• The Korean Journal of Applied Statistics
• /
• v.22 no.3
• /
• pp.499-513
• /
• 2009

An economic signal in the real world usually reflects complex phenomena. One may have difficulty both extracting and interpreting information embedded in such a signal. A natural way to reduce complexity is to decompose the original signal into several simple components, and then analyze each component. Spectral analysis (Priestley, 1981) provides a tool to analyze such signals under the assumption that the time series is stationary. However when the signal is subject to non-stationary and nonlinear characteristics such as amplitude and frequency modulation along time scale, spectral analysis is not suitable. Huang et al. (1998b, 1999) proposed a data-adaptive decomposition method called empirical mode decomposition and then applied Hilbert spectral analysis to decomposed signals called intrinsic mode function. Huang et al. (1998b, 1999) named this two step procedure the Hilbert-Huang transform(HHT). Because of its robustness in the presence of nonlinearity and non-stationarity, HHT has been used in various fields. In this paper, we discuss the applications of the HHT and demonstrate its promising potential for non-stationary financial time series data provided through a Korean stock price index.

• Journal of Internet Computing and Services
• /
• v.20 no.6
• /
• pp.47-54
• /
• 2019

Social dredging of various river facilities, such as dams and agricultural reservoirs currently being constructed, should be done to ensure stable reservoirs. However, it is difficult to find a system that tells the exact amount of water in real-time in a reservoir or dam. These measurements require an automated system to collect and analyze highly accurate data in real time. In this study, we propose a method to measure the amount of water and soil of reservoir in real time through multi-division volume calculation using a drone, and this method can detect sediment conditions in real time and determine the exact timing and scale of dredging.

• The Journal of Korean Academy of Orthopedic Manual Physical Therapy
• /
• v.26 no.2
• /
• pp.19-27
• /
• 2020

Background: The purpose of this study was to examine the effects of a multi-modal exercise program for patients with chronic low back with respect to pain intensity, trunk muscle strength and Oswestry disability index. Methods: Thirty patients with chronic low back pain were recruited and divided equally into two groups. The multi-modal training program comprised a series of exercises such as warm-up, stabilization exercises, stretching, endurance exercises, and cool down whereas the control group performed only stabilization exercises. The both group spent an equal amount of time performing 60 minutes per day, three times per week, for five weeks. Results: The experimental group demonstrated statistically significant improvements in range of motion, trunk muscle strength, the visual analogue scale, and the Oswestry Disability Index (p<.05). Intergroup comparison showed a statistically significant difference in the range of motion of the lumbar spine and the degree of disability in the experimental group. Muscle strength and pain were statistically significant in both groups. Conclusion: The multi-modal exercise program is effective for patients with chronic low back pain, as it reduces lower back pain, increases trunk muscles strength, and decrease the potential for becoming disabled.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.493-496
• /
• 2009

Many AR (Augmented Reality) applications have been interested in a marker-less tracking since the tracking methods give camera poses without attaching explicit markers. In this paper, we propose a new marker-less page recognition and tracking algorithm for an AR book application such as DigilogBook. The proposed method only requires orthogonal images of pages, which need not to be trained for a long time, and the algorithm works in real-time. The page recognition is done in two steps by using SIFT (Scale Invariant Feature Transform) descriptors and the comparison evaluation function. And also, the method provides real-time tracking with 25fps ~ 30fps by separating the page recognition and the frame-to-frame matching into two multi-cores. The proposed algorithm will be extended to various AR applications that require multiple objects tracking.

• Journal of Korea Water Resources Association
• /
• v.44 no.2
• /
• pp.97-107
• /
• 2011

For the purpose of enhancing usability of NWP (Numerical Weather Prediction), the quantitative precipitation prediction scheme was suggested. In this research, precipitation by leading time was predicted using 3-hour rainfall accumulation by meso-scale numerical weather model and AWS (Automatic Weather Station), precipitation water and relative humidity observed by atmospheric sounding station, probability of rainfall occurrence by leading time in June and July, 2001 and August, 2002. Considering the nonlinear process of ranfall producing mechanism, the ANN (Artificial Neural Network) that is useful in nonlinear fitting between rainfall and the other atmospheric variables. The feedforward multi-layer perceptron was used for neural network structure, and the nonlinear bipolaractivation function was used for neural network training for converting negative rainfall into no rain value. The ANN simulated rainfall was validated by leading time using Nash-Sutcliffe Coefficient of Efficiency (COE) and Coefficient of Correlation (CORR). As a result, the 3 hour rainfall accumulation basis shows that the COE of the areal mean of the Korean peninsula was improved from -0.04 to 0.31 for the 12 hr leading time, -0.04 to 0.38 for the 24 hr leading time, -0.03 to 0.33 for the 36 hr leading time, and -0.05 to 0.27 for the 48 hr leading time.

• Annual Conference of KIPS
• /
• 2016.10a
• /
• pp.506-509
• /
• 2016

We present real-time pedestrian detection that exploit accuracy of Faster R-CNN network. Faster R-CNN has shown to success at PASCAL VOC multi-object detection tasks, and their ability to operate on raw pixel input without the need to design special features is very engaging. Therefore, in this work we apply and adjust Faster R-CNN to single object detection, which is pedestrian detection. The drawback of Faster R-CNN is its failure when object size is small. Previously, small sized object problem was solved by Scale-aware Network. We incorporate Scale-aware Network to Faster R-CNN. This made our method Scale-aware Faster R-CNN (DF R-CNN) that is both fast and very accurate. We separated Faster R-CNN networks into two sub-network, that is one for large-size objects and another one for small-size objects. The resulting approach achieves a 28.3% average miss rate on the Caltech Pedestrian detection benchmark, which is competitive with the other best reported results.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.10a
• /
• pp.641-642
• /
• 2013

This research implemented Geostationary Ocean Color Imager (GOCI) observation satellite imagery visualization system on a large high-resolution tiled display. This system is designed to help users observe or analyze satellite imagery more effectively on the tiled display using multi-touch and Kinect motion gesture recognition interaction. We developed the multi-scale image loading and rendering technique for the massive amount of memory management and smooth rendering for GOCI satellite imagery on the tiled display. In this system, the unit of time corresponding to the selected date of the satellite images are sequentially displayed on the screen. Users can zoom-in, zoom-out, move the imagery and select some buttons to trigger functions using both multi-touch or Kinect gesture interaction.

• Journal of Convergence for Information Technology
• /
• v.12 no.1
• /
• pp.31-38
• /
• 2022

This paper applies an expert independent unsupervised neural network learning-based multivariate time series data analysis model, MSCRED(Multi-Scale Convolutional Recurrent Encoder-Decoder), and to overcome the limitation, because the MCRED is based on Auto-encoder model, that train data must not to be contaminated, by using learning data sampling technique, called Subset Sampling Validation. By using the vibration data of power plant equipment that has been labeled, the classification performance of MSCRED is evaluated with the Anomaly Score in many cases, 1) the abnormal data is mixed with the training data 2) when the abnormal data is removed from the training data in case 1. Through this, this paper presents an expert-independent anomaly diagnosis framework that is strong against error data, and presents a concise and accurate solution in various fields of multivariate time series data.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.12
• /
• pp.4439-4455
• /
• 2021

Compared with two-stage object detection algorithms, one-stage algorithms provide a better trade-off between real-time performance and accuracy. However, these methods treat the intermediate features equally, which lacks the flexibility to emphasize meaningful information for classification and location. Besides, they ignore the interaction of contextual information from different scales, which is important for medium and small objects detection. To tackle these problems, we propose an image pyramid network based on dual attention mechanism (DAIPNet), which builds an image pyramid to enrich the spatial information while emphasizing multi-scale informative features based on dual attention mechanisms for one-stage object detection. Our framework utilizes a pre-trained backbone as standard detection network, where the designed image pyramid network (IPN) is used as auxiliary network to provide complementary information. Here, the dual attention mechanism is composed of the adaptive feature fusion module (AFFM) and the progressive attention fusion module (PAFM). AFFM is designed to automatically pay attention to the feature maps with different importance from the backbone and auxiliary network, while PAFM is utilized to adaptively learn the channel attentive information in the context transfer process. Furthermore, in the IPN, we build an image pyramid to extract scale-wise features from downsampled images of different scales, where the features are further fused at different states to enrich scale-wise information and learn more comprehensive feature representations. Experimental results are shown on MS COCO dataset. Our proposed detector with a 300 × 300 input achieves superior performance of 32.6% mAP on the MS COCO test-dev compared with state-of-the-art methods.

• Journal of Advanced Marine Engineering and Technology
• /
• v.25 no.3
• /
• pp.605-616
• /
• 2001

Generally, there are two methods in researching internal combustion engines. One is by experimental research and the other is by computer simulation. The experimental research has many merits that researchers can get data for engine performance, but it has also some demerit of cost and time. If there is an engine simulation code with accuracy for the solution, it is very convenient to predict the performance and optimum design value of the engine. In this study, engine performance simulation program has been improved to predict the transient variation of properties of gas in cylinder, intake and exhaust manifolds, There total program code was developed to calculate the pressure, flame factor and turbulent intensity, As a result of present study, the authors could predicted the in-cylinder pressure, intake manifold pressure and the engine performance in various conditions. The authors also could easily prepare the tool if optimum design of manifold and in-cylinder geometry.