• Journal of the Korea Society of Computer and Information
• /
• v.16 no.12
• /
• pp.23-31
• /
• 2011

The simple bacteria cooperative optimization (sBCO) algorithm that we developed as one of optimization algorithms has shown relatively good performances, but their performances were limited by step-by-step movement of individuals at a time. In order to solve this problem, we proposed a new method that assigned a speed to each individual according to its rank and it was confirmed that it improved the performances of sBCO in some degree. In addition to the assigning of speed to the individuals, we employed a new mutation operation that most existing evolutionary algorithms used in order to enhance the performances of sBCO in this paper. A specific percent of bad individuals are mutated within an area that is proportion to the rank of the individual in the mutation operation. That is, Gaussian noise of large standard deviation is added as the fitness of individuals is low. From this, the probability that the individuals with lower ranks can be located far from its parent will be increased. This causes that the probability of falling into local optimum areas is decreased and the probability of fast escaping the local optimum areas is increased. From experimental results with four function optimization problems, we showed that the performances of sBCO with mutation operation and individual speed were increased. If the optimization function is quite complex, however, the performances are not always better. We should devise a new method for solving this problem as a further work.

• Korean Journal of Plant Taxonomy
• /
• v.49 no.2
• /
• pp.118-126
• /
• 2019

The taxonomic rank of the tiny-leaved terrestrial orchid Cephalanthera subaphylla Miyabe & $Kud{\hat{o}}$ has been somewhat controversial, as it has been treated as a species or as an infraspecific taxon, under C. erecta (Thunb.) Blume [C. erecta var. subaphylla (Miyabe & $Kud{\hat{o}}$) Ohwi and C. erecta f. subaphylla (Miyabe & $Kud{\hat{o}}$) M. Hiro]. Allozyme markers, traditionally employed for delimiting species boundaries, are used here to gain information for determining the taxonomic status of C. subaphylla. To do this, we sampled three populations of five taxa (a total of 15 populations) of Cephalanthera native to the Korean Peninsula [C. erecta, C. falcata (Thunb.) Blume, C. longibracteata Blume, C. longifolia (L.) Fritsch, and C. subaphylla]. Among 20 putative loci resolved, three were monomorphic (Dia-2, Pgi-1, and Tpi-1) across the five species. Apart from C. longibracteata, there was no allozyme variation within the remaining four species. Of the 51 alleles harbored by these 17 polymorphic loci, each of the 27 alleles at 14 loci was unique to a single species. Accordingly, we found low average values of Nei's genetic identities (I) between ten species pairs (from I = 0.250 for C. erecta versus C. longifolia to I = 0.603 for C. falcata vs. C. longibracteata), with C. subaphylla being genetically clearly differentiated from the other species (from I = 0.349 for C. subaphylla vs. C. longifolia to 0.400 for C. subaphylla vs. C. falcata). These results clearly indicate that C. subaphylla is not genetically related to any of the other taxa of Cephalanthera that are native to the Korean Peninsula, including C. erecta. In a principal coordinate analysis (PCoA), C. subaphylla was positioned distant not only from C. falcata, C. longibracteata, and C. longifolia, but also from C. erecta. Finally, K = 5 was the best clustering scheme using a Bayesian approach, with five clusters precisely corresponding to the five taxa. Thus, our allozyme results strongly suggest that C. subaphylla merits the rank of species.

• 한국연소학회:학술대회논문집
• /
• 2012.11a
• /
• pp.107-109
• /
• 2012

In Korean coal power plants, rising coal prices have recently led to the rapid utilization of low lank coals such as sub-bituminous coal with low calorific values and low ash fusion temperatures. Using these coals beyond the design range has resulted in important issues including slagging and fouling, which cause negative effects in boiler performances and unstable operations. The purpose of this study is to observe slagging and fouling characteristics resulted from burning various ranks of pulverized coals. We have tested 3 different coals: FLAME(bituminous), KCH(sub-bituminous) and MOOLARBEN(bituminous)coals in the pilot system $50kW_{th}$ scale. A stainless steel tube with preheated air inside was installed in the downstream in order to simulate water wall. Collected ash on the probe and the slag inside the furnace near burner were analyzed by SEM (scanning electron microscopy) to verify the formation degree, surface features and color changes of the pasty ash particles. Induced coupled plasma and energy dispersive X-ray spectroscopy were also performed to figure out the chemical characteristics of collected samples. As a result, KCH was observed that more slag was developed inside the walls of the furnace and on the probe than the other two kinds of coals, as shown in the calculate slagging and fouling indices as well.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.3
• /
• pp.757-770
• /
• 2022

Palmprint recognition has drawn increasingly attentions in the past decade due to its uniqueness and reliability. Traditional palmprint recognition methods usually use high-resolution images as the identification basis so that they can achieve relatively high precision. However, high-resolution images mean more computation cost in the recognition process, which usually cannot be guaranteed in mobile computing. Therefore, this paper proposes an improved low-resolution palmprint image recognition method based on residual networks. The main contributions include: 1) We introduce a channel attention mechanism to refactor the extracted feature maps, which can pay more attention to the informative feature maps and suppress the useless ones. 2) The ResStage group structure proposed by us divides the original residual block into three stages, and we stabilize the signal characteristics before each stage by means of BN normalization operation to enhance the feature channel. Comparison experiments are conducted on a public dataset provided by the Hong Kong Polytechnic University. Experimental results show that the proposed method achieve a rank-1 accuracy of 98.17% when tested on low-resolution images with the size of 12dpi, which outperforms all the compared methods obviously.

• Physical Therapy Rehabilitation Science
• /
• v.11 no.1
• /
• pp.105-112
• /
• 2022

Objective: Nonspecific low back pain (NS-LBP) causes pain and disability, affecting the neuromuscular system and altering gait patterns. The purpose of this study is to investigate the effect of improvement of low back pain symptoms through physical therapy on foot pressure and spatiotemporal gait parameters. Design: A pilot study. Methods: Participants received manual therapy and supervised therapeutic exercise, which consisted of 12 sessions for 6 weeks. Participants were assessed for pain intensity (a numeric pain rating scale), disability index (oswestry disability index), and spatiotemporal gait parameters before and after intervention. Wilcoxon signed rank test was used to analyze the before-and-after differences in a single group. Results: All seven NS-LBP patients completed the study without dropout. After six weeks of physical therapy, the numeric pain rating scale and oswestry disability index showed significant improvement (Z= -2.388, P=0.017). There was no significant improvement in both static and dynamic conditions in foot pressure (P>0.05). However, in the spatiotemporal gait parameters, there were significant differences in all variables except the right stance phase and left mid stance (P<0.05). Conclusions: In our pilot study, 12 sessions of physical therapy in NS-LBP patients improved gait quilty in spatiotemporal gait parameters. Similarly, it has resulted in clinically positive improvements in pain and disability.

• KDI Journal of Economic Policy
• /
• v.30 no.2
• /
• pp.61-97
• /
• 2008

This paper aims at analyzing the practice of job rotation in the Korean government and putting forward policy suggestions. The Korean government is often accused of low capacity and weak competitiveness, which mainly result from the low expertise of public officials. Considering the high quality of human resources flowing into the public sector in Korea, solutions should be found from the structure of the system. This paper regards frequent position changes due to excessive job rotation as a key factor undermining the accumulation of expertise and conducts in-depth analyses. The current practice of frequent rotation shows that the average tenure period of government officials at director level and above is only about one year, far shorter than those in major developed countries, which causes many problems such as low efficiency, lack of accountability and policy consistency, and low opportunity for accumulating expertise. Simple models are set up to analyze job rotation and other alternative personnel management systems. Analyses find that it would be desirable to have each individual experience various positions during the initial rotation period to find his/her own aptitude, and then accumulate expertise by settling in at a certain specialized field for a prolonged period of time based on the revealed aptitude in mid and high positions. This turns out to be in line with the structure of the Career Development Program which is being introduced. The model-based analysis of this paper distinguishes this study from preceding ones conducted in the traditional framework of personnel management study. Practical measures to mitigate the problems of frequent job rotation include rotating within the area of specialty, narrowing the scope of transfer, and reinforcing the minimum tenure period. However, since the current frequent rotation is fundamentally attributable to the rank system based on seniority, the present civil service classification system needs to be converted into a position classification system in the long run.

• Journal of Korean Medicine Rehabilitation
• /
• v.28 no.4
• /
• pp.21-32
• /
• 2018

Objectives This study was designed to investigate the effects of wet cupping on Wisu (BL12) in non-acute low back pain patients. Methods We recruited 30 participants for this study. Fifteen patients were randomly assigned to the Wisu (BL21) treatment group (WT group) and 15 were assigned to the non-acupoint treatment group (NT group). Both groups were treated with the pricking-cupping bloodletting method three times. Values at baseline and follow-up were analyzed by Wilcoxon signed rank test and the differences between the two groups were determined by Wilcoxon rank sum test. p-values less than 0.05 were considered significant. The primary outcome was the visual analogue scale (VAS), and secondary outcomes were the Oswestry disability index (ODI), Rolland-Morris disability questionnaire (RMDQ), Euroqol-5 dimension questionnaire (EQ-5D) and finger-to-ground distance (FTGD). These outcomes were measured on the day of first treatment before the procedure and on follow-up 7 days after the last treatment. Results Significant changes were identified in the VAS for pain and ODI in each group after wet cupping treatment on Wisu (p<0.05). However, no significant changes were found between groups. Meanwhile, RMDQ and EQ-5D were significantly decreased only in the NT group (p<0.05) without any differences between groups. FTGD was decreased in both groups, but not significantly. Conclusions Wet cupping with both Wisu treatment and non-acupoint had significant effects on non-acute low back pain, although there were no differences between the two groups. A large-scale study is needed to identify the effect of wet cupping on Wisu.

• ETRI Journal
• /
• v.40 no.4
• /
• pp.421-434
• /
• 2018

The application of deep neural networks (DNNs) to connect the world with cyber physical systems (CPSs) has attracted much attention. However, DNNs require a large amount of memory and computational cost, which hinders their use in the relatively low-end smart devices that are widely used in CPSs. In this paper, we aim to determine whether DNNs can be efficiently deployed and operated in low-end smart devices. To do this, we develop a method to reduce the memory requirement of DNNs and increase the inference speed, while maintaining the performance (for example, accuracy) close to the original level. The parameters of DNNs are decomposed using a hybrid of canonical polyadic-singular value decomposition, approximated using a tensor power method, and fine-tuned by performing iterative one-shot hybrid fine-tuning to recover from a decreased accuracy. In this study, we evaluate our method on frequently used networks. We also present results from extensive experiments on the effects of several fine-tuning methods, the importance of iterative fine-tuning, and decomposition techniques. We demonstrate the effectiveness of the proposed method by deploying compressed networks in smartphones.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.5
• /
• pp.2141-2155
• /
• 2020

In order to meet the requirements of background change, illumination variation, moving shadow interference and high accuracy in object detection of moving camera, and strive for real-time and high efficiency, this paper presents an object detection algorithm based on sparse approximation recursion and sparse coding migration in subspace. First, low-rank sparse decomposition is used to reduce the dimension of the data. Combining with dictionary sparse representation, the computational model is established by the recursive formula of sparse approximation with the video sequences taken as subspace sets. And the moving object is calculated by the background difference method, which effectively reduces the computational complexity and running time. According to the idea of sparse coding migration, the above operations are carried out in the down-sampling space to further reduce the requirements of computational complexity and memory storage, and this will be adapt to multi-scale target objects and overcome the impact of large anomaly areas. Finally, experiments are carried out on VDAO datasets containing 59 sets of videos. The experimental results show that the algorithm can detect moving object effectively in the moving camera with uniform speed, not only in terms of low computational complexity but also in terms of low storage requirements, so that our proposed algorithm is suitable for detection systems with high real-time requirements.

• The Journal of Korean Physical Therapy
• /
• v.20 no.4
• /
• pp.21-28
• /
• 2008

Purpose: We measured the Oswestry Low Back Pain Disability Index (OLBPDI) and related factors in patients with low back pain. Methods: The sample consisted of 50 patients who received physical therapy at the physical therapy units of the Andong Seoul Sintong Clinic, St. Luke Clinic, and Yeongju Seoul Sintong Clinic in Andong and Yeongju city from October, 2007, to February, 2008. The OLBPDI questionnaire was administered by 5 physical therapists as a cross-sectional study. Student's t-test and analysis of variance (ANOVA/Tukey and Scheffe) were used to analyze OLBPDI score differences. We also used nonparametric statistic analysis (Wilcoxon rank sum test, Median test). Pearson correlation analysis (Spearman correlation analysis) was used to analyze the relationship between OLBPDI and the visual analogue scale (VAS). Multiple regression analysis was performed to determine the effects of independent variables on pain scores as defined by the OLBPDI. Results: The average patient age was 37.1 years (range: 18$\sim$78 years old), and time from onset was 21.7 months (1$\sim$180). OLBPD and VAS scores were 12.70 (3.0$\sim$28.0) and 5.14 (1$\sim$8), respectively. OLBPDI scores were 14.4 in patients taking medicine and 11.57 in those who did not. There was a statistically significant relationship between OLBPDI and VAS (r=0.54, p=0.0001; r=0.55, p=0.0001 by Spearman coefficient). Gender ($\beta$=6.14, p=0.0124), age ($\beta$=-2.01, p=0.0324), weight ($\beta$=0.31, p=0.0222), time from onset ($\beta$=1.54, p=0.0044), and VAS score ($\beta$=1.59, p=0.0004) were significantly associated with OLBPD by multiple regression analysis. Conclusion: Variables associated with OLBPD were gender, age, weight, time from onset, and VAS score. Collecting information on the pain index using OLBPDI was acceptable to patients with low back pain. Further research should explore the pain index by using larger sample sizes and longer follow-up periods.