• 디지털산업정보학회논문지
• /
• 제18권4호
• /
• pp.1-11
• /
• 2022

Recently, research on license plate recognition, which is a core technology of an intelligent transportation system(ITS), is being actively conducted. In this paper, we propose a method to extract numbers and characters from low-quality license plate images by applying the YOLOv4 algorithm. YOLOv4 is a one-stage object detection method using convolution neural network including BACKBONE, NECK, and HEAD parts. It is a method of detecting objects in real time rather than the previous two-stage object detection method such as the faster R-CNN. In this paper, we studied a method to directly extract number and character regions from low-quality license plate images without additional edge detection and image segmentation processes. In order to evaluate the performance of the proposed method we experimented with 500 license plate images. In this experiment, 350 images were used for training and the remaining 150 images were used for the testing process. Computer simulations show that the mean average precision of detecting number and character regions on vehicle license plates was about 93.8%.

• 한국멀티미디어학회논문지
• /
• 제8권4호
• /
• pp.473-481
• /
• 2005

본 논문은 적응적 양자화 컬러 수와 적응적 병합 임계값을 이용하여 순차적으로 영역을 병합하여 영역의 경계를 보존하며 영상을 분할하는 방법을 제안한다. 제안방법은 먼저 PSNR을 이용하여 영상에 따라 다른 양자화 컬러 수로 영상을 벡터 양자화 한다. 그리고 양자화 영상을 이용하여 초기 영역을 설정한 후 CIE Lab와 RGB 컬러 공간에서 순차적으로 유사한 영역을 병합하여 영상의 주요 영역들로 분할한다. 병합의 각 단계에서는 유사성의 척도로 인접 영역의 컬러 거리를 사용하며 병합 임계값은 분할된 영역과 원영상의 컬러 거리의 평균과 평균 변화량을 이용하여 적응적으로 구하였다. 또한 RGB 컬러 공간에서의 병합 영상이 주요 영역 단위로 병합되지 않은 경우 후처리로서 CIE Lab 영역에서 다시 한번 병합을 수행한다. 이때 초기 영역 영상과 RGB 컬러 공간에서의 병합 영상의 영역간의 컬러 거리를 이용하여 병합 유무를 결정한다. 실험 결과는 제안방법에 의한 결과 영상이 주요 객체를 중심으로 분할되며 객체의 경계가 잘 보존됨을 보여준다. 또한 객관적인 척도에서도 기존의 방법에 비해 좋은 결과를 보여준다.

• 대한전자공학회논문지SP
• /
• 제43권5호
• /
• pp.22-29
• /
• 2006

본 논문에서는 블록 기반 실시간 계수 시스템을 제안한다. 계수 시스템은 쇼핑몰이나 대형건물의 출입구, 엘리베이터, 에스컬레이터 등과 같은 다양한 환경에 적용될 수 있다. 본 논문의 핵심은 세 부분으로 구분 할 수 있다. 첫째, 환경 변화에 강인한 배경 제거 기법, 둘째, 학습을 이용한 블록 기반 계수 결정 방법, 마지막으로 4 채널에서 실시간으로 처리가 가능한 시스템 구현이다. 환경 변화에 강인한 배경 제거 기법으로 MOG(mixture of gaussian) 방법을 적용하였으며, 블록 기반 계수 결정 방법은 영상을 $6{\times}12$개의 영역으로 구분하고, 학습 영상을 이용하여 각각의 블록에서 물체의 크기에 대한 평균과 분산을 계산하고 이를 저장하여 계수에 적용하는 방법을 제안하였으며, 마지막으로 4 채널에서 실시간으로 처리 하기 위하여 추출된 물체 각각을 구분하여 추적하는 것이 아니라 물체가 이동 중에 발생할 수 있는 여러 상황 즉 결합이나 분리되는 현상을 분석하여 계수 판단에 적용하였다. 제안된 방법을 이용하여 에스컬레이터, 복도 그리고 출입문등과 같이 조명환경과 그림자 상황 등이 변하는 환경에서 의 계수 정확도는 $90{\sim}94%$, 채널 당 처리 속도는 15 frames/sec 이상의 성능을 보였다.

• 한국학교ㆍ지역보건교육학회지
• /
• 제19권2호
• /
• pp.43-52
• /
• 2018

Objects: In this study, we analyzed the correlation between stress and depression related to the quality of life of one elderly household, and analyzed the influence of stress and depression on the quality of life. In this study, I tried to make a plan for improving the quality of life for one elderly. Methods: In this study, the frequency and percentage according to the general characteristics of one elderly household were frequency analyzed. And the mean and standard deviation of stress, depression, quality of life were analyzed on average. And continuous analysis of stress, depression, quality of life, etc., conducted a technical analysis. Relationship to stress, depression, quality of life, correlation analysis was done. The effect of stress and depression on the quality of life was linear regression analysis. SPSS Version 23.0 was used for analysis. Result: Increased stress and increased depression in elderly single households was a static (+) correlation. And the results of the regression analysis showed that the higher the stress and depression of one elderly household, the lower the quality of life. Conclusion: For the improvement of the quality of life of one elderly household, the government should support stress and depression prevention programs according to sex, age, number of chronic diseases.

• 기본간호학회지
• /
• 제12권2호
• /
• pp.215-222
• /
• 2005

Purpose: To identify the incidence of cutaneous injury in clinical nurses. Method: From Feb.1 to 28, 2005, 276 clinical nurses were surveyed by questionnaire. Results: 1. Of the nurses, 53.6% had at least one incidence of cutaneous injury, and the mean number of injuries was 1.34. A higher incidence rate for cutaneous injury was found in nurses who were under the age of 25, unmarried and who had less than 3 years career experience. 2. The major causes of injury were syringe needles at 65.0%, and medical instrument were next followed by sharp objects or blades. The injuries occurred when the nurses were rearranging equipment after care (25.2%), taking blood samples (22.8%), separating syringes and needles (17.1%), during surgical operations (14.2%), and distribution of medications, treatments and recapping of needles (5.7% each). The hands were the most common body parts injured, and the most prevalent pathogens contaminating the instruments causing the injury were HBV syphilis, HCV and HIV in that order. 3. Of the injured nurses, 77.9% did not report the accident and 25.8% did not receive any treatment because there were no pathogens, it was a bother or there was difficulty reporting the incident. Conclusion: To reduce cutaneous injuries, intensive training and supervision may be needed for those of nurses under the age of 25, unmarried and with less than 3 years career experience.

• 천문학회보
• /
• 제40권2호
• /
• pp.34.1-34.1
• /
• 2015

Deep archival V and I image data taken with Hubble Space Telescope have been used to investigate compact stellar objects in an anemic spiral galaxy NGC 4921 in the Coma cluster. We resolve a significant fraction of globular clusters based on the reconstructed master drizzled image data. The color distribution of globular clusters (GCs) shows a clear bimodal distribution. The blue and red GC populations show significantly different radial number density profiles. We derive the turnover magnitudes of globular cluster luminosity functions (GCLFs) for the blue and red GCs in the bulge and halo of NGC 4921. We also derive the GCLFs of two Coma cD galaxies, NGC 4874 and NGC 4889, and one coma S0 galaxy, NGC 4923. Turnover magnitudes of GCs in four galaxies agree well within uncertainties. A mean distance of four Coma galaxies is derived from turnover magnitudes of GCLFs. A value of the Hubble constant is determined from this distance estimate and radial velocity of the Coma. We discuss implications of our results in relation with the recent determinations of the Hubble constant.

• 한국전문물리치료학회지
• /
• 제28권2호
• /
• pp.123-131
• /
• 2021

Background: Only 2% of falls in older adults result in serious injuries (i.e., hip fracture). Therefore, it is important to differentiate injurious versus non-injurious falls, which is critical to develop effective interventions for injury prevention. Objects: The purpose of this study was to a. extract the best features of surface electromyography (sEMG) for classification of injurious falls, and b. find a best model provided by data mining techniques using the extracted features. Methods: Twenty young adults self-initiated falls and landed sideways. Falling trials were consisted of three initial fall directions (forward, sideways, or backward) and three knee positions at the time of hip impact (the impacting-side knee contacted the other knee ("knee together") or the mat ("knee on mat"), or neither the other knee nor the mat was contacted by the impacting-side knee ("free knee"). Falls involved "backward initial fall direction" or "free knee" were defined as "injurious falls" as suggested from previous studies. Nine features were extracted from sEMG signals of four hip muscles during a fall, including integral of absolute value (IAV), Wilson amplitude (WAMP), zero crossing (ZC), number of turns (NT), mean of amplitude (MA), root mean square (RMS), average amplitude change (AAC), difference absolute standard deviation value (DASDV). The decision tree and support vector machine (SVM) were used to classify the injurious falls. Results: For the initial fall direction, accuracy of the best model (SVM with a DASDV) was 48%. For the knee position, accuracy of the best model (SVM with an AAC) was 49%. Furthermore, there was no model that has sensitivity and specificity of 80% or greater. Conclusion: Our results suggest that the classification model built upon the sEMG features of the four hip muscles are not effective to classify injurious falls. Future studies should consider other data mining techniques with different muscles.

• 대한간호학회지
• /
• 제30권2호
• /
• pp.259-271
• /
• 2000

This study was attempted to explore meaning in life and it's relationship to problematic life events, health status and depression among older people. 198samples of elderly were over the age 60 (mean age=70.06) conveniently selected in Seoul, Korea. The data were collected through questionaires by interview with subjects from April, 1, 1999 to May 30, 1999. The research instuments were the main framework in the experiment. They involved meaning fulfillment and problematic life events which were developed by Burbank (1988) and short form geriatric depression scale(Sheikh & Yesavage, 1986). The data were analysed by SAS Program. The results are followed; 1. 172 lists were given out and had objects that would bring meaning to one's life. They were categrized as wishing for offsprings' prosperity, their relation with their grand-children, marriage of their offspring, health of family members, strong family ties, caring for the family, relation with husband, health for self, religious activity, helping others, individual hobbies, needs for growth such as writing, education, relationships with friends and personal achievements. In the end, 50% of total list dealt with in the relationship with their family members. 2. The mean fulfillment of the subjects was 38.31(SD=13.58) with a range of 12-60. The degree fulfillment varied according to the subjects' characteristics such as age group(F=4.44, 0P=.0008), education status (F=6.44, P=.0001), economic status (F=10.27, P=.0001), marrital status (F=9.12, P=.0002) and religious background (F=3.68, P=.006). 3. According to the pearson correlation, analysis significant variables were found between meaning fulfillment and health(r=.47, P=.0001), depression and the number of problematic life events for a life time (r=.147, P=.04), the stress score of problamatic life events for the past one year and the stress score of problematic life events for a life time was r=.43, P=.002, meaning fulfillment and the stress score of problamatic life events r=-.26, P=.04, depression and meaning fulfillment r=-.70, P=.0001 and depression and health r=-.521, P=.0001. 4. According to multiful regression analysis, depression was found as a most predictable variable for health(F=12.51, P=.001).

• 전자공학회논문지SC
• /
• 제49권1호
• /
• pp.94-102
• /
• 2012

최근 고령 사회에 들어섬에 따라 고령 운전자의 수가 증가하는 추세이다. 고령 운전자의 교통사고 대부분이 차량 운전자의 부주의로 인해 발생한다. 이러한 부주의들에는 노화에 따른 느린 몸의 움직임으로 차량 조작 미숙, 노안으로 인한 좁은 시야로 낮은 시각정보 검색 문제 그리고 낮은 대비감도로 인한 물체 식별 문제 등으로 기인한다. 본 연구에서는 고령 운전자의 안전 운전 지원을 위해 도로 영상에서 시각적 주의를 가져야 하는 관심물체 영역들을 실시간으로 자동 검출하는 방법을 제안한다. 제안한 방법은 입력 영상으로부터 선택적 시각 주의를 갖는 관심물체후보 영역들을 실시간으로 검출하기 위해 칼라, 기울기, 그리고 밝기 특징정보들의 대비 변화 정도를 3차원으로 표현한 현저함 맵(Saliency map)을 생성하고, 동시에 입력 영상으로부터 물체들의 경계선 획득을 위해 mean-shift 알고리즘을 적용하여 영상을 분할한다. 그리고 분할된 영역에 속한 현저함 픽셀의 유무에 따른 선택적 시각 주의 영역을 검출한다. 제안한 방법을 다양한 실외 환경 조건에서 실험한 결과, 도로 상의 다양한 물체에 빠른 검출율과 함께 비교적 복잡한 도로 환경에서도 강임함을 알 수 있다.

• 대한공업교육학회지
• /
• 제33권2호
• /
• pp.287-299
• /
• 2008

HTTP는 월드 와이드 웹에서 가장 널리 사용되는 프로토콜의 하나로 신뢰성을 제공하기 위해 전송 계층 프로토콜로 TCP를 사용한다. HTTP는 개별적인 파일 요청에 대해 분리된 TCP 연결을 사용하기 때문에 파일 수신에 있어서 불필요한 오버헤드인 헤드-오브-라인 (head-of-line) 블로킹을 유발시킨다. 웹 응용은 일반적으로 전송되는 크기가 작기 때문에 무선 환경에서는 TCP로 인한 핸드오버의 지연이 증가한다. 이에 비해 최근에 제안된 SCTP(stream control transmission protocol)는 멀티-스트리밍과 멀티-호밍과 같은 매력적인 기능을 갖고 있다. SCTP의 이러한 기능들은 TCP의 헤드-오브-라인 블로킹을 제거하고, 무선 환경에서 TCP의 핸드오버 지연을 줄이는 것으로 기대되고 있다. 평균 응답 시간은 대부분의 웹 응용에 있어서 중요한 측정 요소이다. 본 논문에서는 NS-2 시뮬레이터를 이용하여 유무선 인터넷 환경에서 SCTP와 TCP의 평균 응답 시간을 비교하였다. 이를 위해, 유선 환경에서는 패킷 손실률, 대역폭, RTT(Round Trip Time) 및 웹 객체의 개수의 변화에 따른 평균 응답 시간이 비교되었고, 무선 환경에서는 이동속도 및 반경의 크기에 따른 평균 응답 시간과 패킷 손실률이 비교되었다. 시뮬레이션 결과는 SCTP가 TCP의 평균 응답 시간을 감소시키는 것으로 나타났다.