• Korean Journal of Remote Sensing
• /
• v.39 no.2
• /
• pp.207-221
• /
• 2023

This study suggests deep neural network models for estimating air temperature with Level 1B (L1B) datasets of GEO-KOMPSAT-2A (GK-2A). The temperature at 1.5 m above the ground impact not only daily life but also weather warnings such as cold and heat waves. There are many studies to assume the air temperature from the land surface temperature (LST) retrieved from satellites because the air temperature has a strong relationship with the LST. However, an algorithm of the LST, Level 2 output of GK-2A, works only clear sky pixels. To overcome the cloud effects, we apply a deep neural network (DNN) model to assume the air temperature with L1B calibrated for radiometric and geometrics from raw satellite data and compare the model with a linear regression model between LST and air temperature. The root mean square errors (RMSE) of the air temperature for model outputs are used to evaluate the model. The number of 95 in-situ air temperature data was 2,496,634 and the ratio of datasets paired with LST and L1B show 42.1% and 98.4%. The training years are 2020 and 2021 and 2022 is used to validate. The DNN model is designed with an input layer taking 16 channels and four hidden fully connected layers to assume an air temperature. As a result of the model using 16 bands of L1B, the DNN with RMSE 2.22℃ showed great performance than the baseline model with RMSE 3.55℃ on clear sky conditions and the total RMSE including overcast samples was 3.33℃. It is suggested that the DNN is able to overcome cloud effects. However, it showed different characteristics in seasonal and hourly analysis and needed to append solar information as inputs to make a general DNN model because the summer and winter seasons showed a low coefficient of determinations with high standard deviations.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.25 no.3
• /
• pp.55-66
• /
• 2020

Near-inertial waves (NIW) which are primarily generated by wind can contribute to vertical mixing in the ocean. The energetic NIW can be generated by typhoon due to its strong wind and preferable wind direction changes especially on the right-hand side of the typhoon. Here we investigate the generation and distribution of NIW using the output of a real-time ocean forecasting system. Five-year model outputs during 2013-2017 are analyzed with a focus on cases of energetic NIW generation by the passage of three typhoons (Halong, Goni, and Chaba) over the East Sea. Calculations of wind energy input (${\bar{W}}_I$), and horizontal kinetic energy in the mixed layer (${\bar{HKE}}_{MLD}$) reveal that the spatial distribution of ${\bar{HKE}}_{MLD}$, which is strengthened at the right-hand side of typhoon tracks, is closely related with ${\bar{W}}_I$. Horizontal kinetic energy in the deep layer (${\bar{HKE}}_{DEEP}$) shows patch-shaped distribution mainly located at the southern side of the East Sea. Spatial distribution of ${\bar{HKE}}_{DEEP}$ shows a close relationship with negative relative vorticity regions caused by warm eddies in the upper layer. Monthly-mean ${\bar{HKE}}_{MLD}$ and ${\bar{HKE}}_{DEEP}$ during a typhoon passing over the East Sea shows about 2.5-5.7 times and 1.2-1.6 times larger values than those during summer with no typhoons, respectively. In addition, their magnitudes are respectively about 0.4-1.0 and 0.8-1.0 times from those during winter, suggesting that the typhoon-induced NIW can provide a significant energy to enhance vertical mixing at both the mixed and deep layers during summer.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.4 s.316
• /
• pp.60-69
• /
• 2007

The human eye usually experiences a loss of color sensitivity when it is subjected to high levels of luminance, and perceives a discrepancy in color between high and normal-luminance displays, generally known as a hue shift. Accordingly, this paper models the hue-shift phenomenon and proposes a hue-correction method to provide perceptual matching between high and normal-luminance displays. The value of hue-shift is determined by perceived hue matching experiments. At first the phenomenon is observed at three lightness levels, that is, the ratio of luminance is the same between high and normal-luminance display when the perceived hue matching experiments we performed. To quantify the hue-shift phenomenon for the whole hue angle, color patches with the same lightness are first created and equally spaced inside the hue angle. These patches are then displayed one-by-one on both displays with the ratio of luminance between two displays. Next, the hue value for each patch appearing on the high-luminance display is adjusted by observers until the perceived hue for the patches on both displays appears the same visually. After obtaining the hue-shift values, these values are fit piecewise to allow shifted-hue amounts to be approximately determined for arbitrary hue values of pixels in a high-luminance display and then used for correction. Essentially, input RGB values of an image is converted to CIELAB values, and then, LCh (lightness, chroma, and hue) values are calculated to obtain the hue values for all the pixels. These hue values are shifted according to the amount calculated by the functions of the hue-shift model. Finally, the corrected CIELAB values are calculated from corrected hue values, after that, output RGB values for all pixels are estimated. For evaluation, an observer's preference test was performed with hue-shift results and Almost observers conclude that the images from hue-shift model were visually matched with images on normal luminance display.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.9
• /
• pp.512-522
• /
• 2003

For a time-multiplexed FPGA, a circuit is partitioned into several subcircuits, so that they temporally share the same physical FPGA device by hardware reconfiguration. In these architectures, all the hardware reconfiguration information called contexts are generated and downloaded into the chip, and then the pre-scheduled context switches occur properly and timely. Typically, the size of the chip required to implement the circuit depends on both the maximum number of the LUT blocks required to implement the function of each subcircuit and the maximum number of micro-registers to store results over context switches in the same time. Therefore, many partitioning or synthesis methods try to minimize these two factors. In this paper, we present a new estimation technique to find the lower bound on the number of micro-registers which can be obtained by any synthesis methods, respectively, without performing any actual synthesis and/or design space exploration. The lower bound estimation is very important in sense that it greatly helps to evaluate the results of the previous work and even the future work. If the estimated lower bound exactly matches the actual number in the actual design result, we can say that the result is guaranteed to be optimal. In contrast, if they do not match, the following two cases are expected: we might estimate a better (more exact) lower bound or we find a new synthesis result better than those of the previous work. Our experimental results show that there are some differences between the numbers of micro-registers and our estimated lower bounds. One reason for these differences seems that our estimation tries to estimate the result with the minimum micro-registers among all the possible candidates, regardless of usage of other resources such as LUTs, while the previous work takes into account both LUTs and micro-registers. In addition, it implies that our method may have some limitation on exact estimation due to the complexity of the problem itself in sense that it is much more complicated than LUT estimation and thus needs more improvement, and/or there may exist some other synthesis results better than those of the previous work.

• Journal of the Korean Society of Radiology
• /
• v.6 no.6
• /
• pp.447-454
• /
• 2012

To examine the performance of a diagnostic X-ray system, we tested a linearity, reproducibility, and Half Value Layer(HVL). The linearity was examined 4 times of irradiation with a given condition, and we recorded a level of radiation. We then calculated the mR/mAs. And the measured value should not be more than 0.1. If the measured value was more than 0.1, we could know that the linearity was decreased. The reproducibility was analyzed 10 times of irradiations at 80kVp, 200mA, 20mAs and 120kVp, 300mA, 8mAs. The values from these analyses were integrated into CV equation, and we could get outputs. The reproducibility was good if the output was lower than 0.05. HVL was measured 3 times of irradiation without a filter, and we inserted additional HLV filters with 0, 1, 2, 4 mm of thickness. We tested the values until we get the measured value less than a half of the value measured without additional filter. We tested the linearity, the reproducibility, and HVL of 5 diagnostic X-ray generators in this facilities. The linearity of No. 1 and No. 5 generator didn't satisfy the standard for radiation safety around 300mA~400mA and 100mA~200mA, respectively. HVL of No.1 generator was not satisfied at 80kVp. The outputs were higher in the three-phase equipment than the single-phase equipment. The old generators need to maintain and exchange of components based on the these results. Then, we could contribute to getting more exact diagnosis increasing a quality of the image and decreasing an expose dose of radiation.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.11 no.2
• /
• pp.61-71
• /
• 2009

Most deciduous trees in temperate zone are dormant during the winter to overcome cold and dry environment. Dormancy of deciduous fruit trees is usually separated into a period of rest by physiological conditions and a period of quiescence by unfavorable environmental conditions. Inconsistent and fewer budburst in pear orchards has been reported recently in South Korea and Japan and the insufficient chilling due to warmer winters is suspected to play a role. An accurate prediction of the flowering time under the climate change scenarios may be critical to the planning of adaptation strategy for the pear industry in the future. However, existing methods for the prediction of budburst depend on the spring temperature, neglecting potential effects of warmer winters on the rest release and subsequent budburst. We adapted a dormancy clock model which uses daily temperature data to calculate the thermal time for simulating winter phenology of deciduous trees and tested the feasibility of this model in predicting budburst and flowering of Niitaka pear, one of the favorite cultivars in Korea. In order to derive the model parameter values suitable for Niitaka, the mean time for the rest release was estimated by observing budburst of field collected twigs in a controlled environment. The thermal time (in chill-days) was calculated and accumulated by a predefined temperature range from fall harvest until the chilling requirement (maximum accumulated chill-days in a negative number) is met. The chilling requirement is then offset by anti-chill days (in positive numbers) until the accumulated chill-days become null, which is assumed to be the budburst date. Calculations were repeated with arbitrary threshold temperatures from $4^{\circ}C$ to $10^{\circ}C$ (at an interval of 0.1), and a set of threshold temperature and chilling requirement was selected when the estimated budburst date coincides with the field observation. A heating requirement (in accumulation of anti-chill days since budburst) for flowering was also determined from an experiment based on historical observations. The dormancy clock model optimized with the selected parameter values was used to predict flowering of Niitaka pear grown in Suwon for the recent 9 years. The predicted dates for full bloom were within the range of the observed dates with 1.9 days of root mean square error.

• Journal of Korea Water Resources Association
• /
• v.54 no.10
• /
• pp.795-805
• /
• 2021

In this study, after developing an LSTM-based deep learning model for estimating daily runoff in the Soyang River Dam basin, the accuracy of the model for various combinations of model structure and input data was investigated. A model was built based on the database consisting of average daily precipitation, average daily temperature, average daily wind speed (input up to here), and daily average flow rate (output) during the first 12 years (1997.1.1-2008.12.31). The Nash-Sutcliffe Model Efficiency Coefficient (NSE) and RMSE were examined for validation using the flow discharge data of the later 12 years (2009.1.1-2020.12.31). The combination that showed the highest accuracy was the case in which all possible input data (12 years of daily precipitation, weather temperature, wind speed) were used on the LSTM model structure with 64 hidden units. The NSE and RMSE of the verification period were 0.862 and 76.8 m³/s, respectively. When the number of hidden units of LSTM exceeds 500, the performance degradation of the model due to overfitting begins to appear, and when the number of hidden units exceeds 1000, the overfitting problem becomes prominent. A model with very high performance (NSE=0.8~0.84) could be obtained when only 12 years of daily precipitation was used for model training. A model with reasonably high performance (NSE=0.63-0.85) when only one year of input data was used for model training. In particular, an accurate model (NSE=0.85) could be obtained if the one year of training data contains a wide magnitude of flow events such as extreme flow and droughts as well as normal events. If the training data includes both the normal and extreme flow rates, input data that is longer than 5 years did not significantly improve the model performance.