-
Tsubasa TAKAYAMA, Madoka HASEGAWA
Session ID: S1-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
Shortage of honeybees has become a serious problem in recent years. One of the causes of bee shortage is Varroa mites, which are parasitic on honeybees. The monitoring of Varroa mite parasitism is performed manually, therefore, it takes time and efforts. In addition, the accuracy varies depending on the inspector and the environment. In this paper, we report on an image-processing approach for detecting Varroa mites. We detect mites in images of drone pupae which are taken out from drone frames. Our method uses the template matching and several thresholds.
View full abstract
-
Hiroki MATSUMOTO, Madoka HASEGAWA
Session ID: S1-2
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
The parasitic damage of the varroa mite is a serious problem for honey bees, and the early detection
of mites is indispensable for its control. W e study on a basic method for the image based detecti on of mites
on the body surface of adult bees on the hive frame for beekeeping. Our method generates the s uper p ixel
from the hue components of the image and detect s mites based on the shape and color features. We report
the detection accuracy of mites in this paper.
View full abstract
-
Taiki NAGASAKI, Yuta MURAKI, Kenichi KOBORI
Session ID: S1-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
Jumpei NAKATSUKA, Yusuke FUNABIKI, Yuta MURAKI, Ken-ichi KOBORI
Session ID: S1-4
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
Masataka TOZUKA, Kunihiko TAKANO, Koki SATO
Session ID: R1-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
Interference fringe type Computer-Generated Holography, which is one of the technologies of Computer-Generated Holography, is a hologram that can record and reproduce information by interference fringes using amplitude and phase. This technique uses random noise to obtain a high-definition reconstructed image equivalent to the input image. In the Kinoform using only the phase of the conventional technique, the amplitude is constant, so the deterioration of the reconstructed image is larger than that of the former due to the influence of the quantization error. In this study, we realized digital holography that does not require an optical system by multiplexing using the lens term of the interference fringe type CGH and separating the reconstructed image by simulating the lens term. As a multiplexing technique, a method based on hologram information redundancy is well known. However, most of the methods use an optical system for separation, and there is no evidence of the feasibility of a holographic memory using this technology. Therefore, we confirm the problems of digital holography by conducting basic experiments on digital reconstructed image of holography. We also propose to improve the image quality of the reconstructed image by removing noise using image processing.
View full abstract
-
Masataka TOZUKA, Kunihiko TAKANO, Koki SATO
Session ID: R1-2
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
In recent years, the application of interference fringe type Computer-Generated Holography is expected to improve compression transmission technology. Existing modulation methods include phase shift keying for amplitude phase modulation and quadrature amplitude modulation. In this paper, 256 gradations of the input image data on the transmitting side are divided into a certain range, and the center value is obtained in each range. CGH was created from this data and transmitted in a compressed or uncompressed frequency division multiplexing system. Using a text file between transmission and reception, we made it an ideal transmission line that can round numbers to any number of digits after the decimal point. We propose a method for obtaining information based on the gradation of the reconstructed image of CGH. In order to use the proposed method for existing modulation methods, it is necessary to accurately transmit the gradation of the reconstructed image of CGH and faithfully reproduce the gradation before transmission as a prerequisite for communication quality. We have not found any data communication using the gradation of the reproduced image of interference fringe type CGH, or information compression in transmission using FDM of CGH. Therefore, it was examined by simulated transmission by simulation on a computer. In addition, the amount of information was compared with the 8-phase PSK of amplitude phase modulation. From these studies, we report that there is a possibility that it can be applied to data communication with a large amount of information.
View full abstract
-
Kunihiko TAKANO, Kazumi YOKOTA, Yuhki KASAI, Koki SATO, Kikuo ASAI
Session ID: R1-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
We have studied how to improve the characteristics of the displayable region of the reconstructed images using blue-violet laser light when time-sharing multiplex reconstructing technique is employed. In this report, we are going to show some improved displaying characteristics of the object with the curved shape, and to propose some useful basic suggestions for a colorful display of reconstructed holographic images when blue-violet laser light is applied.
View full abstract
-
Katsuya Hasegawa
Session ID: T1-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
COVID-19 has become a major problem in the world due to the global spread of the disease. COVID-19 is not the only infectious disease, but also many other infectious diseases such as malaria, Ebola, and dengue fever. JAXA has been conducting a joint research with Institut Pasteur (France) on prevention of the spread of infectious diseases in Asia and Africa, and we would like to introduce this project.
Translated with www.DeepL.com/Translator (free version)
View full abstract
-
Mitsuji MATSUMOTO, Hideo FUJITA, Abdelmoula BEKKALI, Mitikazu HATTORI
Session ID: T1-2
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
Pham Tien Dat
Session ID: T1-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
Maso ISSHIKI
Session ID: T1-4
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
-Student presentations using You Tube-
Yukiko HAGIO
Session ID: T1-5
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
I will introduce a trial of online remote learning, in which students' presentations were made available exclusively on You Tube
View full abstract
-
Takashi Mori
Session ID: T1-6
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
[in Japanese]
Session ID: T1-7
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
[in Japanese]
Session ID: T1-8
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
Biodegradable protective clothing that decomposes into CO2 and water does not need high-temperature incineration. Because dioxy by combustion does not occur, it is environmentally friend-ly.
In combination with a machine for microbial combustion of protective clothing, contaminated wear
Because it can be processed in a short period of time without moving, it is also possible to pre-vent the spread of infection.
View full abstract
-
Takuya IIYAMA, Youngha CHANG, Nobuhiko Mukai
Session ID: S2-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
Illumination estimation is one of the basic technologies in the field of image recognition . In this
paper, we propose a method to improve the estimation accuracy by selecting representative
illumination estimation method s and learning the estimated illuminations using linear SVR
(Support Vector Regression). We have evaluated the estimation accuracy of the several methods
with hyperspectral images rendered at multiple color temperatures, and have confirmed that the
estimation accuracy was improved.
View full abstract
-
Luis Peralta, Munetoshi Iwakiri, Kiyoshi Tanaka
Session ID: S2-2
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
The registration of 3D point clouds is a well-known and helpful task for digital reconstruction in multiple fields. Although many registration techniques nowadays are available, most do not assess low overlapping conditions and are prone to failure if they rely on point descriptors. Additionally, recently-proposed learning-based registration methods seem to succeed regardless of the overlapping ratio. However, these require large and ideally shaped datasets, making them difficult to utilize in real-world applications. We present the preliminaries to a descriptor-free pairwise registration method for low overlapping point clouds, based on the super voxel segmentation and Hough-voting techniques. The super voxel segmentation provides the basis for a divide-and-conquer principle by dividing the source point cloud into subsets, and Hough-voting the mechanism that looks for ideal transformations without requiring any point description, learning technique, or large datasets.
View full abstract
-
Takaya NAKAZAKI, Yuta MURAKI , Ken-ichi KOBORI
Session ID: S2-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
Hiroyuki IMADA, Naoto KAWAMURA, Hyunho KANG, Keiichi IWAMURA
Session ID: S2-4
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
In the digital watermarking method based on Green-Noise spectrum spread, the embedding amount increases when the block size is reduced, but the false positives increase. Therefore, we tried to improve the accuracy of the classifier by using machine learning. As a result, a high accuracy with a judgment correct answer rate of 99% or more was obtained for an image in which 256 characters (2048 bits) were embedded in a 512x512 image. Also, in the resistance evaluation, it was confirmed that the resistance was stronger than that of the conventional method.
View full abstract
-
Takuto SASSA, Madoka HASEGAWA
Session ID: S3-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
In recent years, the decrease of bee population has become a problem due to the effects of pesticides and other factors. In our laboratory, we are investigating a method to support bee monitoring in the beekeeping. We classify the hexagonal cells, called brood cells, from images of the honeycomb frame using CNN. In this paper, we investigated the classification accuracy for different CNN structures.
View full abstract
-
Shotaro ISHIWATA, Ryosuke FURUTA, Yukinobu TANIGUCHI
Session ID: S3-2
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
In recent years, the number of dairy farms has been decreasing in Japan. In order to reduce the burden on dairy farmers required for early detection of disease signs, methods have been developed for identifying individual dairy cows using image recognition based on machine learning. One of the problems in the conventional methods is that they require much time and effort to prepare training data every time dairy cows are replaced due to the entry of new dairy cows, etc., and it takes time and effort to create training data. In addition, since the judgment process is black-boxed, another problem is that it is difficult to interpret the result. In this paper, we introduce Attention Branch Network to generate visual explanations for individual identification and to improve identification accuracy. Furthermore, to clarify the effect of dairy cow replacement on the identification accuracy, assuming a situation close to practical use, we evaluated the accuracy while changing the proportion of labels not included in the training data.
View full abstract
-
Zhao Wentao, Jun Ohya, Zhang Zelin
Session ID: S3-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
In order to predict pedestrians' movements and help the pedestrian vehicle interaction, it is important to make a computer vision based classification for whether the pedestrian is looking at the vehicle. Conventional methods based on head orientation estimation only estimates the general orientation of the pedestrian head. Besides, the results are not good due to the low resolution of the pedestrian head image. Meanwhile, a large number of new features are extracted from the head images, which would increase the computation complexity. In this thesis, we propose a novel method that can predict the pedestrian's attention. The proposed method classifies whether the pedestrian is looking at the vehicle, using human pose estimation. The algorithm of proposed method consists of the following modules: (1) A subset of JAAD data set is selected as the dataset for this research. (2) As an existing estimation tool, Alphapose is used to obtain key points on the pedestrian's body. (3) In two successive frames, the pedestrian is tracked by Intersection over Union (IOU). (4) Pedestrian features based on the position relationship and confidence of key points in the successive frames are extracted from the result of Alphapose. (5) The model trained by random forest algorithm is used to classify the pedestrian's attention. Experimental results show that the proposed method can achieve a high accuracy for classifying the pedestrian attention.
View full abstract
-
Kai MATSUI, Katsuya HASEGAWA, Shinta MUTO, Jun OHYA
Session ID: S3-4
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
In recent years, due to frequent sediment disasters and lack of manpower in rural areas, some forest roads
have been left collapsed. It is required to reduce the amount of labor needed to patrol forest roads for their
maintenance.
In order to achieve this goal, we have started to study a method of detecting the collapse of the ground
surface by comparing before and after a disaster using a 3D reconstruction of the ground surface from
camera and sensors attached to a drone.
This paper presents a fundamental study of the method for detecting the location of a sediment disaster by
imitating a topographical change.
View full abstract
-
Madoka Inoue, Yusuke Doiguchi, Rio Tamahiro, Tomomi Nakajima, Kazuki M ...
Session ID: T2-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
We conducted a preliminary survey to purify the water quality of the pond at Mitsuike Park, and it was found that the oxygen concentration was decreasing. As a result, sludge accumulates on the bottom of the pond, causing deterioration of the landscape and offensive odors. Therefore, we investigated the formation process and properties of sludge in the literature and examined how to eliminate it.
View full abstract
-
Kazuki Terayama, Sora Ohashi, Kazuma Tsuchiya, Toshiki Sato, Yui Isoga ...
Session ID: T2-2
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
The Tsurumi High School Biology club is going to participate in a water purification plan for a pond in an park. Prior to this plan, we conducted a preliminary survey on the environment of the pond in order to clarify the problem and tasks we should overcome in purifying water. We investigated on water quality and aquatic organisms, and conducted fieldwork in the field to conclude that it is necessary to eliminate the decrease in oxygen concentration and the overgrowth of foreign water lilies.
View full abstract
-
Muneyoshi Shibagaki
Session ID: T2-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
We will explain about the eco― house that we are doing in the United States where you can live only with
natural energy( rain, wind , sun).
View full abstract
-
Etsuko Kawanishi
Session ID: T2-4
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
setsuko warita
Session ID: T2-5
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
[in Japanese]
Session ID: T2-6
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
[in Japanese]
Session ID: T2-7
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
shuhei warita
Session ID: T2-8
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
Keiko Mizukoshi
Session ID: T2-9
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
Makoto J. HIRAYAMA
Session ID: T3-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
Discussions on image museums and digital humanities are done with several cases related to them. Topics are provided from cultural views for discussions in a special session, organized by SIG on digital signages and interactions, at the 2020 IIEEJ annual conference.
View full abstract
-
Kazuhiro Watanabe
Session ID: T3-2
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
Regional revitalization is an urgent problem in Japan.This study considered that drone technology for
problem solving and effective Aerial Photography.
View full abstract
-
Toshiko KIMURA
Session ID: T3-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
A survey of students regarding the Johari window, which is one of the psychological models is conducted by the writer. She tried to use the app and understood its effectiveness for their career. Although specific data could not be introduced this time from the viewpoint of personal information, she will consider a disclosure method.
View full abstract
-
Kunio OHNO
Session ID: T3-4
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
Rhetoric has traditionally focused only on textual information, but in this paper it has been extended to compound document and Web content to include image and video information. The idea has been modeled by applying Minsky's frame theory.
View full abstract
-
Racco
Session ID: T4-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
An affinity of Wikipedia Town and libraries.
View full abstract
-
Takenori NOGUCHI
Session ID: T4-2
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
In 2019, I conducted a questionnaire survey on the actual situation of the reading environment for local governments consisting only of remote islands. I also visited and interviewed several local governments. In this report, I will report on the current situation and issues of public libraries and school libraries from the results of the questionnaire survey, and introduce some cases from the results of the visit interview survey.
View full abstract
-
Yashio UEMURA
Session ID: T4-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
Tsuyoshi OHTA
Session ID: T4-4
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
- JPEG AI, Fake media, DNA storage -
Osamu WATANABE
Session ID: OS1-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
The most recent status of the standardization activities in JPEG (ISO/IEC JTC 1/SC 29/WG 1) is described. In particular, the new explorations such as JPEG AI, Fake media and DNA Storage are summarized with the purposes, use cases and requirements.
View full abstract
-
Osamu WATANABE
Session ID: OS1-2
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
HTJ2K (ISO/IEC 15444-15) defines a new block encoder/decoder as replacement of corresponding EBCOT encoder/decoder. The new block codecs make it possible to achieve significant improvement of codec’s throughput. A summary, trend of adoption and possible applications in industries are given in this talk.
View full abstract
-
: Standardization Status of ISO/IEC 19566 (JPEG Systems)
Junichi HARA
Session ID: OS1-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
Toshiaki FUJII
Session ID: OS1-4
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
Three-dimensional (3D) image has been attracting much attention as a future visual media. This presentation addresses a problem that 3D image requires tremendous amount of data, which can be tens or hundreds times of data than that of 2D image. We first introduce 3D image coding technologies which have been investigated in this field, and then describe trends and future directions of the 3D image coding technologies, especially in the international standardization activities, such as JPEG and MPEG.
View full abstract
-
-The First Results of Event Horizon Telescope Collaboration-
Fumie TAZAKI, Event Horizon Telescope Collaboration
Session ID: Q-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
-
Taisei INOUE, Norimasa YOSHIDA, Motonori ISHIBASHI
Session ID: P-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
From the perspective of improving work efficiency, human characteristics are expected to be measured simply and quickly. In a previous work, we examined the relationship between task of drawing a straight line for a given point cloud and the error proneness (action slip (AS), cognitive narrowing (CN), and impulsive failure (IF)) by utilizing a tablet device. For specific point cloud, we found that there is a relationship between the distance to the barycenter of the point cloud from the drawn line and the AS score (r=0.55). In this study, we estimate the distance to the barycenter of the point cloud from the drawn line, from the scores of the error proneness questionnaire, the number of points in the point cloud referenced and the features of the point cloud, by using multiple regression analysis.
View full abstract
-
Tomoki AKITA, Norimasa YOSHIDA
Session ID: P-2
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
In recent years, there has been a lot of talk about interactive teaching. Some students are passive in their regular lectures. In an interactive class, the faculty member communicates to that student. In this way, students can discover problems and challenges themselves and are given the opportunity to learn on their own. Seminar can be made better by having interactive classes. We propose a system that can determine the status of the seminar in real time. In this study, we propose a system that creates a spectrogram from recorded speech data of a seminar, identifies speakers by deep learning, and summarizes and displays the results.
View full abstract
-
Itsuki Tanaka, Munetoshi Iwakiri, Kiyoshi Tanaka
Session ID: P-3
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
The 3D point cloud is a field of computer vision,and the demand for it has been increasing in recent years due to the spread of 3D surveying instruments.Since point cloud data shows the shape of an object,it is expected to be applied in various fields such as robots,architecture / civil engineering,transportation,and agriculture.In this research,we focused on the technology for reconstructing 3D point cloud data from moving image series among the SfM (Structure from Motion) technologies for reconstructing 3D point cloud data from images.When reconstructing a 3D model from an image,the processing time and accuracy of the reconstruction differ depending on the image used,and it is necessary to select an appropriate image for efficient reconstruction.However,it is difficult to manually select an appropriate image from many image groups.In this study,we investigated a method of reconstructing a three-dimensional point cloud by selecting a frame using the motion vector contained in motion picture.The experimental results using some moving image data are shown and given consideration.
View full abstract
-
Yohei Saito, Kazuma Uenishi, Munetoshi Iwakiri, Aguirre Hernan, Kiyos ...
Session ID: P-4
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
3D point cloud registration is the process of integrating clouds obtained from multiple viewpoints into one coordinate system. In this process, the part that doesn’t overlap among clouds adversely affects the result. To avoid adversely effects, we propose a method to adaptively change the position of the keypoint-patches during registration process. The ICP that incorporation the proposed method is faster than the conventional evolutionary computation registration using the keypoint-patches and less dependence on the initial position than conventional ICP.
View full abstract
-
Vy Nguyen, Yongqing Sun, Hideaki Kimata
Session ID: I-1
Published: 2020
Released on J-STAGE: January 10, 2022
CONFERENCE PROCEEDINGS
RESTRICTED ACCESS
NTT summer internship gave me a great opportunity to
work on the image translation problem between aerial
images and map images . Out of all the state of the art
methods, MUNIT proposed a framework to utilize
disentangle learning in the image translation problem by
decomposing the traditional feature space into the content
sub space and style subspace As aerial images and map
images are supposed to share the same content sub space,
w e implement MUNIT on the problem of image
translation between aerial images and map images and
analyze the experimental results.
View full abstract