• Communications for Statistical Applications and Methods
• /
• 제26권5호
• /
• pp.463-471
• /
• 2019

The variable transformations are useful ways to guarantee the functional relationships in the model. However, the presence of outliers may undermine the accuracy of transformation. This paper deals with response transformations in the partial linear models under the existence of outliers. A new procedure for response transformation and outliers detection is proposed. The procedure uses a sequential method for identifying outliers and dynamic graphical methods for an appropriate transformation. The graphical tools make it possible to catch diagnostic information by monitoring the movement of points in the data. The procedure is illustrated with several examples. Examples show that visual clues regarding the optimal transformation, the fittness of the model and the outlyness of the observations can be checked from the series of plots.

• 반도체디스플레이기술학회지
• /
• 제22권2호
• /
• pp.128-132
• /
• 2023

Recently, with the development of deep learning technology, research on pedestrian attribute recognition technology using deep neural networks has been actively conducted. Existing pedestrian attribute recognition techniques can be obtained in such a way as global-based, regional-area-based, visual attention-based, sequential prediction-based, and newly designed loss function-based, depending on how pedestrian attributes are detected. It is known that the performance of these pedestrian attribute recognition technologies varies greatly depending on the type of backbone network that constitutes the deep neural networks model. Therefore, in this paper, several backbone networks are applied to the baseline pedestrian attribute recognition model and the performance changes of the model are analyzed. In this paper, the analysis is conducted using Resnet34, Resnet50, Resnet101, Swin-tiny, and Swinv2-tiny, which are representative backbone networks used in the fields of image classification, object detection, etc. Furthermore, this paper analyzes the change in time complexity when inferencing each backbone network using a CPU and a GPU.

• KSBB Journal
• /
• 제13권5호
• /
• pp.577-584
• /
• 1998

A sensitive membrane strip assay for plasma lipoprotein cholesterol that can be performed without handling reagents has been investigated. We previously developed an assay system with immobilized enzymes (cholesterol esterase and cholesterol oxidase) on the surfaces of nitrocellulose membrane(1). In such a case, the amount of enzymes present on the membrane was limited by its surface area and, thus, the detection capability was relatively poor (> 50 mg/dL cholesterol). To overcome this problem, we devised a new system with non-immobilized enzymes by placing them within interstitial spaces of a celullose membrane pad in a dry state. Upon contact with sample medium, the enzymes were immediately dissolved and participated in the reactions with cholesterol in a liquid phase. We constructed a user-friendly system consisting of four membrane pads fro sample application, cholesterol decomposition, color development as signal, and medium absorption to invoke a continuous flow (sequential location from the bottom). A sample containing lipoproteins was added into the application pad by capillary action and transferred to the next pad for decomposition. The decomposition pad (namely, enzyme pad) contained a detergent (sodium cholate) for the destruction of lipoprotein particles, the two enzymes for cholesterol decomposition, and a chromogen (3,3'-diaminobenzidine). As a consequence of the enzyme reactions, hydrogen peroxide was produced, and then reacted in the presence of the chromogen with horseradish peroxidase immobilized on the signal generation pad. Finally, a colorimetric signal directly proportional to the cholesterol concentration was produced. The detection limit determined from this system under optimal conditions was at least 2 times lower than of the enzyme-immobilized system.

• 대한간호학회지
• /
• 제42권3호
• /
• pp.424-433
• /
• 2012

Purpose: To investigate the accuracy, precision and validity of fever detection of tympanic membrane (TM), temporal artery (TA) and axillary temperature (AT) compared with pulmonary artery temperature (PA). Methods: Repeated-measures design was conducted for one year on 83 adult cardiac care unit patients with pulmonary artery catheters after open heart surgery. Sequential temperature measurements were taken three times at 20-minute intervals. Accuracy, precision, repeatability, and validity of fever detection were analyzed. Results: Mean pulmonary artery temperature was $37.04^{\circ}C$ (SD $0.70^{\circ}C$). The mean (SD) offsets from PA, with the mean reflecting accuracy and SD reflecting precision, were $-1.31^{\circ}C$ ($0.75^{\circ}C$) for TA, $-0.20^{\circ}C$ ($0.24^{\circ}C$) for TM, and $-0.97^{\circ}C$ ($0.64^{\circ}C$) for AT. Percentage of pairs with differences within ${\pm}0.5^{\circ}C$ was 9.6% for TA, 19.7% for AT, and 91.6% for TM. Repeated measurements with all three methods had mean SD values within $0.04^{\circ}C$. Sensitivity, specificity, and positive and negative predictive values of tympanic measurements were 0.76, 1.0, and 1.0, and 0.90, respectively. Conclusion: Results show that TM best reflects PA, and is most consistent, accurate, and precise. AT tends to underestimate PA, and TA is least accurate and precise. Therefore tympanic membrane measurement is a reliable alternative to other non-invasive methods of measuring temperatures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권2호
• /
• pp.800-816
• /
• 2018

In recent years, video surveillance research has been able to recognize various behaviors of pedestrians and analyze the overall situation of objects by combining image analysis technology and deep learning method. Human Activity Recognition (HAR), which is important issue in video surveillance research, is a field to detect abnormal behavior of pedestrians in CCTV environment. In order to recognize human behavior, it is necessary to detect the human in the image and to estimate the pose from the detected human. In this paper, we propose a novel approach for 2D Human Pose Estimation based on object detection using RGB-D information. By adding depth information to the RGB information that has some limitation in detecting object due to lack of topological information, we can improve the detecting accuracy. Subsequently, the rescaled region of the detected object is applied to ConVol.utional Pose Machines (CPM) which is a sequential prediction structure based on ConVol.utional Neural Network. We utilize CPM to generate belief maps to predict the positions of keypoint representing human body parts and to estimate human pose by detecting 14 key body points. From the experimental results, we can prove that the proposed method detects target objects robustly in occlusion. It is also possible to perform 2D human pose estimation by providing an accurately detected region as an input of the CPM. As for the future work, we will estimate the 3D human pose by mapping the 2D coordinate information on the body part onto the 3D space. Consequently, we can provide useful human behavior information in the research of HAR.

• 전자공학회논문지S
• /
• 제35S권8호
• /
• pp.55-67
• /
• 1998

본 논문에서는 컬러 비디오 시퀀스 상에서 눈과 입에 해당하는 얼굴 특징점을 고속으로 추출하는 방법을 제안한다. 자유로운 움직임을 갖는 얼굴 영역을 안정적으로 추출하기 위해 얼굴 색상 분포를 이용한 색상 변환 영상에 움직임 검출 기법을 적용하여 움직이는 살색 부분만을 효율적으로 검출하는 색상 움직임 개념을 사용하였다. 움직임 정보는 살색의 가능성 정도에 따라 가중치가 주어지며 화소 단위의 움직임 여부를 결정하는 문턱값도 살색의 가능성 정도에 따라 적응적으로 결정된다. 눈의 색상분포와 형태소 연산자를 사용한 움직임 살색 영역에서 눈 후보 영역을 추출하고 눈과 눈썹의 상호 위치 관계를 이용하여 눈의 영역을 최종 결정한다. 입의 영역은 눈의 위치를 기준으로 입 후보 영역을 정하고 색상 히스토그램을 이용하여 입의 영역을 검출한다. 찾아진 눈과 입의 영역에서 정확한 특징점의 위치를 구하기 위해 PCA (Principal Component Analysis)를 사용하였다. 실험 결과 복잡한 배경, 개인적인 편차, 얼굴의 방향과 크기 등에 영향을 받지 않고 고속으로 정확한 얼굴의 특징점을 추출할 수 있었다.

• Journal of Ginseng Research
• /
• 제43권4호
• /
• pp.684-691
• /
• 2019

Background: We have reported that defective nef and gag genes are induced in HIV-1-infected patients treated with Korean Red Ginseng (KRG). Methods: To investigate whether KRG treatment and highly active antiretroviral therapy (HAART) affect genetic defects in the pol gene, we amplified and sequenced a partial pol gene (p-pol) containing the integrase portion (1.2 kb) by nested PCR with sequential peripheral blood mononuclear cells over 20 years and compared it with those patients at baseline, in control patients, those taking ginseng-based combination therapy (GCT; KRG plus combinational antiretroviral therapy) and HAART alone. We also compared our findings to look for the full-length pol gene (pol) (3.0-kb) Results: Twenty-patients infected with subtype B were treated with KRG for $116{\pm}58months$ in the absence of HAART. Internal deletion in the pol gene (${\Delta}pol$) was significantly higher in the KRG group (11.9%) than in the control group and at baseline; its detection was significantly inhibited during GCT as much as during HAART. In addition, the ${\Delta}pol$ in p-pol significantly depended on the duration of KRG treatment. In pol, the proportion of ${\Delta}pol$ was significantly higher in the KRG group (38.7%) than in the control group, and it was significantly inhibited during GCT and HAART. In contrast, the proportion of stop codon appeared not to be affected by KRG treatment. The PCR success rate was significantly decreased with longer GCT. Conclusion: The proportion of ${\Delta}pol$ depends on template size as well as KRG treatment. HAART decreases the detection of ${\Delta}pol$.

• 항공우주시스템공학회지
• /
• 제18권4호
• /
• pp.43-52
• /
• 2024

본 논문은 부유식 풍력터빈의 블레이드 피치 시스템에서 발생하는 이상을 조기에 감지하기 위한 LSTM-Autoencoder 모델 기반의 이상징후 감지 시스템을 설명한다. 발전소 모니터링 시스템에 활용되는 센서 데이터는 주로 시계열 데이터로 구성되며, LSTM 네트워크는 이러한 시계열 데이터를 분석하기 위해 두 개의 단방향 LSTM 네트워크로 구성된다. 이를 통해 순차 데이터에 숨겨진 장기 의존성을 효과적으로 발견할 수 있다. 한편, 오토인코더 메커니즘은 정상상태 데이터로부터만 학습하여 이상상태를 분류될 수 있기 때문에 이 두 가지 네트워크를 결합하여 시스템에 발생하는 이상징후를 효과적으로 감지할 수 있다. 제안된 프레임워크의 효과를 입증하기 위해 풍력 터빈 모델에서 수집한 실제 다변량 시계열 데이터셋을 적용하였다. LSTM-AE 모델은 높은 이상징후 감지 정확도를 달성하여 우수한 성능을 보였다.

• 한국통신학회논문지
• /
• 제19권11호
• /
• pp.2094-2105
• /
• 1994

본 논문은 레이돈(Radon) 변환식으로부터 유도된 투사 근거 허프(Hough) 변환 방식을 사용하여 시간연속 영상상의 이동물체의 궤적을 추정하는 기법을 제안한다. 이때 이동비행물체는 시간연속되는 각 영상 프레임에 몇 개의 화소로 나타나며 그 궤적은 삼차원 직선으로 간주한다. 근래에 제시된 방법들은 합성영상 입단일 궤적도면을 여러 종류의 허프변환 방식을 사용하여 그 궤적을 추정하여 왔으나 본 논문의 방식은 시간연속 영상을 여러 방향에서 투사하여 얻어지는 이차원적 비행체 궤적 지식을 효과적으로 비행체 궤적 재구성에 이용함으로 비행체 탐지 능력은 물론 궤적 추정 능력을 향상하였다. 아울러 투사 근거 허프 변환방식을 평가, 사정하기 위하여 영상 공간의 소음 등으로 야기된 투사 공간상의 허프 파라미터의 추정 error를 분석, 유도하였다. HiCamps라 명칭된 실제 적외선 시간연속 영상 데이타를 대상으로 시뮬레이션한 결과 비행체 궤적 추정이 아주 낮은 SNR 에서도 가능함을 보여준다.

• 폴리머
• /
• 제32권4호
• /
• pp.366-371
• /
• 2008

The controlled/living cationic polymerization of $\alpha$-methylstyrene (${\alpha}MeSt$) and sequential block copolymerization of isobutylene (IB) with ${\alpha}MeSt$ were achieved using 2-chloro-2,4,4-trimethylpentane (TMPCl)/titanium tetrachloride ($TiCl_4$)/titanium isopropoxide ($Ti(OiPr)_4$)/2,6-ditert-butylpyridine (DtBP) initiating system in $CH_3Cl$/hexane(50/50 v/v) solvent mixture at $-80^{\circ}C$. The polymerization rate decreased with increasing $[Ti(OiPr)_4]/[TiCl_4]$ ratio in the homopolymerization of ${\alpha}MeSt$. The effects of $[Ti(OiPr)_4]/[TiCl_4]$ ratios and $PIB^+$ molecular weight on the polymerization rate and blocking efficiency were also investigated. Well-defined poly(isobutylene-b-$\alpha$-methylstyrene)s were demonstrated by $^1H$-NMR and triple detection SEC; refractive index (RI), multiangle laser light scattering (MALLS) and ultraviolet (UV) detectors. Blocking efficiencies for the poly(isobutylene-b-$\alpha$-methylstyrene)s of almost 100% were obtained when ${\alpha}MeSt$ was induced by PIB's of $M_n\;{\geq}\;41000$ at $[Ti(OiPr)_4]/[TiCl_4]=1$. Differential scanning calorimetry (DSC) of the block copolymers showed two glass transition temperatures, thereby demonstrating microphase separation.