• Journal of Internet Computing and Services
• /
• v.7 no.1
• /
• pp.65-74
• /
• 2006

Human tracking is a requirement for the advanced human-computer interface (HCI), This paper proposes a method which uses a component-based human model, detects body parts, estimates human postures, and animates an avatar, Each body part consists of color, connection, and location information and it matches to a corresponding component of the human model. For human tracking, the 2D information of human posture is used for body tracking by computing similarities between frames, The depth information is decided by a relative location between components and is transferred to a moving direction to build a 2-1/2D human model. While each body part is modelled by posture and directions, the corresponding component of a 3D avatar is rotated in 3D using the information transferred from the human model. We achieved 90% tracking rate of a test video containing a variety of postures and the rate increased as the proposed system processed more frames.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.8A
• /
• pp.786-795
• /
• 2002

In this paper, we present a histogram and moment-based vidoe scencd change detection technique using hierarchical Hidden Markov Models(HMMs). The proposed method extracts histograms from a low-frequency subband and moments of edge components from high-frequency subbands of wavelet transformed images. Then each HMM is trained by using histogram difference and directional moment difference, respectively, extracted from manually labeled video. The video segmentation process consists of two steps. A histogram-based HMM is first used to segment the input video sequence into three categories: shot, cut, gradual scene changes. In the second stage, a moment-based HMM is used to further segment the gradual changes into a fade and a dissolve. The experimental results show that the proposed technique is more effective in partitioning video frames than the previous threshold-based methods.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.32 no.5
• /
• pp.518-525
• /
• 2012

In this study, the feasibility of hyperspectral reflectance imaging technique was investigated for the discrimination of viable and non-viable lettuce seeds. The spectral data of hyperspectral reflectance images with the spectral range between 750 nm and 1000 nm were used to develop PLS-DA model for the classification of viable and non-viable lettuce seeds. The discrimination accuracy of the calibration set was 81.6% and that of the test set was 81.2%. The image analysis method was developed to construct the discriminant images of non-viable seeds with the developed PLS-DA model. The discrimination accuracy obtained from the resultant image were 91%, which showed the feasibility of hyperspectral reflectance imaging technique for the mass discrimination of non-viable lettuce seeds from viable ones.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.29-36
• /
• 2021

In this paper, we classify ECG signal data for mobile devices using deep learning models. To classify abnormal heartbeats with high accuracy, three factors of the deep learning model are selected, and the classification accuracy is compared according to the changes in the conditions of the factors. We apply a CNN model that can self-extract features of ECG data and compare the performance of a total of 48 combinations by combining conditions of the depth of model, optimization method, and activation functions that compose the model. Deriving the combination of conditions with the highest accuracy, we obtained the highest classification accuracy of 97.88% when we applied 19 convolutional layers, an optimization method SGD, and an activation function Mish. In this experiment, we confirmed the suitability of feature extraction and abnormal beat detection of 1-channel ECG signals using CNN.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.5
• /
• pp.161-168
• /
• 2021

Cracks affect the robustness of infrastructures such as buildings, bridge, pavement, and pipelines. This paper presents an automated crack detection system which detect cracks in diverse surfaces. We first constructed the combined crack dataset, consists of multiple crack datasets in diverse domains presented in prior studies. Then, state-of-the-art deep learning models in computer vision tasks including VGG, ResNet, WideResNet, ResNeXt, DenseNet, and EfficientNet, were used to validate the performance of crack detection. We divided the combined dataset into train (80%) and test set (20%) to evaluate the employed models. DenseNet121 showed the highest accuracy at 96.20% with relatively low number of parameters compared to other models. Based on the validation procedures of the advanced deep learning models in crack detection task, we shed light on the cost-effective automated crack detection system which can be applied to different surfaces and structures with low computing resources.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.5
• /
• pp.47-54
• /
• 2022

Recently, with the development of deep learning technology, research on recognizing human behavior is in progress. In this paper, a study was conducted to recognize risky behaviors that may occur in a single-person household environment using deep learning technology. Due to the nature of single-person households, personal privacy protection is necessary. In this paper, we recognize human dangerous behavior in privacy protection video with Gaussian blur filters for privacy protection of individuals. The dangerous behavior recognition method uses the YOLOv5 model to detect and preprocess human object from video, and then uses it as an input value for the behavior recognition model to recognize dangerous behavior. The experiments used ResNet3D, I3D, and SlowFast models, and the experimental results show that the SlowFast model achieved the highest accuracy of 95.7% in privacy-protected video. Through this, it is possible to recognize human dangerous behavior in a single-person household environment while protecting individual privacy.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.5
• /
• pp.307-314
• /
• 2023

As supercomputing and hardware technology advances, climate prediction models are improving. The Korean Meteorological Administration adopted GloSea5 from the UK Met Office and now operates an updated GloSea6 tailored to Korean weather. Universities and research institutions use Low-GloSea6 on smaller servers, improving accessibility and research efficiency. In this paper, profiling Low-GloSea6 on smaller servers identified the tri_sor_dp_dp subroutine in the tri_sor.F90 atmospheric model as a CPU-intensive hotspot. Applying linear regression, a type of machine learning, to this function showed promise. After removing outliers, the linear regression model achieved an RMSE of 2.7665e-08 and an MAE of 1.4958e-08, outperforming Lasso and ElasticNet regression methods. This suggests the potential for machine learning in optimizing identified hotspots during Low-GloSea6 execution.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.97-100
• /
• 1999

본 논문은 자동차의 편의성 및 안전성의 동시 확보를 위하여, 보조적 스위치의 조작없이 상시 음성의 입$\cdot$출력이 가능하도록 하였고, 남성과 여성을 구별하기 위하여 피치검출법을 사용하여 속도별로 구분하였다. 또한, band pass filter를 이용하여 자동으로 잡음하에서 정확하게 음성추간 검출(End Point Detection)을 하게 하였다. Reference Pattern은 DMS(Dynaminc Multi-Section)[1]모델을 사용하려고, 음성의 특징 파라미터와 인식 알고리즘은 PLP 13차와 One Stage Dynamic Programming(OSDP)를 사용하였다. 시내주행중인 자동차 환경에서 자주 사용되는 차량제어 명령어 30단어를 가지고 실험한 결과 40-80km에서 화자독립 남성 $96\%$, 여성 $94.4\%$ 화자종속일 때 남성 $97\%$, 여성 $95\%$의 인식률을 얻을수 있었고 남성과 여성을 구분하므로 써 인식률을 향상 시켰다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.53-56
• /
• 2004

본 논문은 잡음 환경하에서 특징 벡터의 차원 축소를 통한 음성 구간 검출에 관한 연구이다. 음성/비음성 분류는 통계적 모델을 이용한 분류-기반 방법을 사용한다. 검출기에서 실시간 적응화를 위해 우도-기반의 특징 벡터에 대한 차원 축소 방법을 제안한다. 이 방법은 음성/비음성 클래스에 대한 가우시안 확률 밀도 함수에 의한 비선형적 우도값을 새로운 특징으로 취하는 방법이다. 음성/비음성 결정은 우도비 검증(Likelihood Ratio Test)의 방법을 이용하며, LDA(Linear Discriminant Analys)에 의한 축소 결과와 성능을 비교한다. 실험 결과 제안된 차원 축소 방법을 통하여 2차원으로 축소된 특징 벡터가 고차원에서의 결과와 대등함을 확인하였다.

• Journal of Satellite, Information and Communications
• /
• v.8 no.4
• /
• pp.1-5
• /
• 2013

CR(Cognitive Radio) technology is maximized frequency reuse using unused frequency band. Conventional cooperative spectrum sensing has fixed false alarms. so all cooperative user has equal threshold. This problem is caused degradation of detection probability. so we propose and analysis the system model that false alarm differential setting through CSI(channel state information) for resonable threshold. Simulation results show improvement of detection probability.