• Annual Conference of KIPS
• /
• 2021.05a
• /
• pp.354-357
• /
• 2021

Infants communicate their feelings and needs to the outside world through non-verbal methods such as crying and displaying diverse facial expressions. However, inexperienced parents tend to decode these non-verbal messages incorrectly and take inappropriate actions, which might affect the bonding they build with their babies and the cognitive development of the newborns. In this paper, we propose an aural-visual two-stream based infant cry recognition system to help parents comprehend the feelings and needs of crying babies. The proposed system first extracts the features from the pre-processed audio and video data by using the VGGish model and 3D-CNN model respectively, fuses the extracted features using a fully connected layer, and finally applies a SoftMax function to classify the fused features and recognize the corresponding type of cry. The experimental results show that the proposed system classification exceeds 0.92 in F1-score, which is 0.08 and 0.10 higher than the single-stream aural model and single-stream visual model.

• Geomechanics and Engineering
• /
• v.34 no.4
• /
• pp.341-380
• /
• 2023

This study evaluates the seismic fragilities in fill slopes in South Korea through parametric finite element analyses that have been barely investigated thus far. We consider three slope geometries for a slope of height 10 m and three slope angles, and two soil types, namely frictional and frictionless, associated with two soil states, loose and dense for frictional soils and soft and stiff for frictionless soils. The input ground motions accounting for four site conditions in South Korea are obtained from one-dimensional site response analyses. By comparing the numerical modeling of slopes using PLAXIS^2D against the previous studies, we compiled suites of the maximum permanent slope displacement (D_max) against two ground motion parameters, namely, peak ground acceleration (PGA) and Arias Intensity (I_A). A probabilistic seismic demand model is adopted to compute the probabilities of exceeding three limit states (minor, moderate, and extensive). We propose multiple seismic fragility curves as functions of a single ground motion parameter and numerous seismic fragility surfaces as functions of two ground motion parameters. The results show that soil type, slope angle, and input ground motion influence these probabilities, and are expected to help regional authorities and engineers assess the seismic fragility of fill slopes in the road systems in South Korea.

• Journal of Plant Biotechnology
• /
• v.47 no.2
• /
• pp.157-163
• /
• 2020

Calmodulin (CaM) mediates cellular Ca²⁺ signals in the defense responses of plants. We previously reported that GmCaM-4 and 5 are involved in salicylic acid-independent activation of disease resistance responses in soybean (Glycine max). Here, we generated a GmCaM-4 cDNA construct under the control of the cauliflower mosaic virus (CaMV) 35S promoter and transformed this construct into potato (Solanum tuberosum L.). The constitutive over-expression of GmCaM-4 in potato induced high-level expression of pathogenesis-related (PR) genes, such as PR-2, PR-3, PR-5, phenylalanine ammonia-lyase (PAL), and proteinase inhibitorII (pinII). In addition, the transgenic potato plants exhibited enhanced resistance against a bacterial pathogen, Erwinia carotovora ssp. Carotovora (ECC), that causes soft rot disease and showed spontaneous lesion phenotypes on their leaves. These results strongly suggest that a CaM protein in soybean, GmCaM-4, plays an important role in the response of potato plants to pathogen defense signaling.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.40 no.4
• /
• pp.193-205
• /
• 2003

There has been so many research activities about robot soccer system in the many research fields, for example, intelligent control, communication, computer technology, sensor technology, image processing, mechatronics. Especially researchers research strategy for attacking in the field of strategy, and develop intelligent strategy. Then, soccer robots cannot defense completely and efficiently by using simple defense strategy. Therefore, intention extraction of attacker is needed for efficient defense. In this thesis, intention extractor of soccer robots is designed and developed based on FMMNN(Fuzzy Min-Max Neural networks ). First, intention for soccer robot system is defined, and intention extraction for soccer robot system is explained.. Next, FMMNN based intention extractor for soccer robot system is determined. FMMNN is one of the pattern classification method and have several advantages: on-line adaptation, short training time, soft decision. Therefore, FMMNN is suitable for soccer robot system having dynamic environment. Observer extracts attack intention of opponents by using this intention exactor, and this intention extractor is also used for analyzing strategy of opponent team. The capability of developed intention extractor is verified by simulation of 3 vs. 3 robot succor simulator. It was confirmed that the rates of intention extraction each experiment increase.

• The Journal of Engineering Geology
• /
• v.11 no.2
• /
• pp.175-190
• /
• 2001

This study investigates the existing theoretical backgrounds in order to examine the behavior of lateral flow according to the plasticity of soils when unsymmetrical surcharge is worked on polluted soft soils by comparing and analyzing the results measured through model tests. Model tests are canied out as follows soil tank, bearing frame and bearing plate are made. By increasing unsymmetrical surcharge to the ground soils with the consistent water content and with gradually increased polluted materials at intervals, the amounts of settlement, lateral displacement and upheaval were respectively observed. In conclusion, the value of critical surcharge was expressed as q$_{cr}$=2.78$_{cu}$ which was similar to those Tschebotarioff(q$_{cr}$=3.0$_{cu}$) and Meyerhof(q$_{cr}$=(B/2H+$\pi$/2)$_{cu}$) had been proposed. The value of ultimate capacity was expressed as q$_{ult}$=4.84$_{cu}$ which was similar to that of Prandtl. The lateral flow pressure is adeQuately calculated by the eQuation(P$_{max}$=K$_o$ r H) and the maximum value of lateral flow pressure is found near O.3H of layer thickness(H) and is higher to ground surface than the ones in composition pattern, Poulos distribution pattern and softclay soils (CL, CH) which is not polluted. The stability control method used in this research followed the management diagram of Tominaga.Hashimoto, Shibata.Sekiguchi, Matsuo.Kawamura who use the amounts of plasticity displacement by lateral flow. As a result, the ultimate capacity values in the diagram {S$_v$-(Y$_m$/S$_v$)} of Matsuo.Kawamura and in the diagram {(q/Y$_m$)-q} of Shibata. Sekiguchi were smaller than in the ones of load-settlement curve (q-S$_v$).

• Journal of Korea Multimedia Society
• /
• v.23 no.8
• /
• pp.940-952
• /
• 2020

The Alzheimer's disease (AD) is a neurodegenerative disease commonly found in the elderly individuals. It is one of the most common forms of dementia; patients with AD suffer from a degradation of cognitive abilities over time. To correctly diagnose AD, compuated-aided system equipped with automatic classification algorithm is of great importance. In this paper, we propose a novel deep learning based classification algorithm that takes advantage of MRI biomarker images including brain areas of hippocampus and cerebrospinal fluid for the purpose of improving the AD classification performance. In particular, we develop a new approach that effectively applies MRI biomarker patch images as input to 3D Deep Convolution Neural Network. To integrate multiple classification results from multiple biomarker patch images, we proposed the effective confidence score fusion that combine classification scores generated from soft-max layer. Experimental results show that AD classification performance can be considerably enhanced by using our proposed approach. Compared to the conventional AD classification approach relying on entire MRI input, our proposed method can improve AD classification performance of up to 10.57% thanks to using biomarker patch images. Moreover, the proposed method can attain better or comparable AD classification performances, compared to state-of-the-art methods.

• Journal of Biomedical Engineering Research
• /
• v.38 no.1
• /
• pp.25-31
• /
• 2017

Testing TB in chest X-ray images is a typical method to diagnose presence and magnitude of PTB lesion. However, the method has limitation due to inter-reader variability. Therefore, it is essential to overcome this drawback with automatic interpretation. In this study, we propose a novel method for detection of PTB using SegNet, which is a deep learning architecture for semantic pixel wise image labelling. SegNet is composed of a stack of encoders followed by a corresponding decoder stack which feeds into a soft-max classification layer. We modified parameters of SegNet to change the number of classes from 12 to 2 (TB or none-TB) and applied the architecture to automatically interpret chest radiographs. 552 chest X-ray images, provided by The Korean Institute of Tuberculosis, used for training and test and we constructed a receiver operating characteristic (ROC) curve. As a consequence, the area under the curve (AUC) was 90.4% (95% CI:[85.1, 95.7]) with a classification accuracy of 84.3%. A sensitivity was 85.7% and specificity was 82.8% on 431 training images (TB 172, none-TB 259) and 121 test images (TB 63, none-TB 58). This results show that detecting PTB using SegNet is comparable to other PTB detection methods.

• Tunnel and Underground Space
• /
• v.12 no.4
• /
• pp.259-267
• /
• 2002

Recently, Umbrella Arch Method(UAM) is commonly used in order to enhance the stability of tunnel itself and stabilize the adjacent surface structure. But quantitative estimation of reinforcement effect is needed because UAM is designed and constructed only on the basis of empirical experience. By using 3-dimensional finite element method, parametric study is performed for elastic modulus of ground and overburden, and reinforcement effect is analyzed quantitatively. From the results, surface settlement decreases about 9%∼27% in soil tunnel, about 4%∼24% in weathered rock tunnel and 4%∼17% in soft rock tunnel when applied with UAM. The prediction equation for final surface settlement is suggested through regression analysis and the equation is expressed as exponential function which has variable Smax, unknown coefficient i and k.

• The Bulletin of The Korean Astronomical Society
• /
• v.37 no.1
• /
• pp.89.1-89.1
• /
• 2012

We have investigated a coronal jet observed near the limb on 2010 June 27 by the Hinode/X-Ray Telescope (XRT), EUV Imaging Spectrograph (EIS), and Solar Optical Telescope (SOT), and the SDO/Atmospheric Imaging Assembly (AIA), Helioseismic and Magnetic Imager (HMI), and on the disk by STEREO-A/EUVI. From EUV (AIA and EIS) and soft X-ray (XRT) images we have identified both cool and hot jets. There was a small loop eruption in Ca II images of the SOT before the jet eruption. Using high temporal and multi wavelength AIA images, we found that the hot jet preceded its associated cool jet by about 2 minutes. The cool jet showed helical-like structures during the rising period. According to the spectroscopic analysis, the jet's emission changed from blue to red shift with time, implying helical motions in the jet. The STEREO observation, which enabled us to observe the jet projected against the disk, showed that there was a dim loop associated with the jet. We have measured a propagation speed of ~800 km/s for the dimming front. This is comparable to the Alfven speed in the loop computed from a magnetic field extrapolation of the HMI photospheric field measured 5 days earlier and the loop densities obtained from EIS Fe XIV line ratios. We interpret the dimming as indicating the presence of Alfvenic waves initiated by reconnection in the upper chromosphere.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.2
• /
• pp.55-60
• /
• 2011

In this paper, we propose a new method for extracting muscles from lumbar images. The proposed method sets areas without distortions with field expert's assistance as areas of measuring interest and removing noises from initial ultrasonic videos. Then, the method emphasizes the brightness contrast with Ends-in search stretching algorithm and separate thoracic vertebra from subcutaneous fat area using morphological characteristics. 4-directions contour tracing algorithm is applied to extract the bottom of subcutaneous fat area. Extracting thoracic vertebra area requires noise removal and morphological characteristics as well among candidate areas obtained by controlling min-max brightness. The thickness of muscles is then defined as the length between subcutaneous fat area and extracted thoracic vertebra. The experiment which consists of 368 image analysis verifies that the proposed method is more effective in measuring the thickness of muscles than before.