• Journal of the Society of Disaster Information
• /
• v.18 no.2
• /
• pp.364-373
• /
• 2022

Purpose: The purpose of this paper is to develope smoke detection using AI model for detecting the initial fire in underground utility tunnels using CCTV Method: To improve detection performance of smoke which is high irregular, a deep learning model for fire detection was trained to optimize smoke detection. Also, several approaches such as dataset cleansing and gradient exploding release were applied to enhance model, and compared with results of those. Result: Results show the proposed approaches can improve the model performance, and the final model has good prediction capability according to several indexes such as mAP. However, the final model has low false negative but high false positive capacities. Conclusion: The present model can apply to smoke detection in underground utility tunnel, fixing the defect by linking between the model and the utility tunnel control system.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.37-42
• /
• 2022

In this paper, AI methods such as deep learning are applied to provide customized exercise therapy for patients with kidney disease. In order to apply deep learning, a dataset that can determine kidney disease is trained to determine whether it is a kidney disease, and 1RM, which is the user's physical information and muscle strength according to whether it is a disease, can also be calculated through deep learning. The calculated muscle strength of 1RM was converted into resistant exercise for each part through a calculation equation for each part of the body, and was configured to be provided with an aerobic exercise amount tailored to the user's body information. If continuous research is conducted in the manner proposed in this paper, customized exercise therapy can be provided for various diseases.

• Journal of IKEEE
• /
• v.26 no.1
• /
• pp.104-110
• /
• 2022

In this paper, we propose a new dataset for AI diagnosis and BLDC motor diagnosis in UAV. In the diagnosis of BLDC motor, PVA(Park's Vector Approach) is difficult to apply due to many ripples of frequency components. However, since the components of ripples are the third harmonics, we propose a method to utilize PVA as circle fitting by applying Savitzky-Golay filter which is excellent for the third harmonics. On the other hand, PVA, a technique to convert from three-phase to two-phase, is always based on the origin during the transformation process. This study demonstrates that the error of the origin and the measured center can be detected and diagnosed in the application process of Circle fitting, and that it can be used as a new data set of AI technology.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.282-284
• /
• 2021

최근 4차 산업혁명으로 의료빅데이터 기반으로 한 AI 기술이 급속도로 발전하고 있다. 특히, 의료영상을 기반으로 병변을 탐색, 분활 및 정량화 그리고 자동진단 및 예측 관련된 기술이 AI 제품으로 출시되고 있다. AI 기술개발은 많은 학습데이터가 요구되며, 임상검증에 단일기관에서 2개 이상 기관의 검증이 요구되고 있다. 그러나 아직까지도 단일기관에서 학습용 데이터와 테스트, 검증용 데이터를 달리하여 기술개발에 활용하고 있다. 본 논문은 AI 기술개발에 필요한 영상데이터에 대한 표준화된 데이터셋 변환 및 관리를 위한 시스템에 대해 기술한다. 다기관 데이터를 수집하기 위해서는 각 기관의 의료영상 데이터 수집 및 저장하는 기준이 명확하지 않아 표준화 작업이 필요하다. 제안한 시스템은 기관 또는 다기관 연구 그룹의 의료영상데이터를 표준화하여 저장할 수 있을 뿐만 아니라 의료영상 뷰어 및 의료영상 리스트를 통해 연구자가 원하는 의료영상 데이터 셋을 검색하여 다양한 데이터셋으로 제공할 수 있기 때문에 수집 및 변환 그리고 관리까지 지원할 수 있는 시스템으로 영상기반의 머신러닝 연구에 활력을 불어넣을 수 있을 것으로 기대하고 있다.

• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• pp.115-122
• /
• 2024

Multi-object tracking (MOT) is a vital component in understanding the surrounding environments. Previous research has demonstrated that MOT can successfully detect and track surrounding objects. Nonetheless, inaccurate classification of the tracking objects remains a challenge that needs to be solved. When an object approaching from a distance is recognized, not only detection and tracking but also classification to determine the level of risk must be performed. However, considering the erroneous classification results obtained from the detection as the track class can lead to performance degradation problems. In this paper, we discuss the limitations of classification in tracking under the classification uncertainty of the detector. To address this problem, a class update module is proposed, which leverages the class uncertainty estimation of the detector to mitigate the classification error of the tracker. We evaluated our approach on the VisDrone-MOT2021 dataset,which includes multi-class and uncertain far-distance object tracking. We show that our method has low certainty at a distant object, and quickly classifies the class as the object approaches and the level of certainty increases.In this manner, our method outperforms previous approaches across different detectors. In particular, the You Only Look Once (YOLO)v8 detector shows a notable enhancement of 4.33 multi-object tracking accuracy (MOTA) in comparison to the previous state-of-the-art method. This intuitive insight improves MOT to track approaching objects from a distance and quickly classify them.

• KIPS Transactions on Software and Data Engineering
• /
• v.13 no.1
• /
• pp.17-27
• /
• 2024

Video frame interpolation is an important technique used in the field of video and media, as it increases the continuity of motion and enables smooth playback of videos. In the study of video frame interpolation using deep learning, Kernel Based Method captures local changes well, but has limitations in handling global changes. In this paper, we propose a new U-Net structure that applies feature map differentiation and two directions to focus on capturing major changes to generate intermediate frames more accurately while reducing the number of parameters. Experimental results show that the proposed structure outperforms the existing model by up to 0.3 in PSNR with about 61% fewer parameters on common datasets such as Vimeo, Middle-burry, and a new YouTube dataset. Code is available at https://github.com/Go-MinSeong/SF-AdaCoF.

• Dementia and Neurocognitive Disorders
• /
• v.23 no.1
• /
• pp.1-10
• /
• 2024

Background and Purpose: Voice, reflecting cerebral functions, holds potential for analyzing and understanding brain function, especially in the context of cognitive impairment (CI) and Alzheimer's disease (AD). This study used voice data to distinguish between normal cognition and CI or Alzheimer's disease dementia (ADD). Methods: This study enrolled 3 groups of subjects: 1) 52 subjects with subjective cognitive decline; 2) 110 subjects with mild CI; and 3) 59 subjects with ADD. Voice features were extracted using Mel-frequency cepstral coefficients and Chroma. Results: A deep neural network (DNN) model showed promising performance, with an accuracy of roughly 81% in 10 trials in predicting ADD, which increased to an average value of about 82.0%±1.6% when evaluated against unseen test dataset. Conclusions: Although results did not demonstrate the level of accuracy necessary for a definitive clinical tool, they provided a compelling proof-of-concept for the potential use of voice data in cognitive status assessment. DNN algorithms using voice offer a promising approach to early detection of AD. They could improve the accuracy and accessibility of diagnosis, ultimately leading to better outcomes for patients.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.4
• /
• pp.57-64
• /
• 2024

The introduction of Smart Factories has transformed manufacturing towards more objective and efficient line management. However, most companies are not effectively utilizing the vast amount of sensor data collected every second. This study aims to use this data to predict product quality and manage production processes efficiently. Due to security issues, specific sensor data could not be verified, so semiconductor process-related training data from the "SAMSUNG SDS Brightics AI" site was used. Data preprocessing, including removing missing values, outliers, scaling, and feature elimination, was crucial for optimal sensor data. Oversampling was used to balance the imbalanced training dataset. The SVM (rbf) model achieved high performance (Accuracy: 97.07%, GM: 96.61%), surpassing the MLP model implemented by "SAMSUNG SDS Brightics AI". This research can be applied to various topics, such as predicting component lifecycles and process conditions.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.5
• /
• pp.476-486
• /
• 2022

Synthetic dynamic infrared image generation from the given virtual environment is being the primary goal to simulate the output of the infra-red(IR) camera installed on a vehicle to evaluate the control algorithm for various search & reconnaissance missions. Due to the difficulty to obtain actual IR data in complex environments, Artificial intelligence(AI) has been used recently in the field of image data generation. In this paper, CycleGAN technique is applied to obtain a more realistic synthetic IR image. We added the Structural Similarity Index Measure(SSIM) loss function to the L1 loss function to generate a more realistic synthetic IR image when the CycleGAN image is generated. From the simulation, it is applicable to the guided-missile flight simulation tests by using the synthetic infrared image generated by the proposed technique.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.9
• /
• pp.1-7
• /
• 2024

In this paper, we proposes a Convolutional Neural Networks(CNN) equipped with Batch Normalization(BN) for handwritten digit recognition training the MNIST dataset. Aiming to surpass the performance of LeNet-5 by LeCun et al., a 6-layer neural network was designed. The proposed model processes 28×28 pixel images through convolution, Max Pooling, and Fully connected layers, with the batch normalization to improve learning stability and performance. The experiment utilized 60,000 training images and 10,000 test images, applying the Momentum optimization algorithm. The model configuration used 30 filters with a 5×5 filter size, padding 0, stride 1, and ReLU as activation function. The training process was set with a mini-batch size of 100, 20 epochs in total, and a learning rate of 0.1. As a result, the proposed model achieved a test accuracy of 99.22%, surpassing LeNet-5's 99.05%, and recorded an F1-score of 0.9919, demonstrating the model's performance. Moreover, the 6-layer model proposed in this paper emphasizes model efficiency with a simpler structure compared to LeCun et al.'s LeNet-5 (7-layer model) and the model proposed by Ji, Chun and Kim (10-layer model). The results of this study show potential for application in real industrial applications such as AI vision inspection systems. It is expected to be effectively applied in smart factories, particularly in determining the defective status of parts.