• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.2
• /
• pp.103-108
• /
• 2022

In this paper, we propose a new image analysis method based on regression manifold 3-D PCA. The proposed method is a new image analysis method consisting of a regression analysis algorithm with a structure designed based on an autoencoder capable of nonlinear expansion of manifold 3-D PCA and PCA for efficient dimension reduction when entering large-capacity image data. With the configuration of an autoencoder, a regression manifold 3-DPCA, which derives the best hyperplane through three-dimensional rotation of image pixel values, and a Bayesian rule structure similar to a deep learning structure, are applied. Experiments are performed to verify performance. The image is improved by utilizing the fine dust image, and accuracy performance evaluation is performed through the classification model. As a result, it can be confirmed that it is effective for deep learning performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.4
• /
• pp.1721-1737
• /
• 2020

Human apparent age estimation has become a popular research topic and attracted great attention in recent years due to its wide applications, such as personal security and law enforcement. To achieve the goal of age estimation, a large number of methods have been pro-posed, where the models derived through the cumulative attribute coding achieve promised performance by preserving the neighbor-similarity of ages. However, these methods afore-mentioned ignore the geometric structure of extracted facial features. Indeed, the geometric structure of data greatly affects the accuracy of prediction. To this end, we propose an age estimation algorithm through joint feature selection and manifold learning paradigms, so-called Feature-selected and Geometry-preserved Least Square Regression (FGLSR). Based on this, our proposed method, compared with the others, not only preserves the geometry structures within facial representations, but also selects the discriminative features. Moreover, a deep learning extension based FGLSR is proposed later, namely Feature selected and Geometry preserved Neural Network (FGNN). Finally, related experiments are conducted on Morph2 and FG-Net datasets for FGLSR and on Morph2 datasets for FGNN. Experimental results testify our method achieve the best performances.

• Nuclear Engineering and Technology
• /
• v.49 no.4
• /
• pp.817-824
• /
• 2017

Residual stress is a critical element in determining the integrity of parts and the lifetime of welded structures. It is necessary to estimate the residual stress of a welding zone because residual stress is a major reason for the generation of primary water stress corrosion cracking in nuclear power plants. That is, it is necessary to estimate the distribution of the residual stress in welding of dissimilar metals under manifold welding conditions. In this study, a cascaded support vector regression (CSVR) model was presented to estimate the residual stress of a welding zone. The CSVR model was serially and consecutively structured in terms of SVR modules. Using numerical data obtained from finite element analysis by a subtractive clustering method, learning data that explained the characteristic behavior of the residual stress of a welding zone were selected to optimize the proposed model. The results suggest that the CSVR model yielded a better estimation performance when compared with a classic SVR model.

• Journal of Korean Society for Atmospheric Environment
• /
• v.19 no.6
• /
• pp.719-731
• /
• 2003

In order to maintain and manage ambient air quality, it is necessary to identify sources and to apportion its sources for ambient particulate matters. The receptor methods were one of the statistical methods to achieve reasonable air pollution strategies. Also, receptor methods, a field of chemometrics, is based on manifold applied statistics and is a statistical methodology that analyzes the physicochemical properties of gaseous and particulate pollutant on various atmospheric receptors, identifies the sources of air pollutants, and quantifies the apportionment of the sources to the receptors. The objective of this study was 1) after obtaining results from the PMF modeling, the existing sources of air at the study area were qualitatively identified and the contributions of each source were quantitatively estimated as well. 2) finally efficient air pollution management and control strategies of each source were suggested. The PMF model was intensively applied to estimate the quantitative contribution of air pollution sources based on the chemical information (128 samples and 25 chemical species). Through a case study of the PMF modeling for the PM-10 aerosols, the total of 11 factors were determined. The multiple linear regression analysis between the observed PM-10 mass concentration and the estimated G matrix had been performed following the FPEAK test. Finally the regression analysis provided quantitative source contributions (scaled G matrix) and source profiles (scaled F matrix). The results of the PMF modeling showed that the sources were apportioned by secondary aerosol related source 28.8 ％, soil related source 16.8％, waste incineration source 11.5%, field burning source 11.0%, fossil fuel combustion source 10%, industry related source 8.3％, motor vehicle source 7.9%, oil/coal combustion source 4.4％, non-ferrous metal source 0.3％. and aged sea- salt source 0.2％, respectively.

• The Korean Journal of Elementary Counseling
• /
• v.5 no.1
• /
• pp.205-226
• /
• 2006

The purpose of this study was to find out the relations between 6th graders' negative cognitive process and anger experience and aggressiveness. To achieve the goal, it conducted a test to examine children's negative cognitive process, anger experience, aggressiveness targeting 100 children of 6th grade in C elementary school, Gyeonggj province. Then it conducted SPSS 12.0 statistical program to get the results of correlation analysis and regression analysis. The outcomes were as follows. First, there was a meaningfully positive relation between a negative cognitive process and anger experience. In other words, children having more negative cognitive process seemed to experience the feelings of anger more often, this presented the important role of cognition while getting into a temper Second, it reported a positive relation between anger experience and aggressiveness. Children who have experienced anger more often showed more violent behaviors, especially there were more significant positive relations between trait-anger and aggressiveness compared to state-anger and aggressiveness. This could explain some possibilities that children with high level of trait-anger might outrage more often than others by recognizing the situations as anger stimulants. Third, when conducting a regression analysis, a negative cognitive process made an effect on anger experience which affected aggressiveness. However, it did not show a negative cognitive process making a direct effect on aggressiveness. This is considered that children could experience an anger while evaluating an event or object in a negative way based on individual belief, and emotional linguistic behavioral aggressiveness would be formalized as they express the sparked fury either internally or externally. In conclusion, this study proved that there were close relations between children's negative cognitive process and anger experience and aggressiveness. A negative cognitive process affects anger experience, and anger experience affects aggressiveness afterwards. A negative cognitive process affects aggressiveness through anger experience indirectly, and especially trait-anger among anger experience is the main factor to influence on aggressiveness. With consideration of these results, it is believed that mediation is important key to moderate the negative cognition and trait-anger in order to diminish children's aggressive behaviors. This study has a meaning to provide searching for manifold mediating methods between negative cognition and trait anger, with a fundamental resource.

• Journal of Digital Contents Society
• /
• v.15 no.6
• /
• pp.737-743
• /
• 2014

For single image super resolution (SR), interpolation based and example based algorithms are extensively used. The interpolation algorithms have the strength of theoretical simplicity. However, those algorithms are tending to produce high resolution images with jagged edges, because they are not able to use more priori information. Example based algorithms have been studied in the past few years. For example based SR, the nearest neighbor based algorithms are extensively considered. Among them, neighbor embedding (NE) has been inspired by manifold learning method, particularly locally linear embedding. However, the sizes of local training sets are always too small. So, NE algorithm is weak in the performance of the visuality and quantitative measure by the poor generalization of nearest neighbor estimation. An improved NE algorithm with Support Vector Regression (SVR) was proposed to solve this problem. Given a low resolution image, the pixel values in its high resolution version are estimated by the improved NE. Comparing with bicubic and NE, the improvements of 1.25 dB and 2.33 dB are achieved in PSNR. Experimental results show that proposed method is quantitatively and visually more effective than prior works using bicubic interpolation and NE.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.12
• /
• pp.2946-2952
• /
• 2014

Among the Example based Super Resolution(SR) techniques, Neighbor embedding(NE) has been inspired by manifold learning method, particularly locally linear embedding. However, the poor generalization of NE decreases the performance of such algorithm. The sizes of local training sets are always too small to improve the performance of NE. We propose the Learning Sparse-Neighbor Image Representation baesd on SVR having an excellent generalization ability to solve this problem. Given a low resolution image, we first use bicubic interpolation to synthesize its high resolution version. We extract the patches from this synthesized image and determine whether each patch corresponds to regions with high or low spatial frequencies. After the weight of each patch is obtained by our method, we used to learn separate SVR models. Finally, we update the pixel values using the previously learned SVRs. Through experimental results, we quantitatively and qualitatively confirm the improved results of the proposed algorithm when comparing with conventional interpolation methods and NE.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.733-735
• /
• 2014

Example based Super Resolution(SR) is using the correspondence between the low and high resolution image from a database. This method uses only one image to estimate a high resolution image and can get the larger image than 2 times. Example based SR is proposed to solve the problem of classical SR. Neighbor embedding(NE) has been inspired by manifold learning method, particularly locally linear embedding. However, the poor generalization of NE decreases the performance of such algorithm. The sizes of local training sets are always too small to improve the performance of NE. We propose the advanced NE baesd on SVR having an excellent generalization ability to solve this problem. Given a low resolution image, we estimate a pixel in its high resolution version by using SVR based NE. Through experimental results, we quantitatively and qualitatively confirm the improved results of the proposed algorithm when comparing with conventional interpolation methods and NE.

• Journal of Distribution Science
• /
• v.12 no.12
• /
• pp.119-130
• /
• 2014

Purpose - Franchise industries are significant both socially and economically. However, with increasing interest, there are manifold problems. It is necessary to seek measures for mature operation constantly despite unprepared franchisors, negative perceptions of the media and society toward franchise, and rapid changes in business start-up trends and propensity to consume that make business start-ups difficult. The paper aims to explain the effects of relational commitment and performance on the franchisor's management characteristics and justice in the food service franchise system. Research design, data, and methodology - This is an exploratory survey examining franchising in Korea. Based on a literature synthesis, we extract five constructs: managerial characteristics, support, fairness, trust, and satisfaction. We hypothesize that these factors influence the trust, satisfaction, and performance of franchisees. To examine these hypotheses empirically, we conducted a survey on the database of the Franchising Council of Korea. The study employs data from May to September 2014. In total, 135 completed questionnaires were received, of which 128 were usable. The data was analyzed with SPSS/PC 22.0. First, to test unidimensionality and nomological validity of the measures of each construct, we employed a scale refinement procedure. The result of a reliability test with Cronbach's α and factor analysis warranted unidimensionality of the measures for each construct. In addition, nomological validity of the measures was warranted from the result of the correlation and regression analysis. By analyzing the data, we can confirm most hypotheses. Results - Frist, franchisor characteristics have a positive effect on trust and satisfaction. Second, franchisor fairness has an effect on trust and satisfaction. Third, franchisor support has an effect on satisfaction. Further, the franchisee trust has an effect on satisfaction. Fourth, the satisfaction of a franchisee with a franchisor affects the performance of a franchisee. Finally, there is a possibility that not only franchisee performance but also increasing the credibility and improving the image of the franchisor through communication between franchisor and franchisee can improve franchisees' performance and satisfaction by motivating the franchisee for sustainable growth. Franchisers should endeavor for franchisees to obtain stable revenue with continuous and practical support. They should recognize that they can expand their business by supporting their franchisees. Franchisors should not only instantly respond to franchisees' troubles with interactive communications but also raise the ability of supervisors for better support. Franchisors should share their visions and goals with their franchisees and provide systematic and continuous support based on trust and clear company management. Franchisees should understand franchisors' position as well as participate in establishing the basic franchise system. Contributions - The paper contributes to understanding franchising in Korea. It offers insights and assistance to franchisors hoping to start franchises. This paper explores measurement issues related to franchisee performance by estimating its determinant factors (managerial characteristics, support, fairness, trust, and satisfaction). This study provides franchisors and practitioners planning to extend their franchising business with some practical knowledge.