• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.91-94
• /
• 2002

음성인식 시스템의 실용화를 위해서 우선적으로 해결되어야 될 문제중 하나로 잡음환경하에서의 끝점검출을 들 수 있다. 잡음이 존재하지 않는 환경에서는 기존의 에너지 파라미터만으로도 어느정도 신뢰성있는 끝점 구간을 검출할 수 있으나 도심 소음과 같은 실제 잡음환경하에서는 대부분 좋지 않은 결과를 보인다. 본 논문에서는 도심환경의 배경잡음을 제거하는 방법으로 입력되는 음성에 대하여 주변소음에 의해 손상된 음성스펙트럼의 크기 성분만을 제거하는 전처리 기법인 Bark scale에 기반한 스펙트럼 차감법을 사용하고, 인간의 청각특성을 고려하여 음성의 주파수 대역을 3개의 대역으로 분리한 후, 대역별로 세밀한 에너지 문턱치값을 설정하여 음성의 끝점을 탐색하는 방법을 제안한다. 제안한 방법의 유효성을 확인하기 위해 실제 사무실 및 지하철역 등의 잡음환경하에서 녹음된 데이터베이스를 이용하여 끝점검출을 수행한 결과 기존의 에너지와 영교차율을 이용한 방법에 비해 평균 $46\%$의 오차율 감소와 대역에너지만을 사용한 경우에 비해 평균 $17\%$의 오차율 감소를 나타내어 제안한 방법의 유효성을 확인할 수 있었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.04a
• /
• pp.478-481
• /
• 2010

컴퓨터 비전에서 객체의 인식, 추적에 앞서 배경으로부터 전경을 분리하는 배경차감 기법과 분리된 전경에 대한 관심 영역(ROI)을 추출하는 것은 일반적인 방법이다. 하지만 전경을 정확히 분리하지 못하면 개별 객체의 관심영역(ROI) 역시 잘못 추출되는 문제가 발생된다. 본 논문에서는 정확하지 않은 전경 분리로 부터 발생되는 개별 객체에 대한 분산된 관심영역을 병합하는 방법을 제안한다. 본 방법은 배경과 분리된 전경에서 한 객체의 일정 거리 이내에 있는 다른 객체를 가상으로 병합하는 단계, 워터쉐드 분할 알고리즘을 적용하는 단계를 거쳐 다시 블럽 레이블링을 수행한다. 제안 방법을 통하여 배경 모델에서 분리된 개별 객체의 병합된 관심영역을 제공한다. 실험에서 기존의 일반적인 블럽 레이블링 방법만을 적용하여 추출한 전경영역과 제안하는 방법에 의한 전경영역을 비교하여 배경 모델에서 분리된 개별 객체의 관심영역이 효과적으로 추출되는 것을 보인다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.2
• /
• pp.173-181
• /
• 2014

In this paper, the algorithm that can automatically segment the lung, the airway and the pulmonary vessels in a chest CT was proposed. The proposed method is progressed in three steps. In the first step, the lung and the airway are segmented by the region growing law through the optimal threshold and three-dimensional labeling. In the second, from the start point to the first carina of the airway is segmented by the deduction operation, and the next airway of the bifurcations are segmented by applying a variable threshold technique. In the third step, the left/right lungs are divided by the restoration process for the lung, and the outside of lungs for abnormal is checked by applying the advanced rolling ball algorithm, and if abnormal is found, that part is removed, and it is restored to the normal lungs by connecting the outside of the lung in the form of second-order polynomial. Finally, pulmonary vessels are segmented by applying the three-dimensional connected component labeling method and three-dimensional region growing method. As the results of simulation, it could be confirmed that the pulmonary vascular is accurately divided without loss of tissue around lung.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.7
• /
• pp.1531-1539
• /
• 2014

Premature ventricular contraction(PVC) is the most common disease among arrhythmia and it may cause serious situations such as ventricular fibrillation and ventricular tachycardia. Nevertheless personalized difference of ECG signal exist, performance degradation occurs because of carrying out diagnosis by general classification rule. In other words, the design of algorithm that exactly detects abnormal signal and classifies PVC by analyzing the persons's physical condition and/or environment and variable QRS pattern is needed. Thus, PVC classification by personalized abnormal signal detection and QRS pattern variability is presented in this paper. For this purpose, we detected R wave through the preprocessing method and subtractive operation method and selected abnormal signal sets. Also, we classified PVC in realtime through QS interval and R wave amplitude. The performance of abnormal beat detection and PVC classification is evaluated by using MIT-BIH arrhythmia database. The achieved scores indicate the average of 98.33% in abnormal beat classification error and 94.46% in PVC classification.

• The KIPS Transactions:PartB
• /
• v.15B no.1
• /
• pp.25-36
• /
• 2008

In this paper, we present a robust hand recognition approach to sudden illumination changes. The proposed approach constructs a background model with respect to hue and hue gradient in HSI color space and extracts a foreground hand region from an input image using the background subtraction method. Eighteen features are defined for a hand pose and multi-class SVM(Support Vector Machine) approach is applied to learn and classify hand poses based on eighteen features. The proposed approach robustly extracts the contour of a hand with variations in illumination by applying the hue gradient into the background subtraction. A hand pose is defined by two Eigen values which are normalized by the size of OBB(Object-Oriented Bounding Box), and sixteen feature values which represent the number of hand contour points included in each subrange of OBB. We compared the RGB-based background subtraction, hue-based background subtraction and the proposed approach with sudden illumination changes and proved the robustness of the proposed approach. In the experiment, we built a hand pose training model from 2,700 sample hand images of six subjects which represent nine numerical numbers from one to nine. Our implementation result shows 92.6% of successful recognition rate for 1,620 hand images with various lighting condition using the training model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.2
• /
• pp.437-444
• /
• 2016

Most methods for detecting arrhythmia require pp interval, diversity of P wave morphology, but it is difficult to detect the p wave signal because of various noise types. Therefore it is necessary to use noise-free R wave. In this paper, we propose algorithm for premature contraction arrhythmia classification through ECG pattern analysis and template threshold. For this purpose, we detected R wave through the preprocessing method using morphological filter, subtractive operation method. Also, we developed algorithm to classify premature contraction wave pattern using weighted average, premature ventricular contraction(PVC) and atrial premature contraction(APC) through template threshold for R wave amplitude. The performance of R wave detection, PVC classification is evaluated by using 6 record of MIT-BIH arrhythmia database that included over 30 PVC and APC. The achieved scores indicate the average of 99.77% in R wave detection and the rate of 94.91%, 95.76% in PVC and APC classification.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.422-422
• /
• 2017

기후변화로 인한 게릴라성 호우 및 국지성 호우의 증가, 급격한 도시화로 인한 불투수면적의 감소 등은 다양한 홍수피해와 물 순환 왜곡문제 등을 유발하고 있다. 서울시의 경우 2010년 9월 21일 시간 최대 75.5mm의 폭우로 광화문광장 및 백운동천 주변상가에서 약 15.7ha의 침수가 발생하였다. 이러한 침수발생에 대응하기 위한 대안으로 서울시에서는 저영향개발(LID, Low Impact Devleopment)기법을 도입하고 있다. 본 연구에서는 효자배수분구(광화문지역)의 LID 시설 도입을 위한 빗물관리 목표량 설정에 따른 홍수방어 개선효과를 분석하였다. 빗물관리량 산정을 위한 유출해석모형은 도시유출모형인 SWMM을 이용하다. SWMM의 입력자료인 우수관망도 자료는 서울시 우수관망 UIS 자료를 이용하여 구성하였으며 소유역분할은 유역의 경사 및 건물 등을 고려하여 정교하게 소유역을 구분하였다. 구분된 소유역은 293개이며 개별 소유역에 대한 토지이용도, 정밀토양도를 검토하여 소유역별 유출곡선지수 산정하여 모형 입력자료를 구축하였다. 효자배수분구(광화문지역) 전체유역 중 전원지역을 제외한 나머지 지역에 대하여 소배수구역별로 유출곡선지수(CN)값을 10-90%까지 감소시키면서 LID시설의 계획에 의한 빗물관리량(mm)에 따른 침수발생량($m^3$)의 변화를 분석하였다. 여기서, LID 시설의 빗물관리량은 유역의 현상태 직접 유출고에서 CN값 변화에 따라 계산된 직접유출고를 차감한 양이다. 연구결과, 효자배수분구의 경우 도시유역 전체에서 20mm의 우수를 관리하면 전체 침수발생량의 약 50%를 30mm의 우수를 관리하면 유역침수 침수발생량의 약 75%를 저감할 수 있는 것으로 분석되었다. 본 연구결과는 향후 개발된 도시지역에 대한 지구단위계획 수립시 LID 시설목표에 따른 홍수저감 효과를 비교적 효율적인 방법으로 검토하는 방안으로 활용될 수 있을 것으로 기대된다.

• Journal of the Korea Computer Graphics Society
• /
• v.16 no.1
• /
• pp.1-7
• /
• 2010

With the widespread use of vision-based surveillance systems, the capability for person identification is now an essential component. However, the CCTV cameras used in surveillance systems tend to produce relatively low-resolution images, making it difficult to use face recognition techniques for person identification. Therefore, an algorithm is proposed for person identification in CCTV camera images based on the clothing. Whenever a person is authenticated at the main entrance of a building, the clothing feature of that person is extracted and added to the database. Using a given image, the clothing area is detected using background subtraction and skin color detection techniques. The clothing feature vector is then composed of textural and color features of the clothing region, where the textural feature is extracted based on a local edge histogram, while the color feature is extracted using octree-based quantization of a color map. When given a query image, the person can then be identified by finding the most similar clothing feature from the database, where the Euclidean distance is used as the similarity measure. Experimental results show an 80% success rate for person identification with the proposed algorithm, and only a 43% success rate when using face recognition.

• The Journal of Korean Institute of Information Technology
• /
• v.15 no.12
• /
• pp.157-164
• /
• 2017

Mobile ECG signal measurement is a technique to measure small signals of several mV, and many studies have been conducted to remove noise including wandering scheme. Removal of the equipotential line noise caused by shaking or movement of the electrode cable is one of the core research contents for the electrocardiogram measurement. In this study, we proposed a modified step-size of combined NLMS(normalized least squares) and DLMS(delayed least squares) adaptive filter to eliminate baseline noise from ECG signals. The proposed method mainly adjusts initial filter step-size to reduce distortion of original ECG signals characteristic after eliminating baseline noise. The modified filter step-size is scaled by filter order size and distortion minimization factor. This method is suitable for portable ECG device with a small processor and less power consumption. This technique also decreases computation time which is essential for real-time filtering. The proposed filter also increase the signal to noise ratio (SNR) compared to conventional NLMS filter.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.8
• /
• pp.383-388
• /
• 2006

This Paper presents a novel approach to single channel microphone speech enhancement in noisy environments. Widely used noise reduction techniques based on the spectral subtraction are generally expressed as a spectral gam depending on the signal-to-noise ratio (SNR). The well-known decision-directed(DD) estimator of Ephraim and Malah efficiently reduces musical noise under the background noise conditions, but generates the delay of the a prioiri SNR because the DD weights the speech spectrum component of the Previous frame in the speech signal. Therefore, the noise suppression gain which is affected by the delay of the a priori SNR, which is estimated by the DD matches the previous frame rather than the current one, so after noise suppression. this degrades the noise reduction performance during speech transient periods. We propose a computationally simple but effective speech enhancement technique based on the sigmoid type function for the weight Parameter of the DD. The proposed approach solves the delay problem about the main parameter, the a priori SNR of the DD while maintaining the benefits of the DD. Performances of the proposed enhancement algorithm are evaluated by ITU-T p.862 Perceptual Evaluation of Speech duality (PESQ). the Mean Opinion Score (MOS) and the speech spectrogram under various noise environments and yields better results compared with the fixed weight parameter of the DD.