• 제어로봇시스템학회논문지
• /
• 제16권10호
• /
• pp.957-962
• /
• 2010

If we want to recognize the human's emotion via the facial image, first of all, we need to extract the emotional features from the facial image by using a feature extraction algorithm. And we need to classify the emotional status by using pattern classification method. The AAM (Active Appearance Model) is a well-known method that can represent a non-rigid object, such as face, facial expression. The Bayesian Network is a probability based classifier that can represent the probabilistic relationships between a set of facial features. In this paper, our approach to facial feature extraction lies in the proposed feature extraction method based on combining AAM with FACS (Facial Action Coding System) for automatically modeling and extracting the facial emotional features. To recognize the facial emotion, we use the DBNs (Dynamic Bayesian Networks) for modeling and understanding the temporal phases of facial expressions in image sequences. The result of emotion recognition can be used to rehabilitate based on biofeedback for emotional disabled.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 1996년도 Proceedings International Workshop on New Video Media Technology
• /
• pp.72-76
• /
• 1996

We propose a novel algorithm for fractal video sequence coding, based on the circular prediction mapping (CPM), in which each range block is approximated by a domain block in the circularly previous frame. In our approach, the size of the domain block is set to be same as that of the range block for exploiting the high temporal correlation between the adjacent frames, while most other fractal coders use the domain block larger than the range block. Therefore the domain-range mapping in the CPM is similar to the block matching algorithm in the motion compensation techniques, and the advantages of this similarity are discussed. Also we show that the CPM can be combined with non-contractive inter-frame mapping (NCIM), improving the performance of the fractal sequence coder further. The computer simulation results on real image sequences demonstrate that the proposed algorithm provides very promising performance at low bit-rate, ranging from 40 Kbps to 250 Kbps.

• 한국작물학회:학술대회논문집
• /
• 한국작물학회 2005년도 국제학술회의
• /
• pp.100-113
• /
• 2005

Leaf color and plant vigor are key indicators of crop health. These visual plant attributes are frequently used by greenhouse managers, producers, and consultants to make water, nutrient, and disease management decisions. Remote sensing techniques can quickly quantify soil and plant attributes, but it requires humans to translate such data into meaningful information. Over time, scientists have used reflectance data from individual wavebands to develop a series of indices that attempt to quantify things like soil organic matter content, leaf chlorophyll concentration, leaf area index, vegetative cover, amount of living biomass, and grain yield. The recent introduction of active sensors that function independent of natural light has greatly expanded the capabilities of scientists and managers to obtain useful information. Characteristics and limitations of active sensors need to be understood to optimize their use for making improved management decisions. Pot experiments involving sand culture were conducted in 2003 and 2004 in a green house to evaluate corn and red pepper biomass. The rNDVI, gNDVI and aNDVI by ground-based remote sensors were used for evaluation of corn and red pepper biomass. The result obtained from the case study was shown that ground remote sensing as a non-destructive real-time assessment of plant nitrogen status was thought to be a useful tool for in season crop nitrogen management providing both spatial and temporal information.

• 한국재료학회:학술대회논문집
• /
• 한국재료학회 2012년도 춘계학술발표대회
• /
• pp.51-51
• /
• 2012

There is a growing interest in innovative chemical synthesis in microreactors owing to high efficiency, selectivity, and yield. In microfluidic systems, the low-volume spatial and temporal control of reactants and products offers a novel method for chemical manipulation and product generation. Glass, silicon, poly(dimethylsiloxane) (PDMS), and plastics have been used for the fabrication of miniaturized devices. However, these materials are not the best due to either of low chemical durability or expensive fabrication costs. In our group, we have recently addressed the demand for economical resistant materials that can be used for easy fabrication of microfluidic systems with reliable durability. We have suggested the use of various specialty polymers such as silicon-based inorganic polymers and fluoropolymer, flexible polyimide (PI) films that have not been used for microfluidic devices, although they have been used for other areas. And inexpensive lithography techniques were used to fabricate Lab-on-a-Chip type of microreactors with differently devised microchannel design. These microreactors were demonstrated for various synthetic reactions: liquid, liquid-gas organic chemical reactions in heterogeneous catalytic processes, syntheses of polymer and non-trivial inorganic materials. The microreactors were inert, and withstand even harsh conditions, including hydrothermal reaction. In addition, various built-in microstructures inside the microchannels, for example Pd decorated peptide nanowires, definitely enhance the uniqueness and performance of microreactors. These user-friendly Lab-on-a-Chip devices are useful alternatives for chemist and chemical engineer to conventional chemical tools such as glass.

• Current Research on Agriculture and Life Sciences
• /
• 제32권3호
• /
• pp.135-143
• /
• 2014

The variability and temporal trends of the annual and seasonal minimum and maximum temperature, rainfall, relative humidity, wind speed, sunshine hours, and runoff were analyzed for 5 major rivers in Korea from 1960 to 2010. A simple regression and non-parametric methods (Mann-Kendall test and Sen's estimator) were used in this study. The analysis results show that the minimum temperature ($T_{min}$) had a higher increasing trend than the maximum temperature ($T_{max}$), and the average temperature increased by about $0.03^{\circ}C\;yr.^{-1}$. The relative humidity and wind speed decreased by $0.02%\;yr^{-1}$ and $0.01m\;s^{-1}yr^{-1}$, respectively. With the exception of the Han River basin, the regression analysis and Mann-Kendall and Sen results failed to detect trends for the runoff and rainfall over the study period. Rapid land use changes were linked to the increase in the runoff in the Han River basin. The sensitivity of the evapotranspiration and ultimately the runoff to the meteorological variables was in the order of relative humidity > sunshine duration > wind speed > $T_{max}$ > $T_{min}$. Future studies should investigate the interaction of the variables analyzed herein, and their relative contributions to the runoff trends.

• 반도체디스플레이기술학회지
• /
• 제22권2호
• /
• pp.6-10
• /
• 2023

High-potential data can be predicted and stored in the cache to prevent cache misses, thus reducing the processor's request and wait times. As a result, the processor can work non-stop, hiding memory latency. By utilizing the temporal/spatial locality of memory access, the prefetcher introduced to improve the performance of these computers predicts the following memory address will be accessed. We propose a prefetcher that applies the GRU model, which is advantageous for handling time series data. Display the currently accessed address in binary and use it as training data to train the Gated Recurrent Unit model based on the difference (delta) between consecutive memory accesses. Finally, using a GRU model with learned memory access patterns, the proposed data prefetcher predicts the memory address to be accessed next. We have compared the model with the multi-layer perceptron, but our prefetcher showed better results than the Multi-Layer Perceptron.

• 한국수자원학회논문집
• /
• 제54권12호
• /
• pp.1317-1328
• /
• 2021

본 연구에서는 예측 모델의 정확성이 비교적 높은 월단위의 GloSea5 자료를 기반으로 예측강수량을 편의보정 및 시공간적으로 상세화하여 연속된 일단위 강우량을 모의하고자 하였다. 이를 위하여 GloSea5를 입력자료로 조건부 Copula와 MNHMM 모형을 적용하여 일단위 시계열 강우량 예측정보를 생산할 수 있는 모델링 체계를 제시하였다. 모의결과 동기간의 자료라도 매주 생산되는 결과가 큰 차이를 나타내는 예측강수량의 변동성이 유의하게 개선되었다. 모형 검증에서 모의된 일강수량, 연속강우확률, 연속무강우확률 및 강우일수가 관측자료와 유사한 값으로 모의되는 등 수문모형의 입력자료로써 활용성이 클 것으로 판단된다. 유역 단위에서의 모의된 강수량 계열간의 상관성 차이가 최소 -0.02에서 최대 0.10로 유역의 강우관측소간 상호종속성을 효과적으로 복원되는 등 수문모형의 입력자료로 활용 시 유역의 수문기상학적 반응을 보다 현실적으로 모의가 가능할 것으로 기대된다.

• 한국통신학회논문지
• /
• 제31권5C호
• /
• pp.534-539
• /
• 2006

본 논문에서는 스케일러블 비디오 코딩을 위한 새로운 프레임 예측 필터링 기법과 하드웨어 구조를 제안하였다. MCTF와 hierarchical B-picture는 비디오 프레임간의 상관성을 제거하는 기술의 일종으로 본 논문에서 다루고자 하는 대상이다. 두 기술은 시간에 대해서 비인과성 시스템에 해당하므로 소프트웨어 및 하드웨어 구현 시에 프레임 버퍼링을 위한 대기지연시간이 매우 길고 대용량의 프레임 버퍼를 요구하는 단점이 있다. 이러한 비인과성 시스템을 인과성 시스템으로 재구성하여 효율적으로 구현할 수 있는 구조를 제안하고자 한다. 동일한 연산이 반복으로 수행되는 특성을 이용하여 단위 연산을 수행할 수 있는 프레임 예측 필터링 셀(FPFC : frame prediction filtering cell)을 제안하고 이를 확장하여 전체 연산구조를 재구성하였다. 먼저, 연산의 동작 순서를 분석하고 하드웨어의 구현을 고려한 인과성을 부여한 후 단위 프레임 처리를 위한 셀을 최적화하였다. 제안한 셀의 단순한 확장을 통해서 FPFC 커널을 구성하고, 이를 이용하여 스케일러블 비디오 코딩을 위한 FPFC 프로세서를 구현하였다.

• 한국물환경학회지
• /
• 제37권6호
• /
• pp.531-538
• /
• 2021

In this study, the TOC in six estuarine reservoirs in the West Sea (Ganwol, Namyang, Daeho, Bunam, Sapkyo, and Asan) was estimated using optically-active water quality factors by the water environment monitoring network. First, specification data and land use maps of each estuarine reservoir were collected. Subsequently, water quality data from 2013 to 2020 were collected. The data comprised of 11 parameters: pH, dissolved oxygen, BOD, COD, suspended solids (SS), total nitrogen, total phosphorus, water temperature, electrical conductivity, total coliforms, and chlorophyll-a (Chl-a). The TOC in the estuarine reservoirs was 4.9~7.0 mg/L, with the highest TOC of 7.0 mg/L observed at the Namyang reservoir, which has a low shape coefficient and high drainage density. The correlation of TOC with water quality factors was also analyzed, and the correlation coefficients of Chl-a and SS were 0.28 and 0.19, respectively, while the correlation coefficients of these factors in the Namyang reservoir were 0.42 and 0.27, respectively. To improve the estimation of TOC using Chl-a and SS, the TOC was averaged in 5 mg/L units, and Chl-a and SS were averaged. Correlation analysis was then performed and the R² of Chl-a-TOC was 0.73. The R² of SS-TOC was 0.73 with a non-linear relationship. TOC had a significant non-linear relationship with Chl-a and SS. However, the relationship should be assessed in terms of the spatial and temporal variations to construct a reliable remote sensing system.

• 한국방재학회 논문집
• /
• 제5권2호
• /
• pp.9-16
• /
• 2005

본 논문에서는 유역 내 침수피해를 예측할 수 있는 실현가능한 수단으로써 인공신경망의 활용에 대해 제안하고자 한다. 유역 내 다양한 환경인자에 의한 침수피해 예측모형의 구축을 위해 108개 중유역을 대상으로 1990년부터 2000년까지 강우량, 침수피해면적, 토지이용 등 총 27개의 매개변수를 선정하여 총 49개의 데이터 세트를 구성하였다. 연구결과, 침수피해는 강우량과 같은 기상정보 뿐 만 아니라 다양한 유역환경의 특성에 영향을 받는 것으로 나타났으며 인공신경망 모형에 의해 R=0.92 수준에서 예측값과 관측값이 잘 일치하는 것으로 나타났다. 따라서 인공신경망은 입력값들과 대응된 출력값들을 알고 있는 경우 과거와 현재의 시공간 정보를 활용하여 특정유역의 강우량에 따른 침수피해면적을 산정 할 수 있으며, 복잡하고 비선형적 역동성을 지니고 있는 유역 내 환경변화에 대한 예측모형으로 활용이 가능하다고 판단된다. 또한 인공신경망은 입력자료의 중요도를 평가하는데 이용될 수 있으며, 기존 모형에서 다루어지는 매개변수중요도를 정량화 시킬 수 있어 다른 모형의 매개변수 추정이나 보정에 도움을 줄 수 있을 것으로 판단된다.