• 한국정보통신설비학회:학술대회논문집
• /
• 한국정보통신설비학회 2009년도 정보통신설비 학술대회
• /
• pp.382-386
• /
• 2009

Online or sequential learning is one of the most basic and powerful method to train neuron network, and it has been widely used in disease detection, weather prediction and other realistic classification problem. At present, there are many algorithms in this area, such as MRAN, GAP-RBFN, OS-ELM, SVM and SMC-RBF. Among them, SMC-RBF has the best performance; it has less number of hidden neurons, and best efficiency. However, all the existing algorithms use signal normal distribution as kernel function, which means the output of the kernel function is same at the different direction. In this paper, we use multi-variable normal distribution as kernel function, and derive EKF learning formulas for multi-variable normal distribution kernel function. From the result of the experience, we can deduct that the proposed method has better efficiency performance, and not sensitive to the data sequence.

• 대한임베디드공학회논문지
• /
• 제13권4호
• /
• pp.179-185
• /
• 2018

Particulate Matter (PM2.5) has various adverse effects on health. Climate and industry activity and traffic volume are the main causes, especially in urban area. In order to construct an effective forecasting system, many measurement systems are required, but it is impossible in reality. Therefore, in this study, we propose a method to infer PM2.5 condition by using rule induction technique. The experimental results showed a classification accuracy of 71%.

• International Journal of Advanced Culture Technology
• /
• 제4권4호
• /
• pp.23-29
• /
• 2016

In this paper, the learning technique for CNN processor on vehicle is proposed. In the case of conventional CNN processors, weighted values learned through training are stored for use, but when there is distortion in the image due to the weather conditions, the accuracy is decreased. Therefore, the method of enhancing the input image for classification is general, but it has the weakness of increasing the processor size. To solve this problem, the CNN performance was improved in this paper through the learning method of the distorted image. As a result, the proposed method showed improvement of approximately 38% better accuracy than the conventional method.

• 환경영향평가
• /
• 제5권1호
• /
• pp.95-105
• /
• 1996

Meteorological Joint Frequency Function required indispensably in long-term air quality prediction models were discussed for practical application in Korea. The algorithm, proposed by Turner(l964), is processed with daily solar insolation and cloudiness and height basically using Pasquill's atmospheric stability classification method. In spite of its necessity and applicability, the computer program, called STAR(STability ARray), had some significant difficulties caused from the difference in meteorological data format between that of original U.S. version and Korean's. To cope with the problems, revised STAR program for Korean users were composed of followings; applicability in any site of Korea with regard to local solar angle modification; feasibility with both of data which observed by two classes of weather service centers; and examination on output format associated with prediction models which should be used.

• 한국조경학회지
• /
• 제27권4호
• /
• pp.73-79
• /
• 1999

This study is on the production and the classification of a new appreciation methods of seascape through materials in the words of Korean popular songs. In advance, it is necessary to understand the popular songs as collective representation and the songs are analytic data. In this study, some essential elements of seascape in popular songs are analyzed and classified. They are; 1. visible elements-weather, time, season and object. 2. all senses-vision, audition, olfaction, tactile sense, and spatial sense. 3. the line of vision-static line of vision and dynamic line of vision. In this way data is produced, and then the result of this study makes appreciation methods of seascape developed. In this way, this study results in developed appreciation of seascape. This study on new understanding of appreciation methods of seascape is on the basis of a design method of water-front that is considered a visible scene, not a design of construction elements.

• 한국항해학회지
• /
• 제18권4호
• /
• pp.155-169
• /
• 1994

The minimum temperatures are important element in the daily human life, the climatic classification, and so on. In this study, the authors aim to make an analysis the distribution characteristics of minimum temperatures of 95 weather stations in Korea by using the Climatological Standard Normals of Korea VolumeI, VolumeII, and the Climatological Standard Normals of North Korea. The important results are as follows 1) The daily, fifthly, and tenthly minimum temperatures show the highest rate of occurrance on 14th of January( Occurrance rate : 56.6% ), 16~20th of January( 37.6% ), and the middle ten days of January( 82.1% ) respectively. 2) In the regional distribution of minimum temperatures in winter, the values of northern part, inland area, and west coastal region are lower than those of southern, coastal, and east coastal regions respectively. And, bigger cities and industrial area( Seoul, Incheon, Daejeon, Daegu, etc. ) have larger values than the its vicinities. 3) When the daily minimum temperature is $0^{\circ}C$ and less, the days of northern part, inland area, and wests coastal region are higher than those of southern, coastal, and east coastal regions respectively.

• Journal of Information Processing Systems
• /
• 제17권2호
• /
• pp.411-425
• /
• 2021

In this paper, a vehicle recognition algorithm based on deep convolutional neural network and compression dictionary is proposed. Firstly, the network structure of fine vehicle recognition based on convolutional neural network is introduced. Then, a vehicle recognition system based on multi-scale pyramid convolutional neural network is constructed. The contribution of different networks to the recognition results is adjusted by the adaptive fusion method that adjusts the network according to the recognition accuracy of a single network. The proportion of output in the network output of the entire multiscale network. Then, the compressed dictionary learning and the data dimension reduction are carried out using the effective block structure method combined with very sparse random projection matrix, which solves the computational complexity caused by high-dimensional features and shortens the dictionary learning time. Finally, the sparse representation classification method is used to realize vehicle type recognition. The experimental results show that the detection effect of the proposed algorithm is stable in sunny, cloudy and rainy weather, and it has strong adaptability to typical application scenarios such as occlusion and blurring, with an average recognition rate of more than 95%.

• Journal of Information Processing Systems
• /
• 제15권5호
• /
• pp.1029-1035
• /
• 2019

In recent years, artificial intelligence (AI) services have become one of the most essential parts to extend human capabilities in various fields such as face recognition for security, weather prediction, and so on. Various learning algorithms for existing AI services are utilized, such as classification, regression, and deep learning, to increase accuracy and efficiency for humans. Nonetheless, these services face many challenges such as fake news spread on social media, stock selection, and volatility delay in stock prediction systems and inaccurate movie-based recommendation systems. In this paper, various algorithms are presented to mitigate these issues in different systems and services. Convolutional neural network algorithms are used for detecting fake news in Korean language with a Word-Embedded model. It is based on k-clique and data mining and increased accuracy in personalized recommendation-based services stock selection and volatility delay in stock prediction. Other algorithms like multi-level fusion processing address problems of lack of real-time database.

• 대기
• /
• 제16권1호
• /
• pp.33-48
• /
• 2006

A classification of snowfall type based on development mechanism is proposed using previous snowfall studies, operational experiences, etc. Five types are proposed: snowfall caused by 1) airmass transformation (AT type), 2) terrain effects in a situation of expanding Siberian High (TE type), 3) precipitation systems associated with extratropical cyclones (EC type), 4) indirect effects of extratropical cyclones passing over the sea to the south of the Korean peninsula (ECS type), and 5) combined effects of TE and ECS types (COM type). Snowfall events during 1981-2001 are classified according to the 5 types mentioned above. For this, 118 events, with at least one station with daily snowfall depth greater than 20 cm, are selected. For the classification, synoptic weather charts, satellite images, and precipitation data are used. For TE and COM types, local sea-level pressure chart is also used to confirm the presence of condition for TE type (this is done for events in 1990 and thereafter). The classification shows that 109 out of 118 events can be classified as one of the 5 types. In the remaining 8 events, heavy snowfall occurred only in Ullung Island. Its occurrence may be due to one or more of the following mechanism: airmass transformation, mesoscale cyclones and/or mesoscale convergence over the East Sea, etc. Each type shows different characteristics in location of snowfall and composition of precipitation (i.e., dry snow, rain, and mixed precipitation). The AT-type snowfall occurs mostly in the west coast, Jeju and Ullung Islands whereas the TE-type snowfall occurs in the East coast especially over the Young Dong area. The ECS-type snowfall occurs mostly over the southern part of the peninsula and some east cost area (sometimes, whole south Korea depending on the location of cyclones). The EC- and COM-type snowfalls occur in wider area, often whole south Korea. Precipitation composition also varies with the type. The AT-type has a snow ratio (SR) higher than the mean value. The TE- and EC-type have SR similar to the mean. The ECS- and COM-type have SR values smaller than the mean. Generally the SR values at high latitude and mountainous areas are higher than those at the other areas. The SR value informs the characteristics of the precipitation composition. An SR value larger than 10 means that all precipitation is composed of snow whereas a zero SR value means that all precipitation is composed of rain.

• 한국콘크리트학회:학술대회논문집
• /
• 한국콘크리트학회 2008년도 춘계 학술발표회 제20권1호
• /
• pp.997-1000
• /
• 2008

동절기 공사시에는 낮은 외기온에 기인하여 콘크리트의 조기강도 발현이 취약하게 된다. 특히, 동절기 공사중, 외부의 절취면을 대상으로 Nail를 삽입함으로서 사면 전체를 일체화시키는 Soil Nailing 공법을 적용하는 경우에는 낮은 외기온에 기인한 Shotcrete의 강도발현에 많은 어려움이 있을 것으로 예상되고 있다. 따라서, 본 연구에서는 상기의 어려움을 해결할 목적으로, 시멘트 생산공정 중 부수적으로 발생되는 초미립자 시멘트(이하 미립자 시멘트라 칭함)를 이용하여 동절기 Shotcrete공사의 조기강도 발현성능을 확보하고자 하였다. 실험결과, 외기온이 최저 $-9^{\circ}C$의 극저온 조건이라 할지라도 미립자 시멘트를 100% 사용하면서, 간단한 비닐보양 양생을 실시한다면 약 5일 정도의 단기재령만으로도 설계기준강도를 만족하는 우수한 품질의 Shotcrete를 경제적으로 제조 할 수 있는 것으로 나타났다.