研究者詳細 - 神谷　亨

論文 - 神谷　亨

分割表示757 件中 301 - 320 件目／全件表示 >>

Swallowing motion analyzing from dental MR imaging based on AKAZE and particle filter algorithm 査読有り国際誌

Suetani K., Lu H., Tan J., Kim H., Tanaka T., Kitou S., Morimoto Y.

International Conference on Control, Automation and Systems 2018-October 1343 - 1346 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. In recent years, dysphagia is problem among elderly people. Therefore, it is necessary to accurately evaluate swallowing function in order to prevent swallowing disorder beforehand or to detect it early. And it is considered that evaluation of swallowing function using Magnetic Resonance Imaging (MRI) is useful. In order to accurately analyzing of the swallowing motion using a computer aided diagnosis (CAD) system on MR imaging, automatic extraction of the esophagus region, which is a region of interest by the image analysis method, is required. Extraction of the spinal region is required as a preliminary step of the esophagus region extraction. Therefore, in this paper, we develop an analysis method of swallowing movement by three steps of extraction of spinal region, extraction of esophageal region, and analysis of swallowing movement. As an analytical method of swallowing movement, we emphasize the liquid part at the time of swallowing movement using an emphasis map, then follow the liquid tracing by using the AKAZE feature quantity and the particle filter algorithm, and analyze the swallowing motion.

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060441189&origin=inward

Wide residual networks for semantic segmentation 査読有り国際誌

Nakayama Y., Lu H., Li Y., Kim H.

International Conference on Control, Automation and Systems 2018-October 1476 - 1480 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. In the task of object recognition, convolutional neural networks (CNNs) have achieved high performance. In addition, these CNNs are also applied to the field of semantic image segmentation. However, applying the classification models to semantic segmentation tasks has a problem, lack of global context and reduction in resolution. In this work, we propose global context module and high resolution path in order to solve above problems. By simply combining them with an existing classification model (wide residual networks), our methods yield high-accuracy segmentation models. Our proposed approaches produce competitive results, the mean intersection over union (IoU) 67.6% and global accuracy 91.1%, on CamVid test set.

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060445374&origin=inward

Extraction of GGO Candidate Regions on Thoracic CT Images using SuperVoxel-Based Graph Cuts for Healthcare Systems 査読有り国際誌

Lu H., Kondo M., Li Y., Tan J., Kim H., Murakami S., Aoki T., Kido S.

Mobile Networks and Applications 23 ( 6 ) 1669 - 1679 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018, Springer Science+Business Media, LLC, part of Springer Nature. In this paper, we propose a method to reduce artifacts on temporal difference images by improving the conventional method using a non-rigid registration method for ground glass opacification (GGO), which is light in concentration and difficult to detect early. In this method, global matching, local matching, and 3D elastic matching are performed on the current image and past image, and an initial temporal difference image is generated. After that, we use an Iris filter, which is the gradient vector concentration degree filter, to determine the initial GGO candidate regions and perform segmentation using SuperVoxel and Graph Cuts in which a superpixel is extended to three dimensions for each region of interest. For each extracted region, a support vector machine (SVM) is used to reduce the over-segmentation. Finally, in the method that greatly reduces artifacts other than the remaining GGO candidate regions, Voxel Matching is applied to generate the final temporal difference image, emphasizing the GGO regions while reducing the artifact. The resulting ratio of artifacts to lung volume is 0.101 with an FWHM of 28.3, which is an improvement over the conventional method and shows the proposed method’s effectiveness.

DOI： 10.1007/s11036-018-1111-2

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85053445997&origin=inward

細胞領域の論理積を用いた蛍光顕微鏡画像からの血中循環がん細胞の自動検出査読有り

辻、陸、タン、金、米田、田中

医用画像情報学会誌 34 ( 4 ) 151 - 155 2018年12月

Segmentation of Spinal Canal Region in CT Images using 3D Region Growing Technique 査読有り国際誌

Fu G., Lu H., Tan J., Kim H., Zhu X., Lu J.

2018 International Conference on Information and Communication Technology Robotics, ICT-ROBOT 2018 2018年11月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2018 IEEE. Tumor of spinal cord (Spinal Cord Neoplasms (SCN)) is considered as one of the life threatening diseases that causes death. Early detection of the SCN plays an important role in the management of the lesions. To analyze the treatment, it is necessary to segment the spinal canal based on accurate three-dimensional image processing technique. This paper presents a segmentation algorithm based on 3D region growing for extracting spinal canal from CT images with high accuracy. Intersection over union (IoU) is used to compare the results of segmentation with the manual segmentation results. In the experiment, the proposed method was tested on 3373 CT slices of 10 patients. The proposed method has an average accuracy of 0.7732 and a variance of 0.0061. Satisfactory results have been achieved rapidly, which demonstrates the effectiveness and superiority of the proposed method.

DOI： 10.1109/ICT-ROBOT.2018.8549913

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060055563&origin=inward

Object Detection on Video Images Based on R-FCN and GrowCut Algorithm 査読有り国際誌

Mouri K., Lu H., Tan J.K., Kim H.

2018 International Conference on Information and Communication Technology Robotics, ICT-ROBOT 2018 2018年11月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Since the declining birthrate and the aging of society, there is concern about the labor shortage in Japan. There is a movement to compensate for the labor shortage by automation of factories by robots. Automation technique is wildly promoted in logistics industry, while there is few studies in objects picking. To solve this issue, we develop an image detection scheme for robotics picking from a video image. It is difficult to recognize and grasp different types of objects in robot vision field. Therefore, in the proposed method, object detection and object recognition method are proposed using Region-based Fully Convolutional Networks that is a type of object detection using deep learning. After detecting the object individually, final target object can select by applying the GrowCut algorithm. As a result, we achieve 0.6773 of the average precision and 0.6395 of Intersection over Union as the segmentation result respecively.

DOI： 10.1109/ICT-ROBOT.2018.8549879

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060010049&origin=inward

実数値GAに基づく指骨CR画像の位置合わせ

川越、村上、陸、タン、金、青木

第38回医療情報学連合大会抄録集 388 - 391 2018年11月

CT temporal subtraction method for detection of sclerotic bone metastasis in the thoracolumbar spine 査読有り国際誌

Ueno M., Aoki T., Murakami S., Kim H., Terasawa T., Fujisaki A., Hayashida Y., Korogi Y.

European Journal of Radiology 107 54 - 59 2018年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018 Elsevier B.V. Purpose: To assess the effectiveness of a CT temporal subtraction (TS) method on radiologists’ performance in sclerotic metastasis detection in the thoracolumbar spine. Materials and methods: 20 pairs (current and previous CTs) of standard-dose CT and their TS images in patients with sclerotic bone metastasis and 20 pairs (current and previous CTs) of those in patients without bone metastasis were used for an observer performance study. A total of 135 lesions were identified as the reference standard of actionable lesions (sclerotic metastasis newly appeared or increased in size or in attenuation). 4 attending radiologists and 4 radiology residents participated in this observer study. Ratings and locations of “lesions” determined by the observers were utilized for assessing the statistical significance of differences between radiologists’ performances without and with the CT-TS images in JAFROC analysis. The statistical significance of differences in the reviewing time was determined by a two-tailed paired t-test. Results: The average figure-of-merit (FOM) values for all but one radiologist increased to a statistically significant degree, from 0.856 without the CT-TS images to 0.884 with the images (P =.037). The average sensitivity for detecting the actionable lesions was improved from 60.7 % to 72.5% at a false-positive rate of 0.15 per case by use of the CT-TS images. The average reading time with CT-TS images was significantly shorter than that without (150.6 s vs. 166.5 s, P =.004). Conclusion: The use of CT-TS would improve the observer performance for the detection of the sclerotic bone metastasis in the thoracolumbar spine.

DOI： 10.1016/j.ejrad.2018.07.017

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85051670298&origin=inward

Statistical shape model generation using K-means clustering 査読有り国際誌

Wu J., Li G., Lu H., Kim H.

ACM International Conference Proceeding Series 207 - 211 2018年09月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2018 Association for Computing Machinery. Statistical shape models (SSMs) is a robust and efficient method in medical image segmentation. In this paper, a novel landmark corresponding method based on k-means clustering and demons registration is proposed to train a 3-D statistical shape model with higher quality. The k-means clustering method is performed on the original geometric surface to obtain a simplified surface as standard set of landmarks to find correspondent landmarks on each mapped spherical surface obtained from demon registration in the training set. Twenty cases of left lung and right lung regions in thoracic MDCT images are used in the experiment to build two SSMs. Performance evaluation results show that SSMs generated by the proposed method achieve better generalization ability and specificity while maintaining the same compactness and accuracy of segmentation as those reported by state-of-the-art methods.

DOI： 10.1145/3277453.3277467

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060643606&origin=inward

R-FCNとGrowCutを用いたボールペンの検出

毛利、陸、金

産業応用工学会全国大会2018講演論文集 85 - 86 2018年09月

Object Detection on Video Images Based on GrowCut Algorithm 査読有り国際誌

Mouri, Lu, Tan, Kim

ICT-Robot 2018年09月

Activity representation by SURF-based templates 査読有り国際誌

Ahad M., Tan J., Kim H., Ishikawa S.

Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization 6 ( 5 ) 573 - 583 2018年09月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018, © 2018 Informa UK Limited, trading as Taylor & Francis Group. This paper presents a method to recognise actions, which are overlapping and multi-dimensionalities. A spatio-temporal representation is illustrated on local interest points to compute global features. Motion history image (MHI) is computed and motion overwriting the motion overwriting problem of the MHI. The main contribution of this paper is that it demonstrates a higher discriminative ability of various complex actions when compared to the other MHI-based approaches. It selects local interest feature points to capture motion information using Speeded-Up Robust Features (SURF). These key interest points are exploited to compute gradient-based optical flow into four channels. RANSAC is exploited to remove outliers. It incorporates frame-subtracted accumulated image so that we can mask out points that are not required. Afterwards, feature vectors are computed based on moments. Actions are recognised by employing a nearest neighbour classifier and leave-one-out cross-validation partitioning scheme. The proposed method provides satisfactory recognition rates over several other approaches for some challenging actions in outdoor scenes.

DOI： 10.1080/21681163.2017.1298472

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85016247850&origin=inward

FDCNet: filtering deep convolutional network for marine organism classification 査読有り国際誌

Lu H., Li Y., Uemura T., Ge Z., Xu X., He L., Serikawa S., Kim H.

Multimedia Tools and Applications 77 ( 17 ) 21847 - 21860 2018年09月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017, Springer Science+Business Media New York. Convolutional networks are currently the most popular computer vision methods for a wide variety of applications in multimedia research fields. Most recent methods have focused on solving problems with natural images and usually use a training database, such as Imagenet or Openimage, to detect the characteristics of the objects. However, in practical applications, training samples are difficult to acquire. In this study, we develop a powerful approach that can accurately learn marine organisms. The proposed filtering deep convolutional network (FDCNet) classifies deep-sea objects better than state-of-the-art classification methods, such as AlexNet, GoogLeNet, ResNet50, and ResNet101. The classification accuracy of the proposed FDCNet method is 1.8%, 2.9%, 2.0%, and 1.0% better than AlexNet, GooLeNet, ResNet50, and ResNet101, respectively. In addition, we have built the first marine organism database, Kyutech10K, with seven categories (i.e., shrimp, squid, crab, shark, sea urchin, manganese, and sand).

DOI： 10.1007/s11042-017-4585-1

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85015699675&origin=inward

BrainNets: Human Emotion Recognition Using an Internet of Brian Things Platform 査読有り国際誌

Lu H., Kim H., Li Y., Zhang Y.

2018 14th International Wireless Communications and Mobile Computing Conference, IWCMC 2018 1313 - 1316 2018年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2018 IEEE. Human wearable helmet is a useful tool for monitoring the status of miners in the mining industry. However, there is little research regarding human emotion recognition in an extreme environment. In this paper, an emotional state evoked paradigm is designed to identify the brain area where the emotion feature is most evident. Next, the correct electrode position is determined for the collection of the negative emotion by the electroencephalograph (EEG) based on the international 10-20 system of electrode placement. And then, a fusion algorithm of the anxiety level is proposed to evaluate the person's mental state using the θ, α, and β rhythms of an EEG. Experiments demonstrate that the position Fp2 is the best electrode position for obtaining the anxiety level parameter. The most visible EEG changes appear within the first two seconds following stimulation. The amplitudes of the θ rhythm increase most significantly in the negative emotional state.

DOI： 10.1109/IWCMC.2018.8450382

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85053937610&origin=inward

Automatic detection of cell regions in microscope images based on BFED algorithm 査読有り国際誌

Nakamichi K., Lu H., Kim H., Yoneda K., Tanaka F.

ACM International Conference Proceeding Series 38 - 42 2018年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2018 Association for Computing Machinery. Circulating tumor cells (CTC) attract attention as a biomarker that can evaluate cancer metastasis and therapeutic effects. The CTC exists in the blood of cancer patients, so pathologists analyze blood by using a fluorescence microscope. However, manual analysis by pathologists is hard-work since the number of CTC to substances contained in the blood is very few and the cell regions are often unclear depending on shooting environments. In addition, there are few studies on automatic identification of CTC. In this paper, we develop an automatic detection method of cell regions in microscope images based on bacterial foraging-based edge detection (BFED) algorithm to analyze CTC. In the first step, we detect the initial cell regions by BFED algorithm. Second, we identify whether the region is a single cell or multiple cells come in connect with other cell(s) by SVM. Third, when a cell is connected with other one, we separate the connecting cells by branch and bound algorithm and obtain the final cell regions. We applied our proposed method to 1680 microscopy images (6 cases). The experimental results demonstrate that the proposed method has a true positive rate of 93.9% and a false positive 1.29 /case.

DOI： 10.1145/3278229.3278232

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85058561021&origin=inward

Environment recognition for electric wheelchair based on YOLOv2 査読有り国際誌

Sakai Y., Lu H., Tan J., Kim H.

ACM International Conference Proceeding Series 112 - 117 2018年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2018 Association for Computing Machinery. At present, the aging population is growing in Japan. Along with that, the need for the utilization of welfare equipment is increasing. Electric wheelchair, a convenient transportation tool, is popularized rapidly. However, many accidents have occurred by using electric wheelchair, and the dangers for driving are pointed out. Therefore, it needs to improve accident factors, reduce accidents and improve the convenience of electric wheelchair by automation. Environmental recognition is the key technology for developing autonomous electric wheelchair. Environmental recognition includes self-position estimation, recognition of sidewalks, crosswalks, traffic lights, and moving object prediction, etc. In order to solve these problems, this paper describes a system for the detection of sidewalks, crosswalks and traffic lights. We develop the object recognition methods using a modified YOLOv2 that is one of object detection algorithms applying convolutional neural networks (CNN). We detect the object through YOLOv2 and perform processing such as unnecessary bounding box deletion and interpolation. The experimental results demonstrate that the area under the curve (AUC) of the detection rate is 0.620.

DOI： 10.1145/3278229.3278231

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85058546828&origin=inward

Preface 招待有り査読有り国際誌

Kim H.

ACM International Conference Proceeding Series 2018年08月

Motor anomaly detection for unmanned aerial vehicles using reinforcement learning 査読有り国際誌

Lu H., Li Y., Mu S., Wang D., Kim H., Serikawa S.

IEEE Internet of Things Journal 5 ( 4 ) 2315 - 2322 2018年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017 IEEE. Unmanned aerial vehicles (UAVs) are used in many fields including weather observation, farming, infrastructure inspection, and monitoring of disaster areas. However, the currently available UAVs are prone to crashing. The goal of this paper is the development of an anomaly detection system to prevent the motor of the drone from operating at abnormal temperatures. In this anomaly detection system, the temperature of the motor is recorded using DS18B20 sensors. Then, using reinforcement learning, the motor is judged to be operating abnormally by a Raspberry Pi processing unit. A specially built user interface allows the activity of the Raspberry Pi to be tracked on a Tablet for observation purposes. The proposed system provides the ability to land a drone when the motor temperature exceeds an automatically generated threshold. The experimental results confirm that the proposed system can safely control the drone using information obtained from temperature sensors attached to the motor.

DOI： 10.1109/JIOT.2017.2737479

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85028968216&origin=inward

Automatic road detection system for an airland amphibious car 査読有り

Li, Lu, Nakayama, Kim, Serikawa

Future Generation Computer Systems 85 51 - 59 2018年08月

Automatic road detection system for an air–land amphibious car drone 査読有り国際誌

Li Y., Lu H., Nakayama Y., Kim H., Serikawa S.

Future Generation Computer Systems 85 51 - 59 2018年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018 Elsevier B.V. In recent years, unmanned aerial vehicle (UAV) technologies have rapidly developed. Drones, which are one type of UAV, are used in many industrial fields, such as photography, delivery and agriculture. However, a commercial drone can fly for only approximately 20 min on one charge. Furthermore, drones are prohibited from flying in some areas, and cannot be operated in bad weather. Due to the development of drone technologies, we must reduce energy consumption and achieve long-range movement. To overcome these limitations, we develop a new air–land amphibious car drone that can fly and requires less power consumption in land mode; this extends the range of mobility of the drone. Moreover, land mode can be used to pass through restricted areas or bad weather conditions by sliding. Furthermore, we develop a Convolutional Neural Network (CNN)-based algorithm for detecting the road in a captured scene. To more accurately segment the road region based on images from the equipped camera of the drone, we propose atrous spatial pyramid pooling (ASPP) ResNet blocks, instead of Resblocks, which were proposed by DeepLab. The experimental results demonstrate that the proposed method improves the pixel accuracy (PA) to 85.6% and achieves a mean Intersection over Union (mIoU) of 55.8%.

DOI： 10.1016/j.future.2018.02.036

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85043254725&origin=inward

　前のページ - 次のページ　

<神谷 亨>

論文 - 神谷 亨

<神谷　亨>

論文 - 神谷　亨