研究者詳細 - 神谷　亨

論文 - 神谷　亨

分割表示751 件中 61 - 80 件目／全件表示 >>

Temporal subtraction technique for phalanges CR images using U-Net and Geometric-matching CNN 査読有り国際誌

Ono, Kamiya, Aoki

Intl. Journal of Computer Assisted Radiology and Surgery 2023 S84 - S85 2023年06月

Multidimensional Deformable Object Manipulation Based on DN-Transporter Networks 査読有り国際誌

Teng, Huimin Lu, Yujie Li, Tohru Kamiya, Yoshihisa Nakatoh, Serikawa Seiichi, Pengxiang Gao

IEEE Transactions on Intelligent Transportation Systems 24 ( 4 ) 4532 - 4540 2023年04月

Research on adaptive formation control of mobile robot based on improved virtual spring method 査読有り国際誌

Chen Y(天津工大)., Li B., Kamiya T., Shi X.

ACM International Conference Proceeding Series 775 - 780 2023年03月

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）

A robot formation control algorithm based on virtual spring model is proposed to solve the problem of path planning and formation keeping in unknown environment. By improving the design of target attraction, obstacle repulsion and interaction force between robots, the formation control is more secure and stable. Furthermore, the leader-follower formation transformation problem is studied. According to the objective function of minimum completion time, minimum total energy consumption and minimum total spring deformation, an adaptive formation transformation strategy is designed. This strategy can choose formation transformation mode adaptively according to the environment characteristics, such as adjust the parameters of the formation and reconfigure the formation, so that the robot formation can complete the task in a short time and energy consumption. Finally, the effectiveness of the proposed method is verified by simulation and real experiment.

DOI： 10.1145/3594315.3594404

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85168241761&origin=inward

Volumetric choroidal segmentation using 3D residual U-Net 査読有り国際誌

Li G(天津工大)., Wang K(天津工大)., Wang X(天津工大)., Sun B(天津工大)., Wang K(天津工大)., Gao Y(天津工大)., Sun S(天津工大)., Kamiya T., Dai Y(天津工大).

ACM International Conference Proceeding Series 145 - 149 2023年03月

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）

Estimating dimensional measurements of the choroid provides diagnostic values which can be used to assess choroidal health. In this paper we describe a methodology of calculating measurements from choroid segmentations automatically generated using convolutional neural network (CNN). We use a three-dimensional (3D) U-Net architecture built from residual units to segment the choroid. A surface fitting phase is jointed to the main process to compensate segmented defects at the area of Optic Nerve Hypoplasia (ONH). Consequently, we process these segmentations to estimate the mean choroidal thickness(MCT). The model is evaluated on volumetric scans from 183 subjects, approximately half of which are thyroid eyes. In the choroidal layer segmentation experiment, the accuracy of the automatic segmentation algorithm proposed in this paper was 98.25% when comparing the manual segmentation results masked by doctors. It showed that the MCT in thyroid eyes were higher than those in normal eyes.

DOI： 10.1145/3594315.3594337

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85168249566&origin=inward

Research on Semantic segmentation co-evolutionary method for Automated Driving Haze Images 査読有り国際誌

Tang C（天津工大）., Wang Y., Kamiya T., Li Y.

ACM International Conference Proceeding Series 157 - 163 2023年03月

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）

In this paper, we address the problem of semantic segmentation of advanced automatic driving image in hazy weather. We mainly adopt the co-evolution method of image defogging model and semantic segmentation model, the defogging model adopts generative adversarial GAN network, the semantic segmentation model adopts BiSeNet,and the training data is a public self-driving data set Cityscapes. The co-evolution of these two models is used to improve the accuracy of semantic segmentation of fog maps. The experimental results show that the semantic segmentation of fog map by co-evolution has a more obvious improvement effect, and to a certain extent, it can meet the target extraction and recognition of autonomous driving requirements.

DOI： 10.1145/3594315.3594339

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85168236599&origin=inward

Automatic Classification Method for Plastic Bottles and Caps Using Multi Attention Eff-UNet 査読有り国際誌

Moritsuka, Kamiya

International Conference on Artificial Lefe and Robotics 980 - 93 2023年02月

CNNを用いた磁粉探傷画像の分類法査読有り

森塚、神谷

バイオメディカル・ファジィ・システム学会誌 24 ( 2 ) 35 - 42 2023年02月

Atomatic Identification of Tumor Cells for Circulating Tumor Cells by Convolutional Neural Netwokrs 査読有り国際誌

Hashimoto, Kamiya, Li（天津工大）, Yoneda, Tanaka

International Journal of Innovative Computing, Information and Control 19 ( 1 ) 1 - 14 2023年02月

AI技術の医用画像解析への応用招待有り査読有り

神谷

計測と制御 62 ( 2 ) 95 - 97 2023年02月

AUTOMATIC IDENTIFICATION OF TUMOR CELLS FOR CIRCULATING TUMOR CELLS BY CONVOLUTIONAL NEURAL NETWORKS 査読有り国際誌

Hashimoto K., Kamiya T., Li G（天津工大）., Yoneda K., Tanaka F.

ICIC Express Letters 19 ( 1 ) 1 - 14 2023年02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）

Liquid biopsy allows non-invasive collection of circulating tumor cells (CTCs) in blood without the need for sampling and can clearly demonstrate their presence in many types of cancer. In this study, we propose a method to automatically identify CTCs from fluorescence microscopy images and enable quantitative analysis based on convolutional neural networks (CNN). In this paper, a cell nucleus region cropping algorithm is applied in addition to a filtering process centered on a selective enhancement filter. Next, identification by SqueezeNet is performed. We performed the proposed method to 5,040 images of 6 samples and conducted experiments to identify CTCs. The number of detected CTCs was 148 (TPR = 100%), and the number of over-detected non-CTCs was 925. For the identification, TPR = 88.51% and FPR = 5.102% for the CNN model using SqueezeNet. The proposed method successfully reduced the number of detections by about 71.4% without missing any correct answers, but the proposed method did not show good results in any of the evaluation metrics.

DOI： 10.24507/ijicic.19.01.1

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85148695370&origin=inward

深層学習による経時的差分像からの結節状陰影の検出

神谷，芳野，寺澤，青木

第１４回呼吸機能イメージング研究会学術集会 70 - 70 2023年01月

Automatic Classification Method for Plastic Bottles and Caps Using Multi Attention Eff-UNet 査読有り国際誌

Moritsuka S., Kamiya T.

Proceedings of International Conference on Artificial Life and Robotics 980 - 983 2023年01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）

In Japan, increasing amounts of waste are becoming a social problem. One of approaches to solve the problem is recycle of the plastic bottles. However, they are thrown away with their caps still attached, and it should be removed by hand. To solve this problem, we developed a method for automatic identification of plastic bottles and caps using deep learning technique. In this paper, we propose a method that combines different numbers of Efficient blocks and adds an attention structure and verify its usefulness through experiments.

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85149463467&origin=inward

Squeeze-and-Excitation Block Based Mask R-CNN for Object Instance Segmentation 査読有り

Nagasawa K., Ishiyama S., Lu H., Kamiya T., Nakatoh Y., Serikawa S., Li Y.

Communications in Computer and Information Science 1732 CCIS 56 - 64 2023年01月

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）

Deep learning-based methods have taken center stage in image recognition, such as AlexNet and deep learning-based method. At present, Image recognition based on deep learning has been widely used in agriculture, factory automation, automated driving, medical fields and so on. In the fields of automated driving and medical care, the accuracy of the image recognition directly affects human lives. For these reasons, the importance of improving the accuracy of image recognition is clear. In this paper, we focus on instance segmentation tasks. The method used is Mask R-CNN, which is the basis of current state-of-the-art methods. The network structure based on ResNet, and we tried to improve the accuracy by adding Squeeze-and-Excitation Block (SE Block). According to the result of experiments, it is proved that this method has certain advantages for object instance segmentation.

DOI： 10.1007/978-981-99-2789-0_5

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85163323022&origin=inward

Pose Estimation of Point Sets Using Residual MLP in Intelligent Transportation Infrastructure 査読有り国際誌

Li Y., Yin Z., Zheng Y., Lu H., Kamiya T., Nakatoh Y., Serikawa S.

IEEE Transactions on Intelligent Transportation Systems 2023年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

6D pose estimation of arbitrary objects is a crucial topic for intelligent transportation infrastructure measurement. However, some external environmental factors and the characteristics of the object itself impact the accuracy of the object’s pose estimation in practical applications. In this paper, we propose a new multi-class dataset ICD-4 (Industrial car Components Dataset) for 6D object pose estimation, which mainly includes four component categories, and every category takes 20,000 different scenarios. ICD-4 dataset delivers quite a few research challenges involving the range of object pose transformations and has significant research value for small-scale pose estimation tasks. We also propose an innovative method PoseMLP, a pose estimation network that uses residual MLP (multilayer perceptron) modules to predict the 6D pose estimation directly. Simultaneously, the experimental results demonstrate the effectiveness and reliability of the proposed method.

DOI： 10.1109/TITS.2023.3250604

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85153361050&origin=inward

Remote Sensing Image Registration Based on Improved Geometric-Matching CNN 査読有り国際誌

Morishima F., Lu H., Kamiya T.

International Conference on Control, Automation and Systems 1745 - 1748 2023年01月

担当区分：最終著者,　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）

Environmental change detection is one of the uses of satellite images. This process is performed by subtracting image pairs obtained by different time series or sensors. Therefore, image registration is an important pre-processing step in detection of environmental changes. Currently, image registration methods based on deep learning are gaining attention. In general, higher satellite image resolution results in more accurate registration. However, the increase in image size leads to higher computational costs during training and estimation of deep learning models. Then, we propose a method that reduces the number of parameters of the model to lower the computational cost while maintaining the accuracy. This method makes it easier to handle high-resolution satellite images. The proposed method modified the GMCNN (Geometric-matching CNN) architecture by adding CSA (Cosine Similarity Attention) and SE (Squeeze-and-Excitation) layers to enhance the feature map, and point-wise convolution to reduce the number of parameters. The improved GMCNN decreases the grid MSE by 0.0037 compared to the conventional GMCNN. It also reduces the number of parameters by 49.6%.

DOI： 10.23919/ICCAS59377.2023.10316818

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179178272&origin=inward

Recognition of Specific Parts of Plastic Bottles Using Improved DeepLab v3+ 査読有り国際誌

Ideta D., Kamiya T.

International Conference on Control, Automation and Systems 1758 - 1761 2023年01月

担当区分：最終著者,　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）

In this paper, we focus on the manpower shortage in factories and conduct an experiment to try to automate the process. Factory automation has many advantages, such as reducing labor costs and improving production efficiency. Among them, we focused on the sorting of plastic bottles at a waste disposal plant. Currently, garbage is sent to the landfill without being sorted, and it is sorted by hand. Unlike cans, plastic bottles cannot be easily automated because they cannot be separated using magnets. Therefore, we propose an image processing technique to recognize where plastic bottles are located, and to use a robot arm to sort them, thereby realizing automation. In this paper, we focus on image processing technique based on deep learning. As basic research, we conducted an experiment to see how well the robot can identify a single plastic bottle in an image. We attempted to use semantic segmentation methods for detection, using DeepLabv3+ as the basic model, and improved it. scSE Block and fine-tuning were introduced to improve the accuracy significantly compared to previous studies.

DOI： 10.23919/ICCAS59377.2023.10316854

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179179143&origin=inward

Identification of Root Resorption on Panoramic X-ray Image Based on EfficientNet 査読有り国際誌

Sakata K., Kamiya T., Oda M., Morimoto Y.

International Conference on Control, Automation and Systems 1749 - 1752 2023年01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）

Root resorption is a disease in which the cells of the root of the tooth are destroyed. This disease may be harmless, depending on the cause. When the root resorption has progressed, early detection and treatment is necessary. If root resorption is left untreated, the lesion can progress to pulp exposure, necrosis, sinus and periodontal defects, and perforation. Diagnosis of root resorption is made by clinical examination and a panoramic x-ray machine, however accurate diagnosis requires computed tomography (CT). CT allows full-plane examination of the area of interest and provides an accurate diagnosis. However, CT is less widely used than panoramic radiography in private dental office. Therefore, there is a need for a method to detect root resorption from panoramic radiographs with a high reliability comparable to that of CT images. In this paper, we propose an image analysis method to identify the presence or absence of root resorption from panoramic radiographs. In particular, we propose a method that uses EfficientNet to extract features with fine-tuning and learns them using L2-softmax loss as an error function. We applied the proposed method to panoramic X-ray images of 59 actual cases and obtained TPR: 71% and FPR: 43% respectively.

DOI： 10.23919/ICCAS59377.2023.10316940

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179183951&origin=inward

Identification of Nodular Shadows from CT Images Using Improved CoAtNet Incorporated Clinical Recording 査読有り国際誌

Nishitaki Y., Kamiya T., Kido S.

International Conference on Control, Automation and Systems 1727 - 1732 2023年01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）

When lung cancer is suspected, a CT scan of the chest is widely used as a method of precise examination. However, the number of CT images obtained in a single examination is enormous, placing a heavy burden on the physician who reads the images. In addition, there is concern about the possibility of undetected lesions due to differences in the skills and experience of the reader. Therefore, computer-aided diagnosis systems have been introduced to reduce their workload and undetected lesions. When physicians make a diagnosis, they consider not only CT images but also information about the patient. Therefore, attempts are being made to improve the accuracy of diagnosis by mimicking this process using artificial intelligence. In this paper, we propose a model for identifying nodular shadows using deep learning, aiming to improve the accuracy of diagnosis by introducing medical record information in addition to image information. Normal tissue is divided into three classes: branched vessels, thin vessels, and round vessels, and a total of four classes are classified, including abnormal tissue. Experimental results show that the accuracy of nodular shade discrimination is improved when medical record information is added.

DOI： 10.23919/ICCAS59377.2023.10316787

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179182295&origin=inward

Facial Symmetry Analysis for Cleft Lip Patients from 4D Point Cloud Data 査読有り国際誌

Kihara N., Kimura-Nomoto N., Okawachi T., Li G（天津工大）., Nakamura N., Kamiya T.

International Conference on Control, Automation and Systems 1502 - 1505 2023年01月

担当区分：最終著者,　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）

Cleft lip is one of the most common birth defects. Several operations are performed to form a natural lip. A problem with these operations is that the criteria for the facial symmetry are unclear. Based on this background, we propose a method for evaluating the facial symmetry using 4D point cloud data. We evaluate the facial symmetry in two ways. One is based on the temporal changes in the face landmarks. Corresponding points are searched using two local geometric descriptors. The results are approximated to the 3D lines and these slopes are compared to obtain the left-right difference of the movement. The other is based on the center of gravity. The lip part is extracted from the point cloud and divided into subregions. The centroid coordinates are then calculated for each subregion and compared to obtain the left-right difference of the facial structure. The experiment is performed using artificially generated 4D data. As an experimental result, it is shown that our method can find point correspondences with smaller error than comparative methods.

DOI： 10.23919/ICCAS59377.2023.10316967

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179178953&origin=inward

Extraction of Lung Tumor Regions from Thoracic CT Images Using An Improved U-Net 査読有り国際誌

Takahashi R., Kamiya T., Terasawa T., Aoki T.

International Conference on Control, Automation and Systems 1489 - 1493 2023年01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）

Lung cancer currently kills the largest number of people in the world, with lung cancer accounting for the highest number of deaths among men and women. Therefore, early detection and treatment are important issues to reduce the number of lung cancer deaths. Genetic testing can confirm the presence or absence of driver gene mutations involved in cancer cell growth and other factors. In the process of identifying the presence or absence of driver gene mutations, lung cancer regions are extracted manually in cooperation with a radiologist. One of the concerns is that manual lung cancer extraction places a heavy burden on the physician who reads the images. Therefore, we propose an automatic extraction of lung tumor regions from chest CT images using deep learning, with the goal of developing a CAD system that reduces the workload of the physician and prevents lesions from being overlooked. We constructed a new model based on U-Net with the addition of CBAM and MultiRes Block. Experimental results on the model using CT images of the chest showed a 4.2% improvement in accuracy for Dice and a 4.6% improvement for IoU compared to the original U-Net model.

DOI： 10.23919/ICCAS59377.2023.10316783

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179183296&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