• 한국전자통신학회논문지
• /
• 제18권2호
• /
• pp.321-327
• /
• 2023

인공 지능이 발전함에 따라 예측 시스템은 우리의 삶에 필수적인 기술 중 하나로 자리를 잡았다. 이러한 기술의 성장에도 불구하고, 21세기 사거리 교통 체증은 계속해서 문제 되어 왔다. 본 논문에서는 Conv-LSTM(: Convolutional-Long Short-Term Memory) 알고리즘을 이용한 사거리 교통 체증 예측 시스템을 제안한다. 제안한 시스템은 교통 체증이 발생하는 사거리에 시간대별 교통 정보를 학습한 데이터를 모델링 한다. 시간의 흐름에 따라 기록된 교통량 데이터로 교통 체증을 예측하며. 예측된 결과를 기반으로 사거리 교통 신호를 제어하고, 일정한 교통량으로 유지한다. VDS(: Vehicle Detection System)센서를 활용하여 도로 혼잡도 데이터를 정의하고, 교통을 원활하게 하기 위하여 각각의 교차로를 Conv-LSTM 알고리즘기반 네트워크 시스템으로 구성하였다.

• Korean Journal of Radiology
• /
• 제23권1호
• /
• pp.139-149
• /
• 2022

Objective: To compare the effects of bone suppression imaging using deep learning (BSp-DL) based on a generative adversarial network (GAN) and bone subtraction imaging using a dual energy technique (BSt-DE) on radiologists' performance for pulmonary nodule detection on chest radiographs (CXRs). Materials and Methods: A total of 111 adults, including 49 patients with 83 pulmonary nodules, who underwent both CXR using the dual energy technique and chest CT, were enrolled. Using CT as a reference, two independent radiologists evaluated CXR images for the presence or absence of pulmonary nodules in three reading sessions (standard CXR, BSt-DE CXR, and BSp-DL CXR). Person-wise and nodule-wise performances were assessed using receiver-operating characteristic (ROC) and alternative free-response ROC (AFROC) curve analyses, respectively. Subgroup analyses based on nodule size, location, and the presence of overlapping bones were performed. Results: BSt-DE with an area under the AFROC curve (AUAFROC) of 0.996 and 0.976 for readers 1 and 2, respectively, and BSp-DL with AUAFROC of 0.981 and 0.958, respectively, showed better nodule-wise performance than standard CXR (AUAFROC of 0.907 and 0.808, respectively; p ≤ 0.005). In the person-wise analysis, BSp-DL with an area under the ROC curve (AUROC) of 0.984 and 0.931 for readers 1 and 2, respectively, showed better performance than standard CXR (AUROC of 0.915 and 0.798, respectively; p ≤ 0.011) and comparable performance to BSt-DE (AUROC of 0.988 and 0.974; p ≥ 0.064). BSt-DE and BSp-DL were superior to standard CXR for detecting nodules overlapping with bones (p < 0.017) or in the upper/middle lung zone (p < 0.017). BSt-DE was superior (p < 0.017) to BSp-DL in detecting peripheral and sub-centimeter nodules. Conclusion: BSp-DL (GAN-based bone suppression) showed comparable performance to BSt-DE and can improve radiologists' performance in detecting pulmonary nodules on CXRs. Nevertheless, for better delineation of small and peripheral nodules, further technical improvements are required.

• 융합신호처리학회논문지
• /
• 제12권4호
• /
• pp.341-347
• /
• 2011

본 논문에서는 유효 빔 폭 개념을 새로이 도입하여 중첩 초음파 센서 링의 성능 평가, 최적 설계, 그리고 복합 장애물 탐지에 대해 기술하도록 한다. 일군의 동종 저지향성 초음파 센서들이 반경이 영이 아닌 원주 상에 일정 간격으로 상호 빔 폭이 중첩되도록 배치된다고 가정한다. 첫째, 중첩 초음파 센서 링의 전역 위치 불확실성을 전체 장애물 탐지 범위 내에서의 국부적 위치 불확실성의 평균값으로 나타낸다. 중첩 초음파 센서 링의 유효 빔 폭을 동일한 전역 위치 불확실성을 갖는 단일 초음파 센서의 빔 폭으로 산정하고 이를 바탕으로 정규화된 장애물 탐지 성능 평가 지수를 정의한다. 둘째, 정의된 성능 평가 지수를 이용하여 장애물 탐지 시 위치 불확실성이 최소화되도록 중첩 초음파 센서 링의 설계 사양을 최적화한다. 주어진 초음파 센서의 사용 개수에 대한 중첩 초음파 센서 링의 최적 반경 그리고 주어진 중첩 초음파 센서 링의 반경에 대한 초음파 센서의 최적 사용 개수를 결정한다. 셋째, 3개의 인접 초음파 센서로부터의 장애물 거리 간의 대소 관계에 의거한 복수 장애물의 위치 불확실성 영역 판정 기준을 제시한다. 제시된 위치 불확실성 영역판정 기준을 이용하여 다양한 형태의 장애물로 구성된 복합 장애물 환경에서의 장애물 윤곽 추출 결과를 보인다.

• 대한핵의학회지
• /
• 제37권1호
• /
• pp.24-33
• /
• 2003

Finding epileptogenic zone is the most important step for the successful epilepsy surgery. F-18 fluorodeoxyglucose positron emission tomography (FDG-PET) and single photon emission computed tomography (SPECT) can be used in the localization of epileptogenic foci. In medial temporal lobe epilepsy, the diagnostic sensitivity of FDG-PET and ictal SPECT is excellent. However, detection of hippocampal sclerosis by MRI is so certain that use of FDG-PET and ictal SPECT in medial temporal lobe epilepsy is limited for some occasions. In neocortical epilepsy, the sensitivities of FDG-PET or ictal SPECT are fair. However, FDG-PET and ictal SPECT can have a crucial role in the localization of epileptogenic foci for non-lesional neocortical epilepsy. Interpretation of FDG-PET has been recently advanced by voxel-based analysis and automatic volume of interest analysis based on a population template. Both analytical methods can aid the objective diagnosis of epileptogenic foci. Ictal SPECT was analyzed using subtraction methods and voxel-based analysis. Rapidity of injection of tracers, ictal EEG findings during injection of tracer, and repeated ictal SPECT were important technical issues of ictal SPECT. SPECT can also be used in the evaluation of validity of Wada test.

• Ocean Systems Engineering
• /
• 제5권3호
• /
• pp.181-198
• /
• 2015

An OSV (Offshore Support Vessel) is being used to install a structure which is laid on its deck or an adjacent transport barge by lifting off the structure with its own crane, lifting in the air, crossing splash zone, deeply submerging, and lastly landing it. There are some major considerations during these operations. Especially, when lifting off the structure, if operating conditions such as ocean environmental loads and lifting off velocity are not suitable, the collision can be occurred due to the relative motion between the structure and the OSV or the transport barge. To solve this problem, this study performs the physics-based simulation of the lifting off step while the OSV installs the structure. The simulation includes the calculation of dynamic responses of the OSV and the structure, including the collision detection between the transport barge and the structure. To check the applicability of the physics-based simulation, it is applied to a problem of the lifting off step by varying the ocean environmental loads and the lifting off velocity. As a result, it is confirmed that the operability of the lifting off step are affected by the conditions.

• 국제학술발표논문집
• /
• The 9th International Conference on Construction Engineering and Project Management
• /
• pp.912-919
• /
• 2022

The study is an image-based assessment that uses image processing techniques to determine the condition of concrete with surface cracks. The preparations of the dataset include resizing and image filtering to ensure statistical homogeneity and noise reduction. The image dataset is then segmented, making it more suited for extracting important features and easier to evaluate. The image is transformed into grayscale which removes the hue and saturation but retains the luminance. To create a clean edge map, the edge detection process is utilized to extract the major edge features of the image. The Otsu method is used to minimize intraclass variation between black and white pixels. Additionally, the median filter was employed to reduce noise while keeping the borders of the image. Image processing techniques are used to enhance the significant features of the concrete image, especially the defects. In this study, the tonal zones of the histogram and its properties are used to analyze the condition of the concrete. By examining the histogram, the viewer will be able to determine the information on the image through the number of pixels associated and each tonal characteristic on a graph. The features of the five tonal zones of the histogram which implies the qualities of the concrete image may be evaluated based on the quality of the contrast, brightness, highlights, shadow spikes, or the condition of the shadow region that corresponds to the foreground.

• Parasites, Hosts and Diseases
• /
• 제54권3호
• /
• pp.375-380
• /
• 2016

Angiostrongyliasis is difficult to be diagnosed for the reason that no ideal method can be used. Serologic tests require specific equipment and are not always available in poverty-stricken zone and are time-consuming. A lateral flow immunoassay (LFIA) may be useful for angiostrongyliasis control. We established a LFIA for the diagnosis of angiostrongyliasis based on 2 monoclonal antibodies (mAbs) against antigens of Angiostrongylus cantonensis adults. The sensitivity and specificity were 91.1% and 100% in LFIA, while those of commercial ELISA kit was 97.8% and 86.3%, respectively. Youden index was 0.91 in LFIA and 0.84 in commercial ELISA kit. LFIA showed detection limit of 1 ng/ml of A. cantonensis ES antigens. This LFIA was simple, rapid, highly sensitive and specific, which opened an alternative approach for the diagnosis of human angiostrongyliasis.

• Food Science and Biotechnology
• /
• 제16권1호
• /
• pp.99-103
• /
• 2007

The purpose of this study was to develop a simple detection method, possibly at the species-level, that allows for large-scale screening of bifidobacteria. Human fecal samples were plated on MRS-raffinose agar containing cysteine and neomycin sulfate, serving as selective pressure for bifidobacteria, and 0.003%(w/v) bromocresol green. All of the test strains grew well on this medium at $37{\pm}1^{\circ}C$, forming white colonies surrounded by yellow halos, which presented a sharp contrast against the green background. In this disc assay, the required incubation time to develop a yellowish zone varied with the species of Bifidobacterium that was tested, allowing for differential counts and easy identification at the species-level: 10-14 hr for B. bifidum, 20-22 hr for B. catenulatum and B. infantis. and 24-25 hr for B. longum and B. breve. No apparent color was observed for B. angulatum and B. adolescentis 28 hr after inoculation. To evaluate the results of pH indicator-based identification, individual isolates were subjected to a colony-PCR experiment with genus-specific primers. The amplified products from the isolates were in good accordance with those from the reference strains at a level of 95% agreement. These results suggest that the present method could be conveniently applied to cell counts, as well as to the preliminary identification of bifidobacteria from a variety of sample types including human feces, dairy products, and commercial probiotic supplements.

• Journal of Multimedia Information System
• /
• 제6권2호
• /
• pp.91-98
• /
• 2019

Many researchers have attempted to recognize three-dimensional faces using feature points extracted from two-dimensional facial photographs. However, due to the limit of flat photographs, it is very difficult to recognize faces rotated more than 15 degrees from original feature points extracted from the photographs. As such, it is difficult to create an algorithm to recognize faces in multiple angles. In this paper, it is proposed a new algorithm to recognize three-dimensional face recognition based on feature points extracted from a flat photograph. This method divides into six feature point vector zones on the face. Then, the vector value is compressed and expanded according to the rotation angle of the face to recognize the feature points of the face in a three-dimensional form. For this purpose, the average of the compressibility and the expansion rate of the face data of 100 persons by angle and face zone were obtained, and the face angle was estimated by calculating the distance between the middle of the forehead and the tail of the eye. As a result, very improved recognition performance was obtained at 30 degrees of rotated face angle.

• 한국콘텐츠학회논문지
• /
• 제16권7호
• /
• pp.689-699
• /
• 2016

LCD 수요 증가에 따라 LCD 생산 효율성 개선을 위한 검사장비의 중요성이 지속적으로 부각되고 있다. 패턴 검사기는 라인 스캔 카메라와 같은 광학 장비를 통해 미세한 패턴 결함을 빠른 속도로 검출하는 장비이다. 이러한 패턴 검사기는 실시간 검사를 위해 패턴 내에서 단일 기준값을 사용하여 픽셀 단위의 결함 여부를 판단하고 있다. 하지만 패턴 내 각 영역별 특징을 반영하여 서로 다른 기준값을 적용하는 적응적 이진화를 이용하는 경우 결함 검출 성능을 크게 향상시킬 수 있다. 이러한 적응적 이진화를 적용하기 위해서는 특정 검사 대상 픽셀이 어떠한 영역에 속하는지에 대한 정보를 필요로 한다. 이를 위해 본 논문에서는 각각의 검사 대상 픽셀이 어떠한 영역에 속하는지를 판단하는 영역 매칭 알고리즘을 제안한다. 제안된 알고리즘은 머신 비전의 실시간성을 고려한 패턴 정합에 기반을 둔 알고리즘으로 실제 시스템에 적용될 수 있도록 GPGPU를 이용하여 구현된다. 모의실험을 통해 제안된 방법이 실제 시스템이 요구하는 처리 속도를 만족시킬 수 있을 뿐만 아니라 결함 검출의 성능을 개선할 수 있음을 보인다.