Nobutaka Kuroki
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Nobutaka Kuroki has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
Masahiro Numa
Graduate School of Engineering Kobe University 1‐1 Rokkodai, Nada Kobe Hyogo 657‐8501 Japan
8 shared publicationsTetsuya Hirose
Graduate School of Engineering Osaka University 2‐1 Yamadaoka Suita Osaka 565‐0871 Japan
5 shared publicationsKazuya Urazoe
Graduate School of Engineering Kobe University 1‐1 Rokkodai‐cho, Nada‐ku, Kobe Hyogo 657‐8501 Japan
4 shared publicationsShinya Ohtani
Graduate School of Engineering Kobe University 1‐1 Rokkodai, Nada Kobe Hyogo 657‐8501 Japan
3 shared publicationsKunihiko Taya
Forensic Science Lab Kyoto Prefectural Police H.Q. Kamigyo‐ku Kyoto Japan
2 shared publications- Tetsuya Hirose
Osaka Univ., Suita, Japan
2 shared publications
Latest Publications
- 25 April 2024
- Published by Wiley in IEEJ Transactions on Electrical and Electronic Engineering
- Vol. 19 (8) , 1360-1368
SRFLOW‐Attention: Super Resolution with Multi‐Head Attention for Effective Use of Low‐Resolution Information
This paper proposes normalizing flow‐based image super‐resolution techniques using attention modules. In the proposed method, the features of the low‐resolution images are extracted using a Swin Transformer. Furthermore, multi‐head attention in the flow layers makes effective use of the feature maps. This architecture enables the efficient injection of low‐resolution image features extracted by the transformer into the flow layer. Experimental results at x4 magnifications showed that the proposed method achieved state‐of‐the‐art performance for quantitative metrics and visual quality among single‐loss architectures. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.
- 25 April 2024
- Published by Wiley in IEEJ Transactions on Electrical and Electronic Engineering
- Vol. 19 (8) , 1360-1368
SRFLOW‐Attention: Super Resolution with Multi‐Head Attention for Effective Use of Low‐Resolution Information
This paper proposes normalizing flow‐based image super‐resolution techniques using attention modules. In the proposed method, the features of the low‐resolution images are extracted using a Swin Transformer. Furthermore, multi‐head attention in the flow layers makes effective use of the feature maps. This architecture enables the efficient injection of low‐resolution image features extracted by the transformer into the flow layer. Experimental results at x4 magnifications showed that the proposed method achieved state‐of‐the‐art performance for quantitative metrics and visual quality among single‐loss architectures. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.
- 27 July 2021
- Published by Wiley in IEEJ Transactions on Electrical and Electronic Engineering
- Vol. 16 (11) , 1510-1517
Automated Fish Bone Detection in X‐Ray Images with Convolutional Neural Network and Synthetic Image Generation
This paper proposes a new fish bone detection technique using a convolutional neural network (CNN) and synthetic image generation. Semantic segmentation CNNs with supervised learning generally require a large number of training images and their pixel‐wise teaching signals. In fish bone detection, there are two problems with using semantic segmentation CNNs. One is the manual annotations of fish bones and the other is the difficulty of sampling all variations of fish bones with various lengths, angles, and thicknesses. The proposed method, however, generates them by drawing virtual fish bones on X‐ray images. This technique is very useful for reducing the cost of collecting and annotating a dataset. Experimental results have shown that the average F‐measure for the proposed method is 0.747, while that for a normal training method is 0.493. In the proposed method, the CNN successfully detected actual fish bones despite its training only with virtual fish bones. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.
- 27 July 2021
- Published by Wiley in IEEJ Transactions on Electrical and Electronic Engineering
- Vol. 16 (11) , 1510-1517
Automated Fish Bone Detection in X‐Ray Images with Convolutional Neural Network and Synthetic Image Generation
This paper proposes a new fish bone detection technique using a convolutional neural network (CNN) and synthetic image generation. Semantic segmentation CNNs with supervised learning generally require a large number of training images and their pixel‐wise teaching signals. In fish bone detection, there are two problems with using semantic segmentation CNNs. One is the manual annotations of fish bones and the other is the difficulty of sampling all variations of fish bones with various lengths, angles, and thicknesses. The proposed method, however, generates them by drawing virtual fish bones on X‐ray images. This technique is very useful for reducing the cost of collecting and annotating a dataset. Experimental results have shown that the average F‐measure for the proposed method is 0.747, while that for a normal training method is 0.493. In the proposed method, the CNN successfully detected actual fish bones despite its training only with virtual fish bones. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.
- 1 January 2021
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Information and Systems
- Vol. E104.D (1) , 183-193
Multi-Category Image Super-Resolution with Convolutional Neural Network and Multi-Task Learning
Japan's largest platform for academic e-journals: J-STAGE is a full text database for reviewed academic papers published by Japanese societies
- 1 January 2021
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Information and Systems
- Vol. E104.D (1) , 183-193
Multi-Category Image Super-Resolution with Convolutional Neural Network and Multi-Task Learning
Japan's largest platform for academic e-journals: J-STAGE is a full text database for reviewed academic papers published by Japanese societies
- 28 October 2020
- Published by Wiley in IEEJ Transactions on Electrical and Electronic Engineering
- Vol. 16 (1) , 161-163
Combination of Convolutional Neural Network Architecture and its Learning Method for Rotation‐Invariant Handwritten Digit Recognition
This letter presents several combinations of a convolutional neural network (CNN) and its learning method for rotation‐invariant digit recognition. Rotation data augmentation is widely used for improving rotation invariance. Data augmentation commonly assigns the same label to all augmented images of the same source. However, this learning method causes some collisions between original and rotated digits. Thus, this letter presents three types of rotation‐invariance learning methods and applies them to five popular CNN architectures. Experimental results indicate that multi‐task learning on ResNet‐50 is the best combination. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.
- 28 October 2020
- Published by Wiley in IEEJ Transactions on Electrical and Electronic Engineering
- Vol. 16 (1) , 161-163
Combination of Convolutional Neural Network Architecture and its Learning Method for Rotation‐Invariant Handwritten Digit Recognition
This letter presents several combinations of a convolutional neural network (CNN) and its learning method for rotation‐invariant digit recognition. Rotation data augmentation is widely used for improving rotation invariance. Data augmentation commonly assigns the same label to all augmented images of the same source. However, this learning method causes some collisions between original and rotated digits. Thus, this letter presents three types of rotation‐invariance learning methods and applies them to five popular CNN architectures. Experimental results indicate that multi‐task learning on ResNet‐50 is the best combination. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.
- 30 August 2020
- Published by Wiley in Electronics and Communications in Japan
- Vol. 103 (10) , 15-25
Detecting tampered region in video using LSTM and U‐Net
In the world today, we can always shoot digital videos with smartphones and share them with family or friends after editing them. It is necessary to check whether these videos have been tampered with to use them as forensic evidence in criminal investigations. This paper proposes a method for detecting tampered regions in a video where a specific subject is removed from a scene. In the proposed method, we use LSTM for the time domain analysis and UNet for the space domain analysis. The detection accuracy achieved 0.98 in terms of F‐measure, even though tampered the region was deformed or moved in the video. The experimental results show superior performance in the detection of tampered regions in digital videos.
- 30 August 2020
- Published by Wiley in Electronics and Communications in Japan
- Vol. 103 (10) , 15-25
Detecting tampered region in video using LSTM and U‐Net
In the world today, we can always shoot digital videos with smartphones and share them with family or friends after editing them. It is necessary to check whether these videos have been tampered with to use them as forensic evidence in criminal investigations. This paper proposes a method for detecting tampered regions in a video where a specific subject is removed from a scene. In the proposed method, we use LSTM for the time domain analysis and UNet for the space domain analysis. The detection accuracy achieved 0.98 in terms of F‐measure, even though tampered the region was deformed or moved in the video. The experimental results show superior performance in the detection of tampered regions in digital videos.
- 1 July 2020
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
- Vol. E103.A (7) , 955-958
Improvement of Luminance Isotropy for Convolutional Neural Networks-Based Image Super-Resolution
Convolutional neural network (CNN)-based image super-resolutions are widely used as a high-quality image-enhancement technique. However, in general, they show little to no luminance isotropy. Thus, we propose two methods, “Luminance Inversion Training (LIT)” and “Luminance Inversion Averaging (LIA),” to improve the luminance isotropy of CNN-based image super-resolutions. Experimental results of 2× image magnification show that the average peak signal-to-noise ratio (PSNR) using Luminance Inversion Averaging is about 0.15-0.20dB higher than that for the conventional super-resolution.
- 1 July 2020
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
- Vol. E103.A (7) , 955-958
Improvement of Luminance Isotropy for Convolutional Neural Networks-Based Image Super-Resolution
Convolutional neural network (CNN)-based image super-resolutions are widely used as a high-quality image-enhancement technique. However, in general, they show little to no luminance isotropy. Thus, we propose two methods, “Luminance Inversion Training (LIT)” and “Luminance Inversion Averaging (LIA),” to improve the luminance isotropy of CNN-based image super-resolutions. Experimental results of 2× image magnification show that the average peak signal-to-noise ratio (PSNR) using Luminance Inversion Averaging is about 0.15-0.20dB higher than that for the conventional super-resolution.
- 1 April 2020
- Published by Institute of Electrical Engineers of Japan (IEE Japan) in IEEJ Transactions on Electronics, Information and Systems
- Vol. 140 (4) , 476-483
Detecting Tampered Region in video using LSTM and U-Net
Recently, we can always shoot digital videos with smartphone and share them with family or friends after editing them. It is necessary to check whether they are tampered by someone or not, if we use these digital videos as forensic evidence in criminal investigation. This paper proposes a method for detecting tampered regions in a video where a specific subject is removed from a scene. In the proposed method, we use LSTM for the time domain analysis and U-Net for the space domain analysis. The detection accuracy achieved 0.98 in terms of F-measure, even though tampered region was deformed or moved in the video. The experimental results show superior performance to detect tampered regions for digital videos.
- 1 April 2020
- Published by Institute of Electrical Engineers of Japan (IEE Japan) in IEEJ Transactions on Electronics, Information and Systems
- Vol. 140 (4) , 476-483
Detecting Tampered Region in video using LSTM and U-Net
Recently, we can always shoot digital videos with smartphone and share them with family or friends after editing them. It is necessary to check whether they are tampered by someone or not, if we use these digital videos as forensic evidence in criminal investigation. This paper proposes a method for detecting tampered regions in a video where a specific subject is removed from a scene. In the proposed method, we use LSTM for the time domain analysis and U-Net for the space domain analysis. The detection accuracy achieved 0.98 in terms of F-measure, even though tampered region was deformed or moved in the video. The experimental results show superior performance to detect tampered regions for digital videos.
- 1 November 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 225-228
A 0.38-μW stand-by power, 50-nA-to-1-mA load current range DC-DC converter with self-biased linear regulator for ultra-low power battery management
This paper proposes a DC-DC converter for ultra-low power battery management. The proposed circuit consists of a switched capacitor converter (SCC) and self-biased linear regulator (SBLR). The SBLR monitors an output load current and controls the switching frequency of the SCC to achieve both ultra-low stand-by power and wide load current capability. Measurement results demonstrated that the converter successfully converts a 4.2-V input into a 0.98-V output. The stand-by power was extremely low, 0.38 μW. The load current range where the power conversion efficiency was higher than 50% was from 2 μA to 1 mA. The efficiency was 54% when the input voltage and output power were 4.0 V and 0.97 mW, respectively.
- 1 November 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 225-228
A 0.38-μW stand-by power, 50-nA-to-1-mA load current range DC-DC converter with self-biased linear regulator for ultra-low power battery management
This paper proposes a DC-DC converter for ultra-low power battery management. The proposed circuit consists of a switched capacitor converter (SCC) and self-biased linear regulator (SBLR). The SBLR monitors an output load current and controls the switching frequency of the SCC to achieve both ultra-low stand-by power and wide load current capability. Measurement results demonstrated that the converter successfully converts a 4.2-V input into a 0.98-V output. The stand-by power was extremely low, 0.38 μW. The load current range where the power conversion efficiency was higher than 50% was from 2 μA to 1 mA. The efficiency was 54% when the input voltage and output power were 4.0 V and 0.97 mW, respectively.