• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.1
• /
• pp.85-92
• /
• 2020

Deep learning shows outstanding performance in image and video analysis, such as object classification, object detection and semantic segmentation. In this paper, it is analyzed that the performances of deep learning models can be affected by characteristics of train dataset. It is proposed as a method for selecting activation function and optimization algorithm of deep learning to classify facial expression. Classification performances are compared and analyzed by applying various algorithms of each component of deep learning model for CK+, MMI, and KDEF datasets. As results of simulation, it is shown that genetic algorithm can be an effective solution for optimizing components of deep learning model.

• Proceedings of the KIPE Conference
• /
• 2003.07a
• /
• pp.173-176
• /
• 2003

TOA(Time of Arrival) 및 TDOA(Time Difference of Arrival)경우 무선국의 시간동기화를 위해서 고도의 기술을 요구하고 있으며, 시간동기오차에 따른 위치검지의 정밀도가 낮아지는 문제가 있어 이를 극복하기 위하여 위상차(PDOA)를 이용한 새로운 열차검지기법의 제안에 따른 구현을 위하여 H/W의 설계에 대하여 기술하고자 한다 본 시스템은 전과의 전달 속도($\lambda$)를 응용하여 기준 주파수인 1.5MHz를 송신 시스템과 수신 시스템의 기준 주파수와 비교하여 그 위상의 차이를 비교하여 지연된 시간을 구한 후 이를 거리로 환산하는 시스템으로서 H/W와 S/W로 구분하여 구현 $\cdot$ 설계되는데 본 논문에서는 H/W설계에 대하여 기숙하였다.

• Structural Monitoring and Maintenance
• /
• v.4 no.2
• /
• pp.115-131
• /
• 2017

Classical flutter of wind turbine blades indicates a type of aeroelastic instability with fully attached boundary layer where a torsional blade mode couples to a flapwise bending mode, resulting in a mutual rapid growth of the amplitudes. In this paper the monitoring problem of onset of flutter is investigated from a detection point of view. The criterion is stated in terms of the exceeding of a defined envelope process of a specific maximum torsional vibration threshold. At a certain instant of time, a limited part of the previously measured torsional vibration signal at the tip of blade is decomposed through the Empirical Mode Decomposition (EMD) method, and the 1st Intrinsic Mode Function (IMF) is assumed to represent the response in the flutter mode. Next, an envelope time series of the indicated modal response is obtained in terms of a Hilbert transform. Finally, a flutter onset criterion is proposed, based on the indicated envelope process. The proposed online flutter monitoring method provided a practical and direct way to detect onset of flutter during operation. The algorithm has been illustrated by a 907-DOFs aeroelastic model for wind turbines, where the tower and the drive train is modelled by 7 DOFs, and each blade by means of 50 3-D Bernoulli-Euler beam elements.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.15 no.3
• /
• pp.119-127
• /
• 2020

The detection and recognition of abnormal driving becomes crucial for achieving safety in Intelligent Transportation Systems (ITS). This paper presents a feature extraction method based on spectral data to train a neural network model for driving behavior recognition. The proposed method uses a two stage signal processing approach to derive time-saving and efficient feature vectors. For the first stage, the feature vector set is obtained by calculating variances from each frequency bin containing the power spectrum data. The feature set is further reduced in the second stage where an intersection method is used to select more significant features that are finally applied for training a neural network model. A stream of live signals are fed to the trained model which recognizes the abnormal driving behaviors. The driving behaviors considered in this study are weaving, sudden braking and normal driving. The effectiveness of the proposed method is demonstrated by comparing with existing methods, which are Particle Swarm Optimization (PSO) and Convolution Neural Network (CNN). The experiments show that the proposed approach achieves satisfactory results with less computational complexity.

• Structural Monitoring and Maintenance
• /
• v.2 no.1
• /
• pp.1-18
• /
• 2015

Traditionally, it is not easy to carry out tests to identify modal parameters from existing railway bridges because of the testing conditions and complicated nature of civil structures. A six year (2007-2012) research program was conducted to monitor a group of 25 railway bridges. One of the tasks was to devise guidelines for identifying their modal parameters. This paper presents the experience acquired from such identification. The modal analysis of four representative bridges of this group is reported, which include B5, B15, B20 and B58A, crossing the Caraj$\acute{a}$s railway in northern Brazil using three different excitations sources: drop weight, free vibration after train passage, and ambient conditions. To extract the dynamic parameters from the recorded data, Stochastic Subspace Identification and Frequency Domain Decomposition methods were used. Finite-element models were constructed to facilitate the dynamic measurements. The results show good agreement between the measured and computed natural frequencies and mode shapes. The findings provide some guidelines on methods of excitation, record length of time, methods of modal analysis including the use of projected channel and harmonic detection, helping researchers and maintenance teams obtain good dynamic characteristics from measurement data.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.24 no.6
• /
• pp.769-776
• /
• 2000

Visible spectral characteristics of cross-sectional emissions from a partially premixed methane/air and propane/air flames have been investigated. An optical train with a two-axis scanning mirror system was used to record line-of-sight emission spectra from 354nm to 618nm, and inversion technique was adapted to obtain cross-sectional emission spectra. By analyzing the reconstructed emission spectra, cross-sectional intensities of CH and $C_2$ radicals were separated from the background emissions. The blue flame edge and yellow flame edge were also obtained by image processing technique for edge detection with color photograph of flame. These edges were compared with radial distributions of CH, $C_2$ radicals and background emissions. The CH radicals were observed at blue flame edge. The background emissions were generated by soot precursor at upstream of flame and by soot at downstream of flame. The $C_2$ radicals in propane/air flame were observed more than those in methane/air flame.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.21 no.2
• /
• pp.146-153
• /
• 2011

Model based machine condition diagnosis methods are generally using a normal and many failure models which need sufficient data to train the models. However, data, especially for failure modes of interest, is very hard to get in real applications. So their industrial applications are either severely limited or impossible when the failure models cannot be trained. In this paper, continuous hidden Markov model(CHMM) with only a normal model has been suggested as a very promising machine condition diagnosis method which can be easily used for industrial applications. Generally hidden Markov model also uses many pattern models to recognize specific patterns and the recognition results of CHMM show the likelihood trend of models. By observing this likelihood trend of a normal model, it is possible to detect failures. This method has been successively applied to arc weld defect diagnosis. The result shows CHMM's big potential as a machine condition monitoring method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.8
• /
• pp.3873-3887
• /
• 2018

In this paper, we propose a novel method to recover images corrupted by impulse noise. The proposed method uses two stages: noise detection and filtering. In the first stage, we use pixel values, rank-ordered logarithmic difference values, and median values to train a neural-network-based impulse noise detector. After training, we apply the network to detect noisy pixels in images. In the next stage, we use group-based weighted couple sparse representation to filter the noisy pixels. During this second stage, conventional methods generally use only clean pixels to recover corrupted pixels, which can yield unsuccessful dictionary learning if the noise density is high and the number of useful clean pixels is inadequate. Therefore, we use reconstructed pixels to balance the deficiency. Experimental results show that the proposed noise detector has better performance than the conventional noise detectors. Also, with the information of noisy pixel location, the proposed impulse-noise removal method performs better than the conventional methods, through the recovered images resulting in better quality.

• Journal of Electrical Engineering and Technology
• /
• v.7 no.1
• /
• pp.115-123
• /
• 2012

A video analysis system used to detect events in video streams generally has several processes, including object detection, object trajectories analysis, and recognition of the trajectories by comparison with an a priori trained model. However, these processes do not work well in a complex environment that has many occlusions, mirror effects, and/or shadow effects. We propose a new approach to a context-aware video surveillance system to detect predefined contexts in video streams. The proposed system consists of two modules: a feature extractor and a context recognizer. The feature extractor calculates the moving energy that represents the amount of moving objects in a video stream and the stationary energy that represents the amount of still objects in a video stream. We represent situations and events as motion changes and stationary energy in video streams. The context recognizer determines whether predefined contexts are included in video streams using the extracted moving and stationary energies from a feature extractor. To train each context model and recognize predefined contexts in video streams, we propose and use a new ensemble classifier based on the AdaBoost algorithm, DAdaBoost, which is one of the most famous ensemble classifier algorithms. Our proposed approach is expected to be a robust method in more complex environments that have a mirror effect and/or a shadow effect.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.11
• /
• pp.55-62
• /
• 2015

In this paper we propose a method to analyze user's motion as a game command, and implement a sports serious game applied the motion analysis method as a command interpreter. Recently, various contents platforms appear in industrial market, the computer game contents plays an important role in these emerging platforms as a killer contents. The computer game has enough values as an independent major cultural product, moreover it has the potential to be applied in various other fields such as education, healthcare, training, and so on. It could motivate users to do something continuously, and it could also support an immersive environment in a certain special game contents such as VR game. The Serious game 'ChoDeungGangHo', implemented in this paper, is the sensory healthcare serious game based on 3D run game and fitness game. It is designed for user to train the various exercise element by just playing the game, and it also supports the user management system and the linkage of social media. We proposes the sensory serious game 'ChoDeungGangHo' as a model of commercial serious game.