiroyuki KASAI, Hideyoshi TOMINAGA
    Article type: Article
    Pages 47-52
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_47
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    We had indicated communication services using multimedia scalability pachage : MSP, which can provide multimedia information accoding to functionality and availability for users. This paper defines data structures of MSP using scalability architectures. Especially, when we made the hierarchical architecurue, we considered temporal feature information. And then, as coding schemes, we attention to video hybrid scalability using spatical scalability and temporal one and scalability profiles are shown.
    View full abstract
    Download PDF (627K)
  • A Study on Playing and Editting Processing of Video Infomation base on the Multimedia Scalability Package Structure
    Hiroyuki KASAI, Mei KODAMA, Hideyoshi TOMINAGA
    Article type: Article
    Pages 53-58
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_53
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    We propose the Multimedia Scalability Packages(MSP) as the packages which are structured by contents with multimedia scalability structure. In this paper, we study the playing processing method and the cditting processing method based on the scalable structure which contents with scalability and the relations among contents with scalability. In particular, we explain the data structure for scalable playing and editting. At last, we state the application to the interacitve video communication service.
    View full abstract
    Download PDF (744K)
  • A Study on the Process of Playing/Editing and Processing Time for Electric Thesis with Hierarchical Structure based on Contents
    Kazuyuki TAKAYA, Hiroyuki KASAI, Mei KODAMA, Hideyoshi TOMINAGA
    Article type: Article
    Pages 59-64
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_59
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    In this paper, we explain the electric thesis which has hierarchical data structure based on contents. At first, we state requests for existing electric thesis, and the functions to play and edit for the content based hierarchical electric thesis. In addition, we explain the processing algorithm to play the electric thesis simply, and to edit simple electric thesis. We made the hierarchical electric thesis with making algorithm we stated. At last, we compare the processing time for decoding multimedia source data from hierarchical electric thesis data and the processing time for generating multimedia source data from non-hierarchical data as input, and we state the validity of hierarchical electric thesis data based on contents.
    View full abstract
    Download PDF (806K)
  • [title in Japanese]
    Article type: Appendix
    Pages App1-
    Published: September 25, 1998
    Released on J-STAGE: June 23, 2017
    DOIhttps://doi.org/10.11485/itetr.22.50.0_App1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (75K)
    • |<
    • <
    • 1
    • >
    • >|
feedback
 Top