• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.9
• /
• pp.1049-1059
• /
• 2019

We propose a learning-based image segmentation algorithm. Starting from super-pixels, our method learns the probability of merging two regions based on the ground truth made by humans. The learned information is used in determining whether the two regions should be merged or not in a segmentation stage. Unlike exiting learning-based algorithms, we use both local and object information. The local information represents features computed from super-pixels and the object information represent high level information available only in the learning process. The object information is considered as privileged information, and we can use a framework that utilize the privileged information such as SVM+. In experiments on the Berkeley Segmentation Dataset and Benchmark (BSDS 500) and PASCAL Visual Object Classes Challenge (VOC 2012) data set, out model exhibited the best performance with a relatively small training data set and also showed competitive results with a sufficiently large training data set.

• The Korean Journal of Applied Statistics
• /
• v.25 no.1
• /
• pp.163-181
• /
• 2012

This paper studies the effect of job training for non-regular employees in the Korea labor market. Using an economically active population data set of statistics Korea, we apply a non-parametric matching and sensitivity analysis method to measure the effect of the training for non-regular employees and to look for the impact of an unobservable variable or confounding factor in regards to the selection effect and outcome effect. In the our empirical results, we conclude that the effect of the training for non-regular employees has a better employment effect for getting a regular job rather than a wage effect; in addition, the impact of unobservable variables or confounding factors do not exercise a statistically strong influence on the baseline ATT.

• The Korean Journal of Food & Health Convergence
• /
• v.7 no.4
• /
• pp.31-41
• /
• 2021

This study conducted 8 weeks of foam roller & mobility, core, and weight training for 9 middle school soccer players 5 times a week for 90 minutes to determine the effect on changes in body composition, thigh circumference, basic physical strength, and functional movement. To analyze the data according to the results of this study, SPSS 25.0 statistical program was used for analysis, and the mean (M) and standard deviation (SD) were calculated to present the descriptive statistics of all dependent variables. In addition, to analyze the difference between before and after exercise, it was verified using a paired t test. The statistical significance level (α) in all reasoning statistics was set to less than 5%. As a result, there were no significant changes in body composition and thigh circumference through 8 weeks of training. However, there were significant changes in agility and muscular endurance, and the total score for functional movement changes was significantly increased, and statistically significant changes were observed in three variables. Therefore, complex training is effective in changing the physical strength and functional movement of middle school soccer players, and further research will be required for a control group and various complex exercises.

• International Journal of Computer Science & Network Security
• /
• v.22 no.10
• /
• pp.107-112
• /
• 2022

Nowadays clusters are recognized as an important instrument for promoting industrial development, innovation, competitiveness and growth. An educational cluster is a set of interrelated vocational educational institutions of various levels that are united by industry with each other and are connected by partnership with industry enterprises. This article attempts to develop and describe cluster model of university students' multilingual training. The purpose of this study is to describe multilingual training of university students and their polycultural competencies formation and to define the process of multilingual training in form of a cluster. The authors consider clusters as an integral part of the educational campus within the concept framework of Shadrinsk State Pedagogical University. To determine the essence of the concept of a cluster model of university students' multilingual training, theoretical, empirical, observational, and diagnostic methods were implemented, such as a review of scientific literature, a compilation of best practices, observation, statistical methods, etc. The authors analyzed the programs of partner universities and organized international webinars and internships for bachelors and masters abroad and developed online courses "Foreign language for undergraduate students and masters". Experimental data obtained during the implementation of cluster training show the effectiveness of the formation of students' polycultural competencies.

• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.98-102
• /
• 2023

This paper presents a study on how augmenting semi-synthetic image data improves the performance of human detection algorithms. In the field of object detection, securing a high-quality data set plays the most important role in training deep learning algorithms. Recently, the acquisition of real image data has become time consuming and expensive; therefore, research using synthesized data has been conducted. Synthetic data haves the advantage of being able to generate a vast amount of data and accurately label it. However, the utility of synthetic data in human detection has not yet been demonstrated. Therefore, we use You Only Look Once (YOLO), the object detection algorithm most commonly used, to experimentally analyze the effect of synthetic data augmentation on human detection performance. As a result of training YOLO using the Penn-Fudan dataset, it was shown that the YOLO network model trained on a dataset augmented with synthetic data provided high-performance results in terms of the Precision-Recall Curve and F1-Confidence Curve.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.3
• /
• pp.195-200
• /
• 2004

In this paper we propose effective speech recognizer through two recognition experiments. In general, SVM is classification method which classify two class set by finding voluntary nonlinear boundary in vector space and possesses high classification performance under few training data number. In this paper we compare recognition performance of HMM and SVM at training data number and investigate recognition performance of each feature parameter while changing feature space of MFCC using Independent Component Analysis(ICA) and Principal Component Analysis(PCA). As a result of experiment, recognition performance of SVM is better than 1:.um under few training data number, and feature parameter by ICA showed the highest recognition performance because of superior linear classification.

• ETRI Journal
• /
• v.39 no.5
• /
• pp.643-651
• /
• 2017

Human age estimation is one of the key factors in the field of Human-Robot Interaction/Human-Computer Interaction (HRI/HCI). Owing to the development of deep-learning technologies, age recognition has recently been attempted. In general, however, deep learning techniques require a large-scale database, and for age learning with variations, a conventional database is insufficient. For this reason, we propose an age estimation method using artificially generated data. Image data are artificially generated through 3D information, thus solving the problem of shortage of training data, and helping with the training of the deep-learning technique. Augmentation using 3D has advantages over 2D because it creates new images with more information. We use a deep architecture as a pre-trained model, and improve the estimation capacity using artificially augmented training images. The deep architecture can outperform traditional estimation methods, and the improved method showed increased reliability. We have achieved state-of-the-art performance using the proposed method in the Morph-II dataset and have proven that the proposed method can be used effectively using the Adience dataset.

• Journal of KIISE:Software and Applications
• /
• v.32 no.9
• /
• pp.912-919
• /
• 2005

When we apply binary classification to multi-class classification for text categorization, we use the One-Against-All method generally, However, this One-Against-All method has a problem. That is, documents of a negative set are not labeled by human. Thus, they can include many noisy documents in the training data. In this paper, we propose that the Sliding Window technique and the EM algorithm are applied to binary text classification for solving this problem. We here improve binary text classification through extracting noise documents from the training data by the Sliding Window technique and re-assigning categories of these documents using the EM algorithm.

• Korean Security Journal
• /
• no.9
• /
• pp.201-235
• /
• 2005

This study is designed to contribute for development of Private Security Business by fact-finding in instruction and training of private security guard serviced in this realm and domestic and foreign guard service and modeling effective and rational instruction and training program based on drawn problem. For this study, basically I collected and analyzed documents, theses, and papers of the inside and outside of the country. For practical use of data, I used materials of private security related institutes and police agency. And for private security educating training programs of the inside and outside of the country, I collected materials on internet, and with the help of police agency and interpol. For korean private security company's educating training programs, I made a study with the interview of private security company's businessmen. This study's conclusion is as follows. In a domestic private security enterprise, when set theory instruction minimize instruction and training program and must set up instruction and training program as practical affairs center enemy instruction, and theory instruction must be composed for instruction me that it is connected to practical affairs instruction too. The instruction course of private security guard instruction and training program composed with a security outline, a security plan, an information-gathering, civilian expenses, a security way, terror and terrorism, a related law, security trial, electronic security, a security analysis technique, company introduction, instruction and training program about a professional tube with theory instruction. Practical affairs instruction composed with the selection and a preventive security, close contact attendance security, vehicle security, security driving the security martial arts and self-protection liquor, first aid, security equipment, a gun and shooting, a security protocol, customer satisfaction, facilities security and expenses, a fire fighting instruction, teamwork training, explosive and a dangerous substance, physical strength, a documentation practical affairs, service, instruction and training program about foreigh language instruction.

• PNF and Movement
• /
• v.10 no.1
• /
• pp.61-70
• /
• 2012

Purpose : The purpose of this study was to investigate thigh muscle-bone CSA and leg strength during low-intensity exercise program with leg blood flow restriction by external compression to reduce muscle outflow. Methods : Eighteen health students gave informed written consent to participate in this investigation. An occlusion cuff was attached to the proximal end of the leg so that blood flow was reduced during the training. The training was conducted one times a day, three times a week, for 8 weeks using one sets of 30 minutes. The training program performed to squat with standing, lunge with standing and heel raise with one leg standing. Measurements of thigh muscle-bone CSA(cross-sectional area) and leg strength were evaluated pre and post-training. Statistical evaluation of these data was accomplished utilizing a paired t-test by SPSS 12.0 program for windows. Significance level was set at p <.05. Results : All data are reported as means and standard deviations(SD) for all variables. The result of the study is followed; After the training, muscle-bone CSA, gluteus maximus m, quadriceps m, hamstring m of both legs were significantly improved but not calf muscle(p<.05). There was no significant difference of change quantity between muscle-bone CSA and leg strength in Lt. and Rt. side. But the variation in leg muscle strength of Rt. leg(dominant) was much more increased than Lt. leg(non-dominant) after 8 weeks training. Conclusion : Low-intensity training with leg blood flow restriction offers a potentially useful method for improving leg muscle strength.