Faculty Profiles - KAMIYA Tohru

Papers - KAMIYA Tohru

Division display381 - 400 of about 751／ All the affair displays >>

Transfer-learning deep convolutional neural network for classification of polyp candidates on CT colonography Reviewed

UEMURA Tomoki, LU Huimin, KIM Hyoungseop, TACHIBANA Rie, HIRONAKA Toru, Janne J. Näppi, YOSHIDA Hiroyuki

Medical Imaging and Information Sciences ( MEDICAL IMAGING AND INFORMATION SCIENCES ) 34 ( 2 ) 80 - 86 2017.01

Authorship：Corresponding author Language：Japanese Publishing type：Research paper (scientific journal)

Computed tomographic colonography（CTC）, also known as virtual colonoscopy, provides a minimally invasive screening method for early detection of colorectal lesions. It can be used to solve the problems of accuracy, capacity, cost,and safety that have been associated with conventional colorectal screening methods. Computer-aided detection（CADe）has been shown to increase radiologists' sensitivity and to reduce inter-observer variance in detecting colonic polyps in CTC. However, although CADe systems can prompt locations of abnormalities at a higher sensitivity than that of radiologists,they also prompt relatively large numbers of false positives（FPs）. In this study, we developed and evaluated the effect of a transfer-learning deep convolutional neural network（TL-DCNN）on the classification of polyp candidates detected by a CADe system from dual-energy CTC images. A deep convolution neural network（DCNN）that had been pre-trained with millions of natural non-medical images was fine-tuned to identify polyps by use of pseudo-colored images that were generated by assigning axial, coronal, and sagittal images of the polyp candidates to the red, green, and blue channels of the images, respectively. The classification performances of the TL-DCNN and the corresponding non-transfer-learning DCNN were evaluated by use of 5-fold cross validation on 20 clinical CTC cases. The TL-DCNN yielded true- and falsepositive rates of 73.6［％］and 1.79［％］, respectively, which were significantly higher than those of the non-transferlearning DCNN. This preliminary result demonstrates the effectiveness of the TL-DCNN in the classification of polyp candidates from CTC images.

DOI： 10.11318/mii.34.80

CiNii Article

CiNii Research

Other Link： https://ci.nii.ac.jp/naid/130006846731

Detection of Ground Glass Opacity Regions on LIDC Datasets Using DCNN Reviewed

HIRAYAMA Kazuki, LU Huimin, TAN Joo Kooi, KIM Hyoungseop, TACHIBANA Rie, HIRANO Yasushi, KIDO Shoji

Medical Imaging and Information Sciences ( MEDICAL IMAGING AND INFORMATION SCIENCES ) 34 ( 2 ) 70 - 74 2017.01

Authorship：Corresponding author Language：Japanese Publishing type：Research paper (scientific journal)

Lung cancer is one of the most important cancer in the world. Among them, Ground Glass Opacity（GGO）has a hazy area of increased attenuation in the lung image. In recent years, development of a Computer Aided Diagnosis （CAD）system for reducing the burden on work load and improving the detection rate of lesions has been advanced. In this paper, we propose a CAD system to extract GGO from CT images. Firstly, we extract the lung region from the input CT images and remove the vessel, and bronchial region based on 3 D line filter algorithm. After that, we extract initial GGO regions using concentration and gradient information. Next, we calculate the statistical features on the segmented regions. After that, we classify GGO regions using support vector machine（SVM）. Finally, we detect the final GGO regions using deep convolutional neural network（DCNN）. The proposed method is tested on 31 cases of CT images from the Lung Image Database Consortium（LIDC）. The results demonstrate that the proposed method has 86.05[%] of true positive rate and 39.03[/case] of false positive number.

DOI： 10.11318/mii.34.70

CiNii Article

CiNii Research

Other Link： https://ci.nii.ac.jp/naid/130006846732

Current distribution based power module screening by new normal/abnormal classification method with image processing Reviewed International journal

Tsukuda M., Yuki D., Tomonaga H., Kim H., Omura I.

Proceedings of the International Symposium on Power Semiconductor Devices and ICs 407 - 410 2017.01

Automatic segmentation of cell candidate region in Microscopy images based on an optimization algorithm Reviewed

International Conference on Control, Automation and Systems 720 - 723 2016.10

Extraction of GGO candidat regions from the LIDC database using deep learning Reviewed

724 - 727 2016.10

Identify rheumatoid arthritis and osteoporosis from phalange CR images based on image registration and ANN Reviewed

10 ( 10 ) 2435 - 2440 2016.10

Identify Rheumatoid Arthritis and Osteoporosis from Phalange CR Images Image　Based on Image Registration and ANN Reviewed International journal

Kajihara, Murakami, Tan, Kim, Aoki

ICIC Express Letters 10 ( 10 ) 2435 - 2440 2016.10

A Computer Aided Diagnosis System to Identify Rheumatoid Arthritis and Osteoporosis from Phalange CR Image Reviewed International journal

Kajihara, Murakami, Tan, Kim, Aoki

11th International Conference on Innovative Computing, Information and Control 2016.08

A method of detecting salient regions employing global and local saliency Reviewed

7 ( 3 ) 555 - 561 2016.03

Detecting pedestrian and extracting their attributes from self-mounted camera views Reviewed

7 ( 2 ) 279 - 286 2016.02

Traffic signs and signals detection employing the my vision system for a visually impaired person Reviewed

7 ( 2 ) 385 - 391 2016.02

Detection method for candidate regions of ground glass opacity on LIDC database using image features Reviewed

70 ( 8 ) J178 - J184 2016.01

Detection Method for Candidate Regions of Ground Glass Opacity on LIDC Database Using Image Features Reviewed

Yokota Keisuke, Kim Hyoungseop, Tan Joo Kooi, Ishikawa Seiji, Tachibana Rie, Hirano Yasushi, Kido Shoji, Aoki Takatoshi

The Journal of The Institute of Image Information and Television Engineers 70 ( 8 ) J178 - J184 2016.01

Detection of Rush-out Pedestrians Using MSC-HOG Features

ONO Yuta, TAN Joo Kooi, KIM Hyoungseop, ISHIKAWA Seiji

Proceedings of the Annual Conference of Biomedical Fuzzy Systems Association ( Biomedical Fuzzy Systems Association ) 29 ( 0 ) 37 - 40 2016.01

Language：Japanese Publishing type：Research paper (conference, symposium, etc.)

In recent years, the development of Intelligent Transportation Systems (ITS) has become more active than ever and many methods of detecting people and automobiles using an in-vehicle camera have been proposed to date. However, a method of predicting a pedestrian's rush-out into a road still faces a mountain of challenges. In this paper, we propose a method of detecting a pedestrian and judging his/her direction of walk employing MSC-HOG, an improved version of the HOG, and AdaBoost. The prediction of a pedestrian's rush-out is carried out by tracking each pedestrian and judging the change of his/her walk directions. The effectiveness of the proposed method is verified by the experiments employing car videos.

DOI： 10.24466/pacbfsa.29.0_37

CiNii Article

Other Link： https://ci.nii.ac.jp/naid/130007979886

Compensation of Vibration on a MY VISION Video Using Normal Distribution

YOSHIDA Naoto, TAN Joo Kooi, KIM Hyoungseop, ISHIKAWA Seiji

Proceedings of the Annual Conference of Biomedical Fuzzy Systems Association ( Biomedical Fuzzy Systems Association ) 29 ( 0 ) 45 - 48 2016.01

Language：Japanese Publishing type：Research paper (conference, symposium, etc.)

In recent years, wearable devices have been developing rapidly as intelligent tools. According to MIC, the wearable devices market is expected to be 15 times larger in 2020 than 2013. Among them, a wearable camera with a PC has been drawing much attention as a substitute for a human real time visual processing system. However, because such a camera is used by being mounted on a user, the vibration generated by the movement of the user affects captured video images and hence resulting in the difficulty in the video processing. This paper proposes a method of compensating the vibration of a video images captured by a wearable camera by detecting the background movement obtained from optical flow analysis. Experimental results are shown and the method is evaluated.

DOI： 10.24466/pacbfsa.29.0_45

CiNii Article

Other Link： https://ci.nii.ac.jp/naid/130007979807

Object Detection from a Moving Camera

MATSUO Takuya, TAN Joo Kooi, KIM Hyoungseop, ISHIKAWA Seiji

Proceedings of the Annual Conference of Biomedical Fuzzy Systems Association ( Biomedical Fuzzy Systems Association ) 29 ( 0 ) 23 - 26 2016.01

Language：Japanese Publishing type：Research paper (conference, symposium, etc.)

This paper presents a method of object detection using a moving camera. In order to perform background subtraction with a moving camera case, the proposed method estimates camera motion and creates a current background model from a previous background model using the camera motion. The background model is a set of a single Gaussian distribution. The method employs the upper and the lower bound of a threshold used for the judgment of background/foreground and color information to achieve precise detection of an object. We have confirmed the effectiveness of the proposed method by the experiments performed using some human motion videos.

DOI： 10.24466/pacbfsa.29.0_23

CiNii Article

Other Link： https://ci.nii.ac.jp/naid/130007979883

Human Action Recognition Using 3D-MHIs for Elderly Care

YAMASHITA Youtarou, TAN Joo Kooi, KIM Hyoungseop, ISHIKAWA Seiji

Proceedings of the Annual Conference of Biomedical Fuzzy Systems Association ( Biomedical Fuzzy Systems Association ) 29 ( 0 ) 33 - 36 2016.01

Language：Japanese Publishing type：Research paper (conference, symposium, etc.)

Nowadays the population of elderly people who live alone has been growing continuously. When they encounter an accident at home, they won't be able to ask for help in the worst case. Under these circumstances, development of an automatic system to care elderly people living alone becomes increasingly important. We have been developing a computer vision system which finds abnormal motions of elderly people among their daily activities indoors. In this paper, we propose a new method of human motion representation called 3D-MHIs (3-dimensional-Motion History Image) based on FoE (Focus of Expansion) and a method of human motion recognition employing 3-D Hu moments by use of a single camera view. The method can recognize not only those motions parallel to a camera lens but also those motions toward depth direction. The effectiveness of the proposed method was experimentally shown.

DOI： 10.24466/pacbfsa.29.0_33

CiNii Article

Other Link： https://ci.nii.ac.jp/naid/130007979895

Recognition of Finger-spelled Letters Using MY VISON System

HIRAKAWA Manabu, TAN Joo Kooi, KIM Hyoungseop, ISHIKAWA Seiji, Hamada SATOSI

Proceedings of the Annual Conference of Biomedical Fuzzy Systems Association ( Biomedical Fuzzy Systems Association ) 29 ( 0 ) 41 - 44 2016.01

Language：Japanese Publishing type：Research paper (conference, symposium, etc.)

This paper proposes a method of recognizing finger-spelled letters using the MY VISION system which contains a camera mounted on the chest of a human and a PC. A finger-speller performs the spelling to the camera. The images of the speller's hand are captured and hand areas are detected employing skin color. For the detection, skin color calibration is performed in advance to cope with various illumination conditions and cluttered backgrounds. The shape and motion of the hand area are analyzed to recognize the finger-spelled letters. The PCA and the nearest neighbor method are employed for the recognition. The high recognition rate with the finger-spelled letters was achieved by the proposed system along with the high speed recognition.

DOI： 10.24466/pacbfsa.29.0_41

CiNii Article

Other Link： https://ci.nii.ac.jp/naid/130007979809

スケール不変な顕著領域特徴に基づくCR画像からの指骨領域の自動位置合わせ Reviewed

梶原，村上，金，タン，石川

バイオメディカル・ファジィ・システム学会誌 17 ( 2 ) 35 - 42 2015.12

IGBTデバイスの目視検査支援システムの開発～不良品の細分類

結城，金，タン，石川，附田，大村

第３４回計測自動制御学会九州支部学術講演会 173 - 174 2015.11

　 PREV - NEXT 　

<KAMIYA Tohru>

Papers - KAMIYA Tohru