ITE Technical Report
Online ISSN : 2424-1970
Print ISSN : 1342-6893
ISSN-L : 1342-6893
Search
Advanced search
OR
Browse
Search
34.22
Displaying 1-12 of 12 articles from this issue
Show all abstracts
    • |<
    • <
    • 1
    • >
    • >|
  • [title in Japanese]
    Article type: Cover
    Pages Cover1-
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_Cover1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (15K)
  • [title in Japanese]
    Article type: Index
    Pages Toc1-
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_Toc1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (43K)
  • Adaptive-Bandwidth Shape From Focus/Defocus Technique
    Tatsuhiro Yamaguchi, Yutaka Nakano, Toshiyuki Yoshida
    Article type: Article
    Session ID: IST2010-29/ME2010-10
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    "Shape from focus or defocus" has been known as one of the 3-D shape reconstruction techniques, which estimates depth information from a set of multi-focus images of a target object. However, the technique suffers from the disadvantage that the luminance variation between adjacent blocks severely affects a focusing measure used for the depth estimation, causing accuracy degradation in the target model. To alleviate the problem, this paper proposes a new technique which adaptively selects three kinds of high-pass filters for the calculation of the focusing measure based on the evaluation of the single/multi-peak behavior of the focusing measure. The proposed technique has been applied to artificial models and actual images to illustrate the usefulness of the proposed technique.
    View full abstract
    Download PDF (833K)
  • Depth estimation of occluded objects for compact compound-eye camera TOMBO
    Keiichiro KAGAWA, Jun TANIDA
    Article type: Article
    Session ID: IST2010-30/ME2010-10
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_5
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    TOMBO (Thin Observation Module by Bound Optics) is a compact three-dimensional camera, which is composed of a single image sensor and a micro-lens array. Compound-eye images include depth information of the subjects as disparities. In this paper, an occlusion-tolerant depth estimation method based on an adaptive support-weight approach and camera selection is presented.
    View full abstract
    Download PDF (776K)
  • Interpolation method based on structural characteristic for coarse-sampled Ray-Space
    Risa Numada, Kousuke Imamura, Hideo Hashimoto
    Article type: Article
    Session ID: IST2010-31/ME2010-10
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_9
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    Image-based rendering is a technique for free-viewpoint image generation from multi-view images. Ray-Space method was proposed as a typical method of image-based rendering, and free-viewpoint image generation generally requires ray interpolation. The quality of free-viewpoint image depends on the accuracy of a ray interpolation in Ray-Space. In this paper, we propose a robust ray interpolation method for coarse-sampled Ray-Space. The proposed method interpolates ray by hough transformation using the straight line structure of Epipolar-Plane Image (EPI).
    View full abstract
    Download PDF (997K)
  • Preliminary Discussion of Image System Composed of Independent Timeline in Pixels
    Junichi Akita
    Article type: Article
    Session ID: IST2010-32/ME2010-10
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_13
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    Movie, or motion picture is conventionally composed of the sequence of still images. However, the photo receptor cells on the retina in our eye sight system asynchronously emit the signal, followed by the image recognition in visual cortex based on asynchronous photo signal. In this paper, the author describes the idea to deal independent, asynchronous time axis for each pixel, as well as the preliminary simulation results for generating binary movie. The author also discuss the application and implementation of the proposed idea.
    View full abstract
    Download PDF (771K)
  • Implementation and Evaluation of Real-time Object Detection System by using Modulated Light
    Nobumasa GINGAWA, Norihisa KIYOFUJI, Takayuki HAMAMOTO, Tomoshi SASAKI ...
    Article type: Article
    Session ID: IST2010-33/ME2010-10
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_17
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    We have been investigating active sensing using modulated light which removes influence of environmental-light change and detects objects high-precisely. Smart image sensor which we already designed and implemented has two capacitors in a pixel circuit. It apportions and accumulates electric charges for each capacitor alternately. The influence of environmental light change can be removed by synchronizing timing between integration of electric charges and switching of modulated light followed by subtraction of two values outside the image sensor. In this paper, we explain the prototype of real-time object detection system and its evaluation. As a result, we confirm the effectiveness of this system. We also describe a smart image sensor newly designed for accuracy improvement.
    View full abstract
    Download PDF (1239K)
  • Japanese Consonant Recognition Based on Temporal Change in Lip Shape
    Yohei Miura, Yoshihiro Watanabe, Takashi Komuro, Masatoshi Ishikawa
    Article type: Article
    Session ID: IST2010-34/ME2010-10
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_21
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    In lip reading for Japanese language, consonant recognition is a still difficult problem. We are assuming that timing of temporal change of the feature generated from each frame has critical information to discriminant Japanese consonants. In this work, we proposed a simple novel feature for consonant recognition, and fed it to clustering process evaluating purity of the clustering and the possibility of discrimination in the feature space.
    View full abstract
    Download PDF (631K)
  • Registration of fish-eye photo and tree model in augmented reality system for forest survey
    Akihiko TANIKAWA, Mamoru KUBO, Ken-ichiro MURAMOTO
    Article type: Article
    Session ID: IST2010-35/ME2010-10
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_25
    CONFERENCE PROCEEDINGS FREE ACCESS
    Show abstractHide abstract
    In the forest survey, it is often that a researcher refers to a map and past archival records while performing a field work. However, the data which was researched in the past is stored as numerical values, and it is used not concurrently with the image photographed on the site. Therefore, this study develops the forest survey support system which exploits a portable terminal with a camera and supports the technology of the augmented reality. Such an equipment allows the researcher to grasp all the information by superimposing the virtual reality based on archival data with the display of verity environment. Additionally, geometric position adjustment is done between fish-eye photo photographed on the site and tree surveying data.
    View full abstract
    Download PDF (788K)
  • [title in Japanese]
    Article type: Appendix
    Pages App1-
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_App1
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (100K)
  • [title in Japanese]
    Article type: Appendix
    Pages App2-
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_App2
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (100K)
  • [title in Japanese]
    Article type: Appendix
    Pages App3-
    Published: June 22, 2010
    Released on J-STAGE: September 21, 2017
    DOIhttps://doi.org/10.11485/itetr.34.22.0_App3
    CONFERENCE PROCEEDINGS FREE ACCESS
    Download PDF (100K)
    • |<
    • <
    • 1
    • >
    • >|
feedback
 Top