• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.3
• /
• pp.181-189
• /
• 2004

This paper describes the method for detecting vehicles in the rear and rear-side at night by using headlight features. A headlight is the outstanding feature that can be used to discriminate a vehicle from a dark background. In the segmentation process, a night image is transformed to a binary image that consists of black background and white regions by gray-level thresholding, and noise in the binary image is eliminated by a morphological operation. In the feature extraction process, the geometric features and moment invariant features of a headlight are defined, and they are measured in each segmented region. Regions that are not appropriate to a headlight are filtered by using geometric feature measurement. In region classification, a pair of headlights is detected by using relational features based on the symmetry of a pair of headlights. Experimental results show that this method is very applicable to an approaching vehicle detection system at nighttime.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.7
• /
• pp.1556-1567
• /
• 2009

The death rate of pedestrian in car accidents in Korea is 2.5 times higher than the average of OECD countries'. If a system that can detect pedestrians and send alarm to drivers is built and reduces the rate, it is worth developing such a pedestrian detection system (PDS). Since the accident rate in which pedestrians are involved is higher at nighttime than in daytime, the adoption of nighttime PDS is being standardized by big auto companies. However, they are usually using night visions or multiple sensors, which are usually expensive. In this paper we suggest a method for nighttime PDS using single wide dynamic range (WDR) monochrome camera in visible spectrum band. In our experiments, pedestrians were accurately detected if only most edges of pedestrians could be obtained.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.9
• /
• pp.2280-2289
• /
• 2009

The death rate of pedestrians in car accidents in Korea is about 2.5 times higher than the average of OECD countries'. If a system that can detect pedestrians and send alarm to driver is built and reduces the rate, it is worth developing such a pedestrian detection system (PDS). Since the accident rate in which pedestrians are involved is higher at nighttime than in daytime, the adoption of nighttime PDS is being standardized by big auto companies. However, they are usually using expensive night visions or multiple sensors for their PDS. In this paper we propose a method for nighttime PDS using a monochrome visible spectrum camera. We could verify its superiority in both performance and real?time operation to existing algorithm through tests against video data taken in several different environments.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.48-48
• /
• 2017

도시지역에서는 이상강우, 돌발홍수와 더불어 급속한 도시화에 따라 침수 발생의 위험성이 증가하고 있다. 지자체에서는 빗물펌프장, 지하저류조 등을 이용하여 적극적으로 침수대책을 강구하고 있지만, 저지대 침수피해는 계속적으로 발생하고 있다. 2013년 7월 서울 경기북부 강원영서에 발생한 집중호우로 1명이 사망하고 피해액은 94,036백만원이 발생하였다. 춘천시 효자동 저지대지역의 주택침수는 인근 하천의 수위가 높아져 내수배제 및 하수도 처리 능력이 부족하여 침수가 발생하였다. 2014년 8월 경남지역에 발생한 집중호우로 2명 사망하였고, 피해액은 134,158백만원에 이르렀으며, 도시화 토지피복변화로 인한 홍수량이 증가하여 피해가 가중되었다. 이에 따라 도시 내 정확한 도시유출 및 침수해석을 통하여 과거 침수양상을 재현하고 앞으로 발생할 수 있는 침수피해를 방지할 수 있도록 침수 예 경보 시스템을 개발하여 도시침수에 대비하고 도시주민들의 인명과 재산피해를 경감하는데 그 목적이 있다. 이를 위해 도시지역의 관망해석에 적합한 모형을 선택하여야 하며, 그 모형의 매개변수를 결정하여야 한다. 도시유출해석 및 관망해석을 위하여 SWMM 모형을 선정해서 유역분할조건, 매개변수에 대한 최적 검정과정을 제시하여, 1D-2D 연계모형을 통해서 침수지역예측의 정확도를 증대시키고자 한다. 본 연구에서는 서울시 강남역 주변 일대의 5개 배수분구을 대상으로 도시유출해석을 위한 최적화 모의를 위하여 1D-2D모형의 연계해석하였다. 실제 강우사상을 적용하여 매개변수와 소유역 개수를 달리하여 자동최적화기법인 PEST를 이용하여 최적인자를 도출하여, 실제 배수맨홀의 수위관측 자료를 이용하여 비교 보정을 하였다. 도시유출해석뿐만 아니라 내수침수시의 최적인자 도출을 위해 2차원 범람해석을 통하여 NDMS 자료를 이용하여 비교 보정을 한 뒤 다른 강우사상을 이용하여 검증을 하여 도시침수 해석의 정확도를 개선을 위한 최적인자들을 검토하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.136-136
• /
• 2018

최근 기후변화로 인해 설계 빈도 이상의 강우가 빈번하게 발생하고 있으며 이러한 집중호우에 의한 피해는 지역별로 토지이용, 사회 환경적인 조건에 따라 차이가 있다. 특히 인구와 경제 제반 시설이 밀집되어 있는 도시지역의 경우 홍수가 발생하게 되면 피해가 과거에 비해 큰 규모로 나타나고 있다. 현재 각 지자체는 홍수 대응을 위해 기상청에서 제공하고 있는 6시간, 12시간 강우량을 기준으로한 호우주의보 및 경보를 사용하고 있다. 하지만 도시지역의 경우 지속시간이 1~2시간인 짧은 호우에서 피해가 발생하여 기상청에서 제공하고 있는 홍수특보를 활용하는데 있어 한계가 있다. 또한 이러한 기준은 각 지역별 사회 경제적 특성을 반영하지 않아 지역별 실질적인 홍수 대응에 효과적이지 않다. 따라서 각 지역별로 특성을 파악하여 홍수에 대한 피해를 저감하기 위한 대책 수립이 필요한 실정이다. 본 연구의 공간적 범위로 한강권역에서 서울시와 인천시의 경우 '구 단위', 경기도와 강원도, 충청북도는 '시 군 단위'로 총 88개 행정구역으로 하고 도시별 특성에 따라 군집화를 위해 환경적, 사회적 및 경제적 특성 중 홍수에 대한 취약성을 반영할 수 있는 인자를 선정하였다. 선정된 인자들을 주성분분석을 통해 공통된 특성이나 유사한 성질에 따라 공통인자를 추출하고 이를 이용하여 각 도시별 군집분석을 시행하였다. 그 후 각 도시 군집별 특성에 따라 홍수에 의한 피해를 가장 잘 나타낼 수 있는 인자를 위험영향(Risk Impact)으로 산정하였다. 이때 위험영향은 강우규모에 따른 잠재 피해 정도를 나타내며 홍수위험도표를 작성하기 위해서는 이러한 위험영향을 정량적으로 설명할 수 있어야 한다. 본 연구에서는 각 도시 군집별 특성에 따른 위험영향의 임계값을 추정하고 그에 따른 홍수위험도표를 작성하였다.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2023.05a
• /
• pp.204-205
• /
• 2023

This study studied a prediction model for sending collision alarms for small fishing boats based on semi-supervised learning(SSL). The supervised learning (SL) method requires a large number of labeled data, but the labeling process takes a lot of resources and time. This study used service data collected through a data pipeline linked to 'intelligent maritime traffic information service' and data collected from real-sea experiment. The model accuracy was improved as a result of learning not only real-sea experiment data with labeling determined based on actual user satisfaction but also service data without label determined together.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.67-77
• /
• 2023

This study represents an innovative research conducted in the smart farm environment, developing a deep learning-based disease and pest detection model and applying it to the Intelligent Internet of Things (IoT) platform to explore new possibilities in the implementation of digital agricultural environments. The core of the research was the integration of the latest ImageNet models such as Pseudo-Labeling, RegNet, EfficientNet, and preprocessing methods to detect various diseases and pests in complex agricultural environments with high accuracy. To this end, ensemble learning techniques were applied to maximize the accuracy and stability of the model, and the model was evaluated using various performance indicators such as mean Average Precision (mAP), precision, recall, accuracy, and box loss. Additionally, the SHAP framework was utilized to gain a deeper understanding of the model's prediction criteria, making the decision-making process more transparent. This analysis provided significant insights into how the model considers various variables to detect diseases and pests.

• Journal of Industrial Convergence
• /
• v.22 no.2
• /
• pp.23-32
• /
• 2024

This study utilized minimum number of demographic variables and first-semester GPA of students to predict the final academic status of students at a vocational college in Seoul. The results from XGBoost and LightGBM models revealed that these variables significantly impacted the prediction of students' dismissal. This suggests that early academic performance could be an important indicator of potential academic dropout. Additionally, the possibility that academic years required to award an associate degree at the vocational college could influence the final academic status was confirmed, indicating that the duration of study is a crucial factor in students' decisions to discontinue their studies. The study attempted to model without relying on psychological, social, or economic factors, focusing solely on academic achievement. This is expected to aid in the development of an early warning system for preventing academic dropout in the future.

• Korean journal of communication and information
• /
• v.43
• /
• pp.247-281
• /
• 2008

The purpose of this article is to re-examine the issue of Chobo's origin, which has not been clearly determinated thus far. Also, the research investigates Chobo's name, time point of discontinuance, character of the article on disaster, and civil-printed incident of Chobo during King Sunjo. As an internal bulletin, handwritten Chobo began to come out at the end of 15th century, roughly King Sungjong period of Chosun Dynasty. Systematically developed in the period of King Jungjong, it had continuously been released until November, 1907 when fairly organized and competitive commercial daily newspapers were published and when the office of secretary named Biseogam was abolished. Because handwritten Chobo was exclusive and narrow in its communication nature and difficult to read it's fast handwritten Nancho calligraphic style, in August, 1577 (10th year of King Sunjo) a few civilian in Seoul obtained a permission from authorities (Uijungboo and Sahunboo) published a wooden type printed Chobo. Unfortunately, privately printed Chobo was forced to cease in 3 month because of King Sunjo's oppressive measure. However, considering it was published for a profit by civilian and used the world's first type printing, it is highly probable that Chobo seems to be the world’s first type printing commercial daily newspaper.

• Research in Plant Disease
• /
• v.25 no.4
• /
• pp.196-204
• /
• 2019

Rice panicle blast occurred severely in southern provinces of Korea in 2014. The proportion of panicle blast incidence area to cultivated area of rice were 11.0% and 14.6% in Jeollanam-do and Gyeongsangnam-do, respectively. To identify the causal factors of the outbreak, we investigated weather conditions in August, amount of cultivated area of mainly grown cultivars, and nitrogen contents in plants with different disease incidences in 2014. 'Saenuri,' 'Ilmibyeo,' 'Unkwang,' 'Dongjin 1 ho,' 'Nampyeongbyeo,' and 'Hwangkeumnuri' were mainly grown cultivars. Monthly average of daily air temperature in August 2014 was 3.2℃ and 3.1℃ less than 2018 in Haenam and Miryang, respectively. Rainfall in August 2014 was 70.0% and 42.0% greater than 2018 in Haenam and Miryang, respectively. The numbers of blast warning days in August calculated nationwide using a forecast model for blast infection were higher in 2014 than in 2018, and they were in high level throughout the country in 2014. Nitrogen contents in plant samples from high-incidence plots were significantly higher than those from low-incidence plots. Consequently, excessive use of nitrogen fertilizers was the main factor for the disease outbreak at the level of specific farms, in addition to the collective cultivation of susceptible cultivar, low temperatures and frequent rainfalls in August.