• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.11
• /
• pp.4413-4425
• /
• 2020

In order to completely record all the information of realistic scenes, high dynamic range (HDR) images have been widely used in virtual reality, photography and computer graphics. A simple yet effective tone mapping method based on total variation is proposed so as to reproduce realistic scenes on low dynamic range (LDR) display devices. The structural component and texture component are obtained using total variation model in logarithmic domain. Then, the dynamic range of the structural component is compressed with an adaptive arcsine function. The texture component is processed by Taylor series. Finally, we adjust the saturation component using sigmoid function and restore the color information. Experimental results demonstrate that our method outperforms existing methods in terms of quality and speed.

• Journal of Yeungnam Medical Science
• /
• v.11 no.1
• /
• pp.82-93
• /
• 1994

Clinical study was carried out on the 64 hemodialysis patients(HD) with chronic renal failure who had been treated from December 1992 to July 1993 in Yeungnam University Hospital. The following results were obitained. In hematologic parameters, MCH was $28.8{\pm}2.0pg$, and MCV was $92.4{\pm}4.7fl$. Result revealed normochromic and normocytic anemia. Mean values of serum ferritin were $657.4{\pm}292.0ng/ml$ in men and $511.5{\pm}370g$ in women. Mean valuses of serum iron were $145.5{\pm}63.7{\mu}g/dl$. Mean values of transferrin saturation was $61.6{\pm}28.4%$. Serum frerritin, serum iron and transferrin saturation were higher in HD group than normal reference. In erythropoeitin treatment group, Hb and Hct were significantly higher than non-erythropoietin treatment group. Amount of transfusion was significantly higher in non-erythropoietin treatment group than erythropoeitin treatment group(p<0.05). Values of iron, transferrin saturation were significantly higher in abnormal liver function test(LFT) hemodialysis group than normal LFT group(p<0.05). Transfusion amounts revealed positive correlation with ferritin(r=0.4675), transferrin satruation (r=0.3823) and iron(r=0.3386)(p<0.05). In conclusion, erythropoietin treatment cna reduce requirement of blood transfusion and transfusion related side effects such as iron overload, hemosiderosis and hemochromatosis.

• Journal of Korean Biological Nursing Science
• /
• v.4 no.2
• /
• pp.93-112
• /
• 2002

Recently, the number of the elderly has increased according to the improvement of socioeconomic status and the efficient medical care system. In spite of the development of medicine, the elderly suffers from the various health problems caused by weakness of musculoskeletal system, cardiopulmonary function and immunologic dysfunction. Regular rhythmic exercise program is known to be one of the effective tools to enhance the health condition in the elderly. However, there has been few studies to evaluate the comprehensive effects of rhythmic movement program on the elderly. This study was focused to evaluate the indices of cardiopulmonary function, life satisfaction, calcium, phosphorous, osteocalcin and deoxypyridinoline which are the essential factors of health problems in the elderly women. Twenty six subjects, aged between 68 and 72, who can do the ordinary activities and do not have cardiovascular dysfunction and mental disorder, participated in this study. They were divided into two groups: 13 in the experimental group and 13 in the control group. The experimental group participated in the rhythmic movement program at the welfare center located in G-city. The program were consisted of three sessions a week during 10 weeks. Each session had three parts: warming up(10 minutes), main exercise(40 minutes), finishing(10 minutes). Heart rate, blood pressure and peripheral arterial oxygen saturation were measured for the evaluation of cardiopulmonary function. Serum calcium, phosphorous, osteocalcin and urine deoxypyridinoline were measured as the indices of bony metabolism. Data were analyzed with mean, standard deviation, $x^2$-test, t-test, paired t-test using SPSS PC+ program. The results of this study were as follows. 1) Heart rate of the experimental group showed significant decrease following the rhythmic movement program. Peripheral arterial oxygen saturation of the experimental group showed significant increase following the program. 2) The degree of life satisfaction of the experimental group showed significant increase following the program. 3) Calcium showed significant decrease following the program but remained within normal range. There was no significant difference of phosphorous between two groups. 4) Osteocalcin, the index of bone formation, showed no significant change following the program, but significant increase in the experimental group comparing with the control group. 5) Deoxypyridinoline, the index of bone resorption, in urine of the experimental group showed significant decrease following the program. In conclusion, the rhythmic exercise program in the elderly showed the improvement of physiologic function and favorable effects on life satisfaction and bony metabolism. According to the above results, the regular rhythmic movement program can be strongly recommended for the improvement of health in the elderly women.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.6
• /
• pp.67-75
• /
• 1998

Although the nCE(n-th order cross-entropy) error function resolves the incorrect saturation problem of conventional EBP(error back-propagation) algorithm, the performance of MLP's (multilayer perceptrons) trained using the nCE function depends heavily on the order of the nCE function. In this paper, we propose an adaptive learning rate to make the performance of MLP's insensitive to the order of the nCE error. Additionally, we propose a limited error signal of output node to prevent unstable learning due to the adaptive learning rate. The effectiveness of the proposed method is demonstrated in simulations of handwritten digit recognition and thyroid diagnosis tasks.

• Sleep Medicine and Psychophysiology
• /
• v.3 no.1
• /
• pp.38-46
• /
• 1996

The purpose of this study is to investigate the neurocognitive dysfunction and the degree of severity according to the oxygen desaturation in obstructive sleep apnea patients. We performed nocturnal polysomnographic recording and administered 3 Vienna Test System subtest of Reaction unit, Continuous attention and Cognitrone to 11 obstructive sleep patients and 13 controls. The result were as follows: 1) On Continous attentin and Cognitrone, patients with obstructive sleep apnea showed significant lower score on correct answer, higher score on missed answer and prolonged reaction time than control. But, there was no significant difference on Reaction unit between two groups. 2) Among 3 groups divided by degree of oxygen desaturation, there were no significant differences on Reaction unit. As the oxygen saturation decreased however, the obstructive sleep apnea group revealed significantly poor performance score on continuous attention and cognitrone. We suggested that obstructive sleep apnea patients showed disturbed neurocognitive function with complex cognitive process and the severity of neurocognitive dysfunction was also correlated with oxygen desaturation.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.3
• /
• pp.249-255
• /
• 2010

The paper proposes a new type of robust speed control strategy for permanent magnet synchronous motor by using PD-sliding mode hybrid control. The PD control has a good performance in the transient region while the sliding mode controller provides the robustness against system uncertainties. Taking advantages of the two control strategies, the proposed control method utilizes the PD control in the approaching region to the sliding surface and the sliding mode control near at the sliding surfaces. The chattering problem of the sliding mode controller is eliminated by applying the saturation function for the switching function of the sliding mode control. The stability of the sliding mode control is verified by using Lyapunov function with the proper selection of variable gains. It is shown that with this simple switching algorithm, stability of the overall hybrid control system is ensured. Through the simulations, the PD-sliding mode algorithm is shown to have a good performance in the transient response as well as being robust against disturbances. The robustness of the PD-sliding mode algorithm is further demonstrated against various external disturbances in the real experiments of SPMSM motor control.

• Journal of Power System Engineering
• /
• v.21 no.4
• /
• pp.18-25
• /
• 2017

This paper deals with vibration control of a small mount with MR(Magneto-Rheological) fluid as a functional fluid mount for precision equipment of automobiles. Damping and stiffness coefficients of the mount with MR fluid are changed by variations of the applied magnetic field strength. We present the robust control scheme, based on a conventional sliding mode control theory, for the design of a stable controller that is capable of vibration control due to various disturbances such as impact and periodic excitations, and is insensitive to dynamic properties of the mount. We got stable controller by using Lyapunov stability theory. The controller is then realized by using a semi-active control condition in simulations. Chattering problem of the sliding mode control is eliminated by saturation function instead of signum function. The sliding mode control with Lyapunov stability theory is superior to passive and Sky-Hook control in performance.

• Progress in Medical Physics
• /
• v.28 no.4
• /
• pp.135-143
• /
• 2017

Chemical Exchange Saturation Transfer (CEST) imaging is a method to detect solutes based on the chemical exchange of mobile protons with water. The solute protons exchange with three different patterns, which are fast, slow, and intermediate rates. The CEST contrast can be obtained from the exchangeable protons, which are hydroxyl protons, amine protons, and amide protons. The CEST MR imaging is useful to evaluate tumors, strokes, and other diseases. The purpose of this study is to review the mathematical model for CEST imaging and for measurement of the chemical exchange rate, and to measure the chemical exchange rate using a 3T MRI system on several amino acids. We reviewed the mathematical models for the proton exchange. Several physical models are proposed to demonstrate a two-pool, three-pool, and four-pool models. The CEST signals are also evaluated by taking account of the exchange rate, pH and the saturation efficiency. Although researchers have used most commonly in the calculation of CEST asymmetry, a quantitative analysis is also developed by using Lorentzian fitting. The chemical exchange rate was measured in the phantoms made of asparagine (Asn), glutamate (Glu), ${\gamma}-aminobutyric$ acid (GABA), glycine (Gly), and myoinositol (MI). The experiment was performed at a 3T human MRI system with three different acidity conditions (pH 5.6, 6.2, and 7.4) at a concentration of 50 mM. To identify the chemical exchange rate, the "lsqcurvefit" built-in function in MATLAB was used to fit the pseudo-first exchange rate model. The pseudo-first exchange rate of Asn and Gly was increased with decreasing acidity. In the case of GABA, the largest result was observed at pH 6.2. For Glu, the results at pH 5.6 and 6.2 did not show a significant difference, and the results at pH 7.4 were almost zero. For MI, there was no significant difference at pH 5.6 or 7.4, however, the results at pH 6.2 were smaller than at the other pH values. For the experiment at 3T, we were only able to apply 1 s as the maximum saturation duration due to the limitations of the MRI system. The measurement of the chemical exchange rate was limited in a clinical 3T MRI system because of a hardware limitation.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.19 no.7
• /
• pp.741-753
• /
• 2008

This paper explores the throughput performance of CSMA/CA-based DCF protocol over both ideal channels and fading channels with payload size at the MAC layer in the 802.11a wireless LAN. In the ideal channel, there are no errors and at the transmission cycle there is one and only one active station which always has a packet to send and other stations can only accept packets and provide acknowledgements. In the fading channel, bit errors appear in the channel randomly and the number of stations is assumed to be fixed. And each station always has packets for transmission. In other words, we operate in saturation conditions. Up to now conventional research work about DCF throughput analysis of IEEE 802.11 a wireless LAN has been done over the ideal channel, but this paper is done over the Rayleigh/Ricean fading channel. So, the ratio of received average energy per bit-to-noise power spectral density $E_b/N_o$ is set to 25 dB and the ratio of direct-to-diffuse signal power in each sub-channel $\xi$ is set to 6 for combined Rayleigh/Ricean fading channel. In conclusion, it is shown that the saturation throughput is always less than the maximum throughput at all the payload size and the higher the transmission rate be, the higher the decreasing rate of saturation throughput compared to the maximum throughput be.

• The Korean Journal of Applied Statistics
• /
• v.32 no.5
• /
• pp.693-702
• /
• 2019

Deep learning has gained popularity for the classification and prediction task. Neural network layers become deeper as more data becomes available. Saturation is the phenomenon that the gradient of an activation function gets closer to 0 and can happen when the value of weight is too big. Increased importance has been placed on the issue of saturation which limits the ability of weight to learn. To resolve this problem, Glorot and Bengio (Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, 249-256, 2010) claimed that efficient neural network training is possible when data flows variously between layers. They argued that variance over the output of each layer and variance over input of each layer are equal. They proposed a method of initialization that the variance of the output of each layer and the variance of the input should be the same. In this paper, we propose a new method of establishing initialization by adopting truncated normal distribution and truncated cauchy distribution. We decide where to truncate the distribution while adapting the initialization method by Glorot and Bengio (2010). Variances are made over output and input equal that are then accomplished by setting variances equal to the variance of truncated distribution. It manipulates the distribution so that the initial values of weights would not grow so large and with values that simultaneously get close to zero. To compare the performance of our proposed method with existing methods, we conducted experiments on MNIST and CIFAR-10 data using DNN and CNN. Our proposed method outperformed existing methods in terms of accuracy.