• 한국멀티미디어학회논문지
• /
• 제20권8호
• /
• pp.1282-1290
• /
• 2017

Since the mitigation of fear of crime significantly enhances the consumptions in a city, studies focusing on urban safety analysis have received much attention as means of revitalizing the local economy. In addition, with the development of computer vision and machine learning technologies, efficient and automated analysis methods have been developed. Previous studies have used global features to predict the safety of cities, yet this method has limited ability in accurately predicting abstract information such as safety assessments. Therefore we used a Convolutional Context Neural Network (CCNN) that considered "context" as a decision criterion to accurately predict safety of cities. CCNN model is constructed by combining a stacked auto encoder with a fully connected network to find the context and use it in the CNN model to predict the score. We analyzed the RMSE and correlation of SVR, Alexnet, and Sharing models to compare with the performance of CCNN model. Our results indicate that our model has much better RMSE and Pearson/Spearman correlation coefficient.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권2호
• /
• pp.366-385
• /
• 2013

In this paper, we propose a compressed sensing (CS) based Wyner-Ziv (WZ) codec using motion-aligned auto regressive model (MAAR) based side information (SI) extrapolation to improve the compression performance of low-delay distributed video coding (DVC). In the CS based WZ codec, the WZ frame is divided into small blocks and CS measurements of each block are acquired at the encoder, and a specific CS reconstruction algorithm is proposed to correct errors in the SI using CS measurements at the decoder. In order to generate high quality SI, a MAAR model is introduced to improve the inaccurate motion field in auto regressive (AR) model, and the Tikhonov regularization on MAAR coefficients and overlapped block based interpolation are performed to reduce block effects and errors from over-fitting. Simulation experiments show that our proposed CS based WZ codec associated with MAAR based SI generation achieves better results compared to other SI extrapolation methods.

• Advances in Computational Design
• /
• 제4권4호
• /
• pp.367-379
• /
• 2019

Multiple Inputs and Multiple Outputs (MIMO) Fuzzy logic model is developed to predict the engine performance and emission characteristics of pongamia pinnata biodiesel with hydrogen injection. Engine performance and emission characteristics such as brake thermal efficiency (BTE), brake specific energy consumption (BSEC), hydrocarbon (HC), carbon monoxide (CO), carbon dioxide ($CO_2$) and nitrous oxides ($NO_X$) were considered. Experimental investigations were carried out by using four stroke single cylinder constant speed compression ignition engine with the rated power of 5.2 kW at variable load conditions. The performance and emission characteristics are measured using an Exhaust gas analyzer, smoke meter, piezoelectric pressure transducer and crank angle encoder for different fuel blends (Diesel, B10, B20 and B30) and engine load conditions. Fuzzy logic model uses triangular and trapezoidal membership function because of its higher predictive accuracy to predict the engine performance and emission characteristics. Computational results clearly demonstrate that, the proposed fuzzy model has produced fewer deviations and has exhibited higher predictive accuracy with acceptable determination correlation coefficients of 0.99136 to 1 with experimental values. The developed fuzzy logic model has produced good correlation between the fuzzy predicted and experimental values. So it is found to be useful for predicting the engine performance and emission characteristics with limited number of available data.

• 인터넷정보학회논문지
• /
• 제22권3호
• /
• pp.9-16
• /
• 2021

This study aims to develop a human activity recognition (HAR) system as a Deep-Learning (DL) classification model, distinguishing various human activities. We solely rely on the signals from a wristband accelerometer worn by a person for the user's convenience. 3-axis sequential acceleration signal data are gathered within a predefined time-window-slice, and they are used as input to the classification system. We are particularly interested in developing a Deep-Learning model that can outperform conventional machine learning classification performance. A total of 13 activities based on the laboratory experiments' data are used for the initial performance comparison. We have improved classification performance using the Convolutional Neural Network (CNN) combined with an auto-encoder feature reduction and parameter tuning. With various publically available HAR datasets, we could also achieve significant improvement in HAR classification. Our CNN model is also compared against Recurrent-Neural-Network(RNN) with Long Short-Term Memory(LSTM) to demonstrate its superiority. Noticeably, our model could distinguish both general activities and near-identical activities such as sitting down on the chair and floor, with almost perfect classification accuracy.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 춘계학술발표대회
• /
• pp.706-708
• /
• 2023

In this study, we embark on a journey to uncover the essence of emotions by exploring the depths of transfer learning on three pre-trained transformer models. Our quest to classify five emotions culminates in discovering the KLUE (Korean Language Understanding Evaluation)-BERT (Bidirectional Encoder Representations from Transformers) model, which is the most exceptional among its peers. Our analysis of F1 scores attests to its superior learning and generalization abilities on the experimental data. To delve deeper into the mystery behind its success, we employ the powerful SHAP (Shapley Additive Explanations) method to unravel the intricacies of the KLUE-BERT model. The findings of our investigation are presented with a mesmerizing text plot visualization, which serves as a window into the model's soul. This approach enables us to grasp the impact of individual tokens on emotion classification and provides irrefutable, visually appealing evidence to support the predictions of the KLUE-BERT model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권9호
• /
• pp.2334-2347
• /
• 2023

Metaverse services generate text data, data of ubiquitous computing, in real-time to analyze user emotions. Analysis of user emotions is an important task in metaverse services. This study aims to classify user sentiments using deep learning and pre-trained language models based on the transformer structure. Previous studies collected data from a single platform, whereas the current study incorporated the review data as "Metaverse" keyword from the YouTube and Google Play Store platforms for general utilization. As a result, the Bidirectional Encoder Representations from Transformers (BERT) and Robustly optimized BERT approach (RoBERTa) models using the soft voting mechanism achieved a highest accuracy of 88.57%. In addition, the area under the curve (AUC) score of the ensemble model comprising RoBERTa, BERT, and A Lite BERT (ALBERT) was 0.9458. The results demonstrate that the ensemble combined with the RoBERTa model exhibits good performance. Therefore, the RoBERTa model can be applied on platforms that provide metaverse services. The findings contribute to the advancement of natural language processing techniques in metaverse services, which are increasingly important in digital platforms and virtual environments. Overall, this study provides empirical evidence that sentiment analysis using deep learning and pre-trained language models is a promising approach to improving user experiences in metaverse services.

• 동력기계공학회지
• /
• 제17권3호
• /
• pp.82-88
• /
• 2013

For commercialized servo drives of the motion stage to include embedded controller, external terminal is provided for tracking command and encoder output, but internal terminal is not for control input. Thus, it is difficult to combine out signal of embedded controller with that of external compensator such as disturbance observer. In this study, for precise tracking control of motion stage without hardware change of the servo drive, tacking control system is composed of an inner loop of servo drive and an outer loop of disturbance observer. Then, the control system is designed so that the output response of actual plant corresponds with nominal model's in transient state as well as in steady state. Finally, the experiment results show that the designed control system is effective to reconcile actual plant behavior with nominal model under nonlinear friction and parameter perturbation.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.1512-1515
• /
• 1997

One of the advantages of master-slave teleoperation is scaling concept such as position scaling, force scaling Meanuhile, lots of quantization effects are generated from position and force sensors in the master and slave manipulator. In this paper, to show the output error caused by the quantizaion effects from the position sensor and position scaling factor, simulation is done for free motion without contact in slave side. Transfer functiion model in which the quantization effect is assumed to be a disturbance input to the system is derived. Model shows that Jacobian, scaling factors, and controller affect the output by quantization effects form esnsors. One dof master and slave are used for simulation. In our study, the higher sensor resolution decreases the output error form quantization. Scaling factors can amplify the quantizatiion effects form the sensors in master and slave manipulators.

• 제어로봇시스템학회논문지
• /
• 제9권5호
• /
• pp.354-360
• /
• 2003

For the convenient use of high performance microprocessor in motor control, peripheral devices are needed for converting its control signals to compatible ones for motor drive. Customized devices are not plentiful far these purposes and their functions do not usually satisfied designers specification. The designers used to implement these functions on FPGA or CPLD using hardware description language. Then, in this case unessential programs are needed for control the peripherals. In this paper, a unified device model that links peripheral devices, including especially the pulse width modulation controller and the quadrature encoder interface device, to an interrupt controller is proposed. Advanced functions of peripherals could be achieved by this model and unessential programs can be simplified. Block diagrams and flowcharts are presented to illustrate the advanced functions. This unified device was designed using VHDL. The simulation results were presented to demonstrate the effectiveness of the proposed scheme.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -1
• /
• pp.289-292
• /
• 2002

This paper provides a frame-layer method for controlling bit rate of compressed video data in real time. Our approach is easy to operate and can store encoded video data in real time without deteriorating the quality of an image. To provide ameliorated and consistent visual quality, a new concept named SOP (Set Of Pictures) and a new quantization parameter variation control algorithm based on a second-order rate-distortion model 〔2〕 are introduced. The total bit-budget is allocated efficiently to cope with unpredictable recording time by using the proposed algorithm and it is distributed to each frame. In the end, we show improved and consistent video quality with experimental results obtained from C-model of a MPEG-4 (simple-profile) encoder.