• Asian-Australasian Journal of Animal Sciences
• /
• v.29 no.9
• /
• pp.1215-1221
• /
• 2016

The prediction of carcass composition in Hanwoo steers is very important for value-based marketing, and the improvement of prediction accuracy and precision can be achieved through the analyses of independent variables using a prediction equation with a sufficient dataset. The present study was conducted to develop a prediction equation for Hanwoo carcass composition for which data was collected from 7,907 Hanwoo steers raised at a private farm in Gangwon Province, South Korea, and slaughtered in the period between January 2009 and September 2014. Carcass traits such as carcass weight (CWT), back fat thickness (BFT), eye-muscle area (EMA), and marbling score (MAR) were used as independent variables for the development of a prediction equation for carcass composition, such as retail cut weight and percentage (RC, and %RC, respectively), trimmed fat weight and percentage (FAT, and %FAT, respectively), and separated bone weight and percentage (BONE, and %BONE), and its feasibility for practical use was evaluated using the estimated retail yield percentage (ELP) currently used in Korea. The equations were functions of all the variables, and the significance was estimated via stepwise regression analyses. Further, the model equations were verified by means of the residual standard deviation and the coefficient of determination ($R^2$) between the predicted and observed values. As the results of stepwise analyses, CWT was the most important single variable in the equation for RC and FAT, and BFT was the most important variable for the equation of %RC and %FAT. The precision and accuracy of three variable equation consisting CWT, BFT, and EMA were very similar to those of four variable equation that included all for independent variables (CWT, BFT, EMA, and MAR) in RC and FAT, while the three variable equations provided a more accurate prediction for %RC. Consequently, the three-variable equation might be more appropriate for practical use than the four-variable equation based on its easy and cost-effective measurement. However, a relatively high average difference for the ELP in absolute value implies a revision of the official equation may be required, although the current official equation for predicting RC with three variables is still valid.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.2
• /
• pp.203-208
• /
• 2006

The main research conducted previously on GPS ionosphere in China is first introduced. Besides, the current investigations include as follows: (1) GPS-based spatial environmental, especially the ionosphere, monitoring, modeling and analysis, including ground/space-based GPS ionosphere electron density (IED) through occultation/tomography technologies with GPS data from global/regional network, development of a GNSS-based platform for imaging ionosphere and atmosphere (GPFIIA), and preliminary test results through performing the first 3D imaging for the IED over China, (2) The atmospheric and ionospheric modeling for GPS-based surveying, navigation and orbit determination, involving high precisely ionospheric TEC modeling for phase-based long/median range network RTK system for achieving CM-level real time positioning, next generation GNSS broadcast ionospheric time-delay algorithm required for higher correction accuracy, and orbit determination for Low-Earth-orbiter satellites using single frequency GPS receivers, and (3) Research products in applications for national significant projects: GPS-based ionospheric effects modeling for precise positioning and orbit determination applied to China's manned space-engineering, including spatial robot navigation and control and international space station intersection and docking required for related national significant projects.

• The korean journal of orthodontics
• /
• v.54 no.1
• /
• pp.48-58
• /
• 2024

Objective: To quantify the effects of midline-related landmark identification on midline deviation measurements in posteroanterior (PA) cephalograms using a cascaded convolutional neural network (CNN). Methods: A total of 2,903 PA cephalogram images obtained from 9 university hospitals were divided into training, internal validation, and test sets (n = 2,150, 376, and 377). As the gold standard, 2 orthodontic professors marked the bilateral landmarks, including the frontozygomatic suture point and latero-orbitale (LO), and the midline landmarks, including the crista galli, anterior nasal spine (ANS), upper dental midpoint (UDM), lower dental midpoint (LDM), and menton (Me). For the test, Examiner-1 and Examiner-2 (3-year and 1-year orthodontic residents) and the Cascaded-CNN models marked the landmarks. After point-to-point errors of landmark identification, the successful detection rate (SDR) and distance and direction of the midline landmark deviation from the midsagittal line (ANS-mid, UDM-mid, LDM-mid, and Me-mid) were measured, and statistical analysis was performed. Results: The cascaded-CNN algorithm showed a clinically acceptable level of point-to-point error (1.26 mm vs. 1.57 mm in Examiner-1 and 1.75 mm in Examiner-2). The average SDR within the 2 mm range was 83.2%, with high accuracy at the LO (right, 96.9%; left, 97.1%), and UDM (96.9%). The absolute measurement errors were less than 1 mm for ANS-mid, UDM-mid, and LDM-mid compared with the gold standard. Conclusions: The cascaded-CNN model may be considered an effective tool for the auto-identification of midline landmarks and quantification of midline deviation in PA cephalograms of adult patients, regardless of variations in the image acquisition method.

• Journal of Korea Water Resources Association
• /
• v.49 no.6
• /
• pp.565-577
• /
• 2016

The objective of this study is to propose a method for computing the Natural Drought Index (NDI) that does not consider man-made drought facilities. Principal Component Analysis (PCA) was used to estimate the NDI. Three monthly moving cumulative runoff, soil moisture and precipitation were selected as input data of the NDI during 1977~2012. Observed precipitation data was collected from KMA ASOS (Korea Meteorological Association Automatic Synoptic Observation System), while model-driven runoff and soil moisture from Variable Infiltration Capacity Model (VIC Model) were used. Time series analysis, drought characteristic analysis and spatial analysis were used to assess the utilization of NDI and compare with existing SPI, SRI and SSI. The NDI precisely reflected onset and termination of past drought events with mean absolute error of 0.85 in time series analysis. It explained well duration and inter-arrival time with 1.3 and 1.0 respectively in drought characteristic analysis. Also, the NDI reflected regional drought condition well in spatial analysis. The accuracy rank of drought onset, termination, duration and inter-arrival time was calculated by using NDI, SPI, SRI and SSI. The result showed that NDI is more precise than the others. The NDI overcomes the limitation of univariate drought indices and can be useful for drought analysis as representative measure of different types of drought such as meteorological, hydrological and agricultural droughts.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1125-1139
• /
• 2022

In an unmanned aerial vehicles (UAVs) system, a physical offset can be existed between the global positioning system/inertial measurement unit (GPS/IMU) sensor and the observation sensor such as a hyperspectral sensor, and a lidar sensor. As a result of the physical offset, a misalignment between each image can be occurred along with a flight direction. In particular, in a case of multi-sensor system, an observation sensor has to be replaced regularly to equip another observation sensor, and then, a high cost should be paid to acquire a calibration parameter. In this study, we establish a precise sensor model equation to apply for a multiple sensor in common and propose an independent physical offset estimation method. The proposed method consists of 3 steps. Firstly, we define an appropriate rotation matrix for our system, and an initial sensor model equation for direct-georeferencing. Next, an observation equation for the physical offset estimation is established by extracting a corresponding point between a ground control point and the observed data from a sensor. Finally, the physical offset is estimated based on the observed data, and the precise sensor model equation is established by applying the estimated parameters to the initial sensor model equation. 4 region's datasets(Jeon-ju, Incheon, Alaska, Norway) with a different latitude, longitude were compared to analyze the effects of the calibration parameter. We confirmed that a misalignment between images were adjusted after applying for the physical offset in the sensor model equation. An absolute position accuracy was analyzed in the Incheon dataset, compared to a ground control point. For the hyperspectral image, root mean square error (RMSE) for X, Y direction was calculated for 0.12 m, and for the point cloud, RMSE was calculated for 0.03 m. Furthermore, a relative position accuracy for a specific point between the adjusted point cloud and the hyperspectral images were also analyzed for 0.07 m, so we confirmed that a precise data mapping is available for an observation without a ground control point through the proposed estimation method, and we also confirmed a possibility of multi-sensor fusion. From this study, we expect that a flexible multi-sensor platform system can be operated through the independent parameter estimation method with an economic cost saving.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.33 no.3
• /
• pp.937-945
• /
• 2013

The accuracy of flood inundation maps is determined by the uncertainty propagated from all variables involved in the overall process including input data, model parameters and modeling approaches. This study investigated the uncertainty arising from key variables (flow condition and Manning's n) among model variables in flood inundation mapping for the Missouri River near Boonville, Missouri, USA. Methodology of this study involves the generalized likelihood uncertainty estimation (GLUE) to quantify the uncertainty bounds of flood inundation area. Uncertainty bounds in the GLUE procedure are evaluated by selecting two likelihood functions, which is two statistic (inverse of sum of squared error (1/SAE) and inverse of sum of absolute error (1/SSE)) based on an observed water surface elevation and simulated water surface elevations. The results from GLUE show that likelihood measure based on 1/SSE is more sensitive on observation than likelihood measure based on 1/SAE, and that the uncertainty propagated from two variables produces an uncertainty bound of about 2% in the inundation area compared to observed inundation. Based on the results obtained form this study, it is expected that this study will be useful to identify the characteristic of flood.

• Food Science and Preservation
• /
• v.10 no.2
• /
• pp.125-130
• /
• 2003

This study was carried out to investigate the thermo-physical properties and design Freezing time prediction model from data of freezing test of Kimchi. Density of Kimchi were measured as 1001.9 ${\pm}$0.03 kg/㎥ at unfrozen state, 987.0 ${\pm}$0.07 kg/㎥ at frozen state and volume of the Kimchi expanded 4.67% at -l5$^{\circ}C$. Initial freezing point of Kimchi and seasoning were -4.0$^{\circ}C$ and -2.5$^{\circ}C$, respectively. Freezing ratio of Kimchi were estimated more than 50％ at -5.0$^{\circ}C$, more than 75% at -l0$^{\circ}C$ and approximately 90% at -25$^{\circ}C$. To obtain equation for freezing time prediction of Kimchi, freezing time(Y) was regressed against the reciprocal( $X_3$) of difference of initial freezing point and freezing medium temperature, reciprocal( $X_4$) of surface heat transfer coefficient, the initial temperature( $X_1$) and thickness( $X_2$) of samples. As results of the multiple regression analysis, equations were obtained as follows. Y$_{kimchi}$=3.856 $X_1$+13982.8 $X_2$+8305.166 $X_3$+ 3559.181 $X_4$-639.189( $R^2$=0.9632). These equations shown better results than previous models, and the accuracy of its was very high as average absolute difference of about 10% in the difference between the fitted and experimental results.

• Journal of Korea Water Resources Association
• /
• v.53 no.12
• /
• pp.1159-1172
• /
• 2020

This study applied deep convolution neural network based on U-Net and SegNet using long period weather radar data to very short-term rainfall prediction. And the results were compared and evaluated with the translation model. For training and validation of deep neural network, Mt. Gwanak and Mt. Gwangdeoksan radar data were collected from 2010 to 2016 and converted to a gray-scale image file in an HDF5 format with a 1km spatial resolution. The deep neural network model was trained to predict precipitation after 10 minutes by using the four consecutive radar image data, and the recursive method of repeating forecasts was applied to carry out lead time 60 minutes with the pretrained deep neural network model. To evaluate the performance of deep neural network prediction model, 24 rain cases in 2017 were forecast for rainfall up to 60 minutes in advance. As a result of evaluating the predicted performance by calculating the mean absolute error (MAE) and critical success index (CSI) at the threshold of 0.1, 1, and 5 mm/hr, the deep neural network model showed better performance in the case of rainfall threshold of 0.1, 1 mm/hr in terms of MAE, and showed better performance than the translation model for lead time 50 minutes in terms of CSI. In particular, although the deep neural network prediction model performed generally better than the translation model for weak rainfall of 5 mm/hr or less, the deep neural network prediction model had limitations in predicting distinct precipitation characteristics of high intensity as a result of the evaluation of threshold of 5 mm/hr. The longer lead time, the spatial smoothness increase with lead time thereby reducing the accuracy of rainfall prediction The translation model turned out to be superior in predicting the exceedance of higher intensity thresholds (> 5 mm/hr) because it preserves distinct precipitation characteristics, but the rainfall position tends to shift incorrectly. This study are expected to be helpful for the improvement of radar rainfall prediction model using deep neural networks in the future. In addition, the massive weather radar data established in this study will be provided through open repositories for future use in subsequent studies.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.249-263
• /
• 2023

Collaborative Filtering, a representative recommendation system methodology, consists of two approaches: neighbor methods and latent factor models. Among these, the latent factor model using matrix factorization decomposes the user-item interaction matrix into two lower-dimensional rectangular matrices, predicting the item's rating through the product of these matrices. Due to the factor vectors inferred from rating patterns capturing user and item characteristics, this method is superior in scalability, accuracy, and flexibility compared to neighbor-based methods. However, it has a fundamental drawback: the need to reflect the diversity of preferences of different individuals for items with no ratings. This limitation leads to repetitive and inaccurate recommendations. The Adaptive Deep Latent Factor Model (ADLFM) was developed to address this issue. This model adaptively learns the preferences for each item by using the item description, which provides a detailed summary and explanation of the item. ADLFM takes in item description as input, calculates latent vectors of the user and item, and presents a method that can reflect personal diversity using an attention score. However, due to the requirement of a dataset that includes item descriptions, the domain that can apply ADLFM is limited, resulting in generalization limitations. This study proposes a Generalized Adaptive Deep Latent Factor Recommendation Model, G-ADLFRM, to improve the limitations of ADLFM. Firstly, we use item ID, commonly used in recommendation systems, as input instead of the item description. Additionally, we apply improved deep learning model structures such as Self-Attention, Multi-head Attention, and Multi-Conv1D. We conducted experiments on various datasets with input and model structure changes. The results showed that when only the input was changed, MAE increased slightly compared to ADLFM due to accompanying information loss, resulting in decreased recommendation performance. However, the average learning speed per epoch significantly improved as the amount of information to be processed decreased. When both the input and the model structure were changed, the best-performing Multi-Conv1d structure showed similar performance to ADLFM, sufficiently counteracting the information loss caused by the input change. We conclude that G-ADLFRM is a new, lightweight, and generalizable model that maintains the performance of the existing ADLFM while enabling fast learning and inference.

• The Korean Journal of Financial Management
• /
• v.20 no.1
• /
• pp.93-124
• /
• 2003

This paper studies the financial analyst's forecasting activities on the firm's operating performance during the period from 1999 to 2003. In this study, financial analyst's forecasting activities are focused on the sales, operating income and net income and financial analyst's forecasting accuracy, forecasting revising patterns and forecasting activities to the unexpected firm's operating performance are studied. Some empirical findings in this study are as follows. First, standard estimate error on the sales, operating income and net income are all significantly negative value and so financial analyst's forecast on the firm's operating performance are upwardly biased. Second, domestic financial analyst's forecasting activities is relatively more accuracy than foreign financial analyst's forecasting activities. Third, forecasting time is more close to the end of the operating performance announcement day, forecasting activities are more accuracy. Fourth, comparing with individual financial analyst's forecast, consensus forecast is more accuracy. Fifth, in the comparative forecasting activities study according to the prior firm's operating performance, financial analyst's forecasting revision activities are found to be upward or downward. Sixth, financial analysts overreact in the sales forecast and underreact in the operating income and net income forecast. Seventh, in the empirical analysis on the Easterwood-Nutt's test model(1999) which the firm's performance change are divided into the expected performance change and the unexpected performance change, it is found that financial analyst's forecasting activities on the firm's operating performance are systematically optimistic.