IIEEJ Transactions on Image Electronics and Visual Computing
Online ISSN : 2188-1901
Print ISSN : 2188-1898
ISSN-L : 2188-191X
7 巻, 2 号
IIEEJ Transactions on Image Electronics and Visual Computing
選択された号の論文の9件中1~9を表示しています
  • Eleni ALOUPOGIANNI, Hiroyuki SUZUKI, Takaya ICHIMURA, Atsushi SASAKI, ...
    原稿種別: Contributed Papers --Special Issue on Journal Track Papers in IEVC2019--
    2019 年 7 巻 2 号 p. 58-66
    発行日: 2019/12/15
    公開日: 2021/03/31
    ジャーナル 認証あり

    This study suggests an analysis procedure on macropathology multi-spectral images(macroMSI), for visual representation of grossly malignant regions of skin samples during excision margin pathological diagnosis. We implemented binary malignancy classification on a database of ten highresolution 7-channel macroMSI tissue samples, captured before and after formalin fixing. We reconstructed spectral reflectance by Wiener estimation and described texture using local binary patterns (LBP). Highlighted malignancy regions were derived from an optimal classifier selected by cross-validated performance.The results show that malignant regions are highlighted fairly accurately and indicate the importance of analyzing unfixed tissue in conjunction with fixed tissue.

  • Jaime SANDOVAL, Kazuma UENISHI, Munetoshi IWAKIRI, Kiyoshi TANAKA
    原稿種別: Contributed Papers --Special Issue on Journal Track Papers in IEVC2019 --
    2019 年 7 巻 2 号 p. 67-77
    発行日: 2019/12/15
    公開日: 2021/03/31
    ジャーナル 認証あり

    Planes detection in unorganized point clouds is a fundamental and essential task in 3D computer vision. It is a prerequisite in a wide variety of tasks such as object recognition, registration,and reconstruction. Conventional plane detection methods are remarkably slow because they require the computation of point-wise normal vectors and are non-deterministic due to their dependency on random sampling. Therefore, we propose a drastically more efficient and deterministic approach based on slidingvoxels. A sliding voxel is an overlapping grid structure in which we analyze the planarity of the points distributions to extract hypothetical planes efficiently. Each possible plane is validated globally by weighing and comparing its co-planarity with other sliding voxels’ planes. Experimental results with simulated and realistic point clouds confirmed that the proposed method is several times faster, more accurate, and more robust to noise than conventional methods.

  • Luis PERALTA, Jaime SANDOVAL, Munetoshi IWAKIRI, Kiyoshi TANAKA
    原稿種別: Contributed Papers --Special Issue on Journal Track Papers in IEVC2019 --
    2019 年 7 巻 2 号 p. 78-87
    発行日: 2019/12/15
    公開日: 2021/03/31
    ジャーナル 認証あり

    Since its introduction, the Iterative Closest Points (ICP) algorithm has led to developing a wide range of registration methods, most of these variations of ICP itself. Notwithstanding the efforts on improving the speed and accuracy of ICP, these variations cannot correctly align point clouds which overlapping ratio is considered low (under 40%) due to an inherited local minima convergence. Furthermore,more advanced registration techniques that rely on point descriptors also cannot overcome this problem because the tuning of their parameters tends to be volatile, which leads to making false point correspondences and consequently failing to perform an accurate registration. In order to solve this problem, we propose a pairwise registration approach that does not entirely rely on point descriptors and leverages the local minima convergence of ICP to correctly align 3D point clouds with overlapping ratios as low as about 20%.Our method uses the supervoxel segmentation technique to divide the point clouds into subsets and finds those which registration maximizes the overlapping ratio between correct correspondences in the full point clouds. We verified the effectiveness of the proposed method through tests in dense models and real-world scan datasets.

  • Natsuki TAKAYAMA, Hiroki TAKAHASHI
    原稿種別: Contributed Papers --Special Issue on Journal Track Papers in IEVC2019 --
    2019 年 7 巻 2 号 p. 88-96
    発行日: 2019/12/15
    公開日: 2021/03/31
    ジャーナル 認証あり

    The reduction of the manual work of annotation is an essential part of sign language recognition research. This paper describes one weakly-supervised learning approach for continuous sign language word recognition. The proposed method consists of forced alignment based on dynamic time warping and isolated word hidden markov model adjustment using ‘embedded training’. While the proposed forced alignment only requires one manual annotation for each isolated sign language word, it can generate sufficient quality of the annotation to initialize isolated word hidden markov models. ‘Embedded training’adjusts initial hidden markov models to recognize continuous sign language words using only ordered word labels. The performance of the proposed method is evaluated statistically using a dataset that includes 5,432 isolated sign language word videos and 4,621 continuous sign language word videos. The averaged alignment error of the proposed forced alignment was 4.02 frames. The averaged recognition performances of the initial models were 74.82% and 91.14% in the signer-opened and trial-opened conditions, respectively.Moreover, the averaged recognition performances of the adjusted models were over 65.00% for all conditions.The evaluation shows significant improvements compared to the previous weakly-supervised learning.

  • Toshiki KIKUCHI, Yuko OZASA
    原稿種別: Contributed Papers --Special Issue on Journal Track Papers in IEVC2019--
    2019 年 7 巻 2 号 p. 97-105
    発行日: 2019/12/15
    公開日: 2021/03/31
    ジャーナル 認証あり

    Despite the recent success in deep neural networks on the visual domain, we need a large amount of data to train the networks. Previous works addressed this issue as the few-shot learning which is the task to identify the class of an example in new classes not seen in a training phase with only a few examples of each new class. Some methods performed well on the few-shot tasks, but need a complex architecture and/or specialized loss functions, such as metric loss, meta learner, and memory. In this paper,we evaluate the performance of the ensemble approach aggregating a huge number of simple neural network models (up to 128 models) on standard few-shot datasets. Surprisingly, although the approach is simple, our experimental results show that the ensemble approach is competitive with state-of-the-art methods among similar architecture methods in some settings.

  • Saeko SHINOZAKI, Masanori NAKAYAMA, Issei FUJISHIRO
    原稿種別: Contributed Papers --Special Issue on Journal Track Papers in IEVC2019--
    2019 年 7 巻 2 号 p. 106-115
    発行日: 2019/12/15
    公開日: 2021/03/31
    ジャーナル 認証あり

    Real paper gets deformed anisotropically due to a uni-directional placement of the inner fibers. Few existing CG researchers allow for such fibers to represent virtual papers, and thus anisotropic paper deformation has not been well represented. In this study, we present a two-dimensional visual simulation model for anisotropic papers, which abstracts mutual mechanical relationships among intersecting fibers by a network of filler and hinge springs, and incorporates connection points for keeping the shape of each bended fiber. By releasing the network connections when the filler and hinge spring extend above a certain limit, the paper provides a plausible tear. We succeeded in generating a different appearance of the torn-off line of papers according as the pulling direction.

  • Malik Olivier BOUSSEJRA, Rikuo UCHIKI, Shunya TAKEKAWA, Kazuya MATSUBA ...
    原稿種別: System Development Paper --Special Issue on Journal Track Papers in IEVC2019--
    2019 年 7 巻 2 号 p. 116-127
    発行日: 2019/12/15
    公開日: 2021/03/31
    ジャーナル 認証あり

    This paper describes an extendable graphical framework, “aflak”, which provides a collaborative visualization environment for the analysis of multi-spectral astronomical datasets. aflak allows the astronomer to share and define analytics pipelines through a node editing interface, in which the user can compose together a set of built-in transforms (e.g. dataset import, integration, Gaussian fit) over astronomical datasets. Not only is aflak fast and responsive, but its macro can be conveniently exported, imported and shared among researchers using a custom data interchange format. aflak, while providing domain-specific features for astronomy, enables collaboration through shareability, guarantees end-to-end provenance management and achieves astronomer-in-the-loop. This paper compares aflak with conventional tools used in astronomy for specific use cases, such as the computing of equivalent widths. Finally, the paper demonstrates that aflak provides a reference implementation for the ProvenanceDM data model.

  • Ji WANG, Yoshiyuki YASHIMA
    原稿種別: Contributed Papers
    2019 年 7 巻 2 号 p. 128-141
    発行日: 2019/12/15
    公開日: 2021/03/31
    ジャーナル 認証あり

    Sparse coding is a technique that represents an input signal as a linear combination of a small number of atoms in the dictionary. When sparse coding is applied to image compression, it is necessary to perform efficient code assignment taking into account the statistical properties of weighting factors for each atom. In this paper, we analyze in detail the position indices and magnitude of non-zero coefficients in a dictionary designed by K-SVD. Based on the analyzed results, we propose an efficient entropy coding method introducing sparsity adaptation and atom reordering. Simulation results show that the proposed method can reduce the amount of generated bits by up to 6.2% compared to the conventional methods.

  • Yoshimitsu NAGAO, Yanan GAO, Jiang LIU, Shigeru SHIMAMOTO
    原稿種別: Contributed Papers
    2019 年 7 巻 2 号 p. 142-150
    発行日: 2019/12/15
    公開日: 2021/03/31
    ジャーナル 認証あり

    It is impossible to estimate arterial oxygen saturation (i.e., SpO2) for individuals by using conventional approaches unless the given sensor of the pulse oximeter is attached to an individual’s finger. This study introduces a novel method to solve this problem. This study has focused on realizing SpO2 measurements by using non-contact space measurements, and the success of the approach is validated through experiments. Finally, despite a few problems including the susceptibility of the proposed approach to other light interference, the study offers an initial method to utilize laser wavelengths for the fore-mentioned purposes. As the characteristic of elderly individuals involves the hardening of the fingertips’ skin, it is difficult for the light of a probe to enter the same. Therefore this study can be applied to medical care, elder care, and other related fields. .Additionally, there are cases that symptoms are unmeasurable. The light receiving property to other light interference in space constitutes a problem for the fore-mentioned method.

feedback
Top