KYKota Yoshida

Kota Yoshida

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kota Yoshida has a total of 27 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Compression-Efficient Feature Extraction Method for a CMOS Image Sensor

    by Keiichiro Kuroda, Yu Osuka, Ryoya Iegaki, Ryuichi Ujiie, Hideki Shima, Kota Yoshida, Shunsuke Okura

    To address the power constraints of the emerging Internet of Things (IoT) era, we propose a compression-efficient feature extraction method for a CMOS image sensor that can extract binary feature data. This sensor outputs six-channel binary feature data, comprising three channels of binarized luminance signals and three channels of horizontal edge signals, compressed via a run length encoding (RLE) method. This approach significantly reduces data transmission volume while maintaining image recognition accuracy. The simulation results obtained using a YOLOv7-based model designed for edge GPUs demonstrate that our approach achieves a large object recognition accuracy (APL50) of 60.7% on the COCO dataset while reducing the data size by 99.2% relative to conventional 8-bit RGB color images. Furthermore, the image classification results using MobileNetV3 tailored for mobile devices on the Visual Wake Words (VWW) dataset show that our approach reduces data size by 99.0% relative to conventional 8-bit RGB color images and achieves an image classification accuracy of 89.4%. These results are superior to the conventional trade-off between recognition accuracy and data size, thereby enabling the realization of low-power image recognition systems.

  • Open Access

    Hash-Based Message Authentication Code with a Reverse Fuzzy Extractor for a CMOS Image Sensor

    by Yuki Rogi, Manami Hagizaki, Tatsuya Oyama, Hiroaki Ogawa, Kota Yoshida, Takeshi Fujino, Shunsuke Okura

    The MIPI (Mobile Industry Processor Interface) Alliance provides a security framework for in-vehicle network connections between sensors and processing electronic control units (ECUs). One approach within this framework is data integrity verification for sensors with limited hardware resources. In this paper, the security risks associated with image sensor data are described. Adversarial examples (AEs) targeting the MIPI interface can induce misclassification, making image data integrity verification essential. A CMOS image sensor with a message authentication code (CIS-MAC) is then proposed as a defense mechanism starting from the image sensor to protect image data from malicious manipulations, such as AE attacks. Evaluation results of the physically unclonable function (PUF) response and random number, which are utilized for generating the cryptographic key and MAC tag, are presented using a 2 Mpixel CMOS image sensor. The area of the CIS-MAC circuit is estimated based on a Verilog HDL design and synthesis using a 0.18 μm CMOS process. Various hash topologies are evaluated to select a hash function suitable for key generation and MAC tag generation within the CMOS image sensor. The estimated area of the CIS-MAC circuit is 67 kGE and 0.86mm2, demonstrating feasibility for implementation in a CMOS image sensor typically fabricated using analog process technology.


  • Open Access

    Realtime Multispectral Pedestrian Detection With Visible and Far-Infrared Under Ambient Temperature Changing

    by Masato Okuda, Kota Yoshida, Takeshi Fujino

    In recent intelligent transportation systems (ITS), it is important to recognize pedestrians and avoid collisions. Various sensors are used to detect pedestrians, and some research on pedestrian detection uses a visible light (RGB) camera and a far-infrared (FIR) camera. FIR cameras are significantly affected by ambient temperatures such as summer and winter. However, few studies have focused on this property when evaluating pedestrian detection accuracy. Therefore, this paper investigates the effect of temperature change in real-time multispectral pedestrian detection. We created an original dataset with three subsets, Hot, Intermediate, and Cold, and evaluated temperature effects by changing the subsets during training and testing. We first evaluated YOLOv8s-4ch, which simply extended the input layer of YOLOv8 from 3 channels of RGB to 4 channels of RGB-FIR. To further improve detection performance, we built a new model called YOLOv8s-2stream. This model has two backbones for RGB and FIR, and fuses their feature maps in each resolution. We found that the model trained on a specific temperature subset dropped the test accuracy in other subsets. On the other hand, when training using a Mix set covering all temperature sets (Hot, Inter., Cold), the model achieved the highest accuracy through all conditions. Moreover, our YOLOv8s-2stream has improved by 3.9 points of accuracy (AP@0.5:0.95) compared to YOLOv8s-4ch, and achieved 73FPS inference speed on Jetson.


  • Open Access

    Image Classification Using Neural Network for Feature-Extractable CMOS Image Sensor

    by Yu Osuka, Kota Yoshida, Shunsuke Okura

    For the forthcoming Internet of Things (IoT) era, it will be important to reduce the data output from sensors as well as their energy efficiency. Since conventional image sensor output data for photography are often redundant in AI applications, we propose a CMOS image sensor that can generate both RGB color images for humans and feature data for deep learning (DL). Use of feature data allows reducing the energy efficiency of the image classification system and saving storage space for imaging data. We performed experiments to demonstrate that the simulated feature data are suitable for use in image classification tasks. A five-layer convolutional neural network (CNN) classifier was trained and tested using the aggressively quantized feature data generated from a person dataset, where image classification accuracy was also improved when applying contrast enhancement. According to the experimental results, an accuracy of 95.9% was achieved using 1-bit feature data, resulting in 93.75% reduction in the amount of data compared to RGB color images.


  • Open Access

    On-Chip Data Reduction and Object Detection for a Feature-Extractable CMOS Image Sensor

    by Yudai Morikaku, Ryuichi Ujiie, Daisuke Morikawa, Hideki Shima, Kota Yoshida, Shunsuke Okura

    In order to improve image recognition technologies in an IoT environment, we propose a data reduction scheme for a feature-extractable CMOS image sensor and present simulation results for object recognition using feature data. We evaluated the accuracy of the simulated feature data in object recognition based on YOLOX trained with a feature dataset. According to our simulation results, the obtained object recognition accuracy was 56.6% with the large-scale COCO dataset, even though the amount of data was reduced by 97.7% compared to conventional RGB color images. When the dataset was replaced with the RAISE RAW image dataset for more accurate simulation, the object recognition accuracy improved to 76.3%. Furthermore, the feature-extractable CMOS image sensor can switch its operation mode between RGB color image mode and feature data mode. When the trigger for switching from feature data mode to RGB color image mode was set to the detection of a large-sized person, the feature data achieved an accuracy of 93.5% with the COCO dataset.


  • Open Access

    Data Poisoning Attack against Neural Network-Based On-Device Learning Anomaly Detector by Physical Attacks on Sensors

    by Takahito Ino, Kota Yoshida, Hiroki Matsutani, Takeshi Fujino

    In this paper, we introduce a security approach for on-device learning Edge AIs designed to detect abnormal conditions in factory machines. Since Edge AIs are easily accessible by an attacker physically, there are security risks due to physical attacks. In particular, there is a concern that the attacker may tamper with the training data of the on-device learning Edge AIs to degrade the task accuracy. Few risk assessments have been reported. It is important to understand these security risks before considering countermeasures. In this paper, we demonstrate a data poisoning attack against an on-device learning Edge AI. Our attack target is an on-device learning anomaly detection system. The system adopts MEMS accelerometers to measure the vibration of factory machines and detect anomalies. The anomaly detector also adopts a concept drift detection algorithm and multiple models to accommodate multiple normal patterns. For the attack, we used a method in which measurements are tampered with by exposing the MEMS accelerometer to acoustic waves of a specific frequency. The acceleration data falsified by this method were trained on an anomaly detector, and the result was that the abnormal state could not be detected.