• Korean Institute of Interior Design Journal
• /
• v.23 no.3
• /
• pp.117-124
• /
• 2014

This study has inferred the mechanism of psychological observation activities through comparison analysis of the observation data acquired from eye-tracking and their post-estimation. The results of their analysis can be summarized as the followings. First, even though the frame of analysis has been set up so that there might not be any change to the number of the sections even with any change of consecutive observation times, the fact that the time by area decreases along with the change of consecutive observation from three times to six and nine times means that the time spent on "recognition" of space information reduces in the course that the feature of observing for space information switches from "perception to recognition". Second, the subjects moves their eyes incessantly in order to acquire space information while observing the space, when it was confirmed that there was a difference between "the space which the subjects searched for information by means of observation activities" and "that which they thought they observed that remaining in their consciousness". The appreciation of this kind of difference is very significant for the analysis of observation features. Third, the short observation (0.1 second, three times of consecutive observations) is consistent with "Ares I, intensively searched = that marked as having been observed consciously" by 60%, while the long-time observation (0.3 second, 9 times of consecutive observations) had 56%, which was relatively high, of "Area I, searched intensively ${\neq}$ that marked as having been observed consciously", which means that the observation feature seen at the activities of "consciousness : unconsciousness" and "observation : search" had some change in the course of changing from "perception to recognition".

• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.1
• /
• pp.10-17
• /
• 2017

We propose classification algorithms for human and dog movement. The proposed algorithms use micro-Doppler signals obtained from humans and dogs moving in four different directions. A two-stage classifier based on a support vector machine (SVM) is proposed, which uses a radial-based function (RBF) kernel and $16^{th}$-order linear predictive code (LPC) coefficients as feature vectors. With the proposed algorithms, we obtain the best classification results when a first-level SVM classifies the type of movement, and then, a second-level SVM classifies the moving object. We obtain the correct classification probability 95.54% of the time, on average. Next, to deal with the difficult classification problem of human and dog running, we propose a two-layer convolutional neural network (CNN). The proposed CNN is composed of six ($6{\times}6$) convolution filters at the first and second layers, with ($5{\times}5$) max pooling for the first layer and ($2{\times}2$) max pooling for the second layer. The proposed CNN-based classifier adopts an auto regressive spectrogram as the feature image obtained from the $16^{th}$-order LPC vectors for a specific time duration. The proposed CNN exhibits 100% classification accuracy and outperforms the SVM-based classifier. These results show that the proposed classifiers can be used for human and dog classification systems and also for classification problems using data obtained from an ultra-wideband (UWB) sensor.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.30 no.1
• /
• pp.1-6
• /
• 2016

In recent years, there were frequent exogenous contagious diseases in Eastasia like SARS(severe acute respiratory syndrome), Avian influenza, Swine influenza, MERS etc. But there are various interpretations about their pathological differentiations and lead to controversy to diagnosis and medicinal use. So there needs universal and consistent understanding methods. Several conclusions are obtained from the research on differentiation theories of various epidemic diseases. Essential elements of differential diagnostic system are pathogen, characters and matters of disease and loci, especially three yin and three yang has close affinity with constitutional features or body shape. Binding these 3 categories, an integrated differentiation 3 dimensional coordinates are made. Out of these, each elements of 3 pathogen-axial lines are related with names of exogenous disease, and those of 3 feature-axial lines are related with 8 principal patterns. And those of 3 locus-axial lines implicating therapeutic method are related with steps and location of exterior and interior, 3 yin 3 yang, Defense, Qi, Nutrient and Blood, five viscera and six bowels and tissues. Additionally, 3 lines of each axis consist of factors which have their own affinity each other, so classification of pathogen, feature, locus of disease has layered interconnectedness. This classification system is included in constitutional features of individual patient. Afterwards, these cognitive structure can be used as a general theory guiding method of therapy, prevention and aftercure healthcare.

• Biomedical Science Letters
• /
• v.18 no.4
• /
• pp.399-405
• /
• 2012

The brain is the body's most organized and controlled organ, and it governs various psychological and mental functions. A brain abnormality could greatly affect one's physical and mental abilities, and consequently one's social life. Brain disorders can be broadly categorized into three main afflictions: stroke, brain tumor, and dementia. Among these, stroke is a common disease that occurs owing to a disorder in blood flow, and it is accompanied by a sudden loss of consciousness and motor paralysis. The main types of strokes are infarction and hemorrhage. The exact diagnosis and early treatment of an infarction are very important for the patient's prognosis and for the determination of the treatment direction. In this study, texture features were analyzed in order to develop a prototype auto-diagnostic system for infarction using computer auto-diagnostic software. The analysis results indicate that of the six parameters measured, the average brightness, average contrast, flatness, and uniformity show a high cognition rate whereas the degree of skewness and entropy show a low cognition rate. On the basis of these results, it was suggested that a digital CT image obtained using the computer auto-diagnostic software can be used to provide valuable information for general CT image auto-detection and diagnosis for pre-reading. This system is highly advantageous because it can achieve early diagnosis of the disease and it can be used as supplementary data in image reading. Further, it is expected to enable accurate medical image detection and reduced diagnostic time in final-reading.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.6 no.4
• /
• pp.1130-1147
• /
• 2014

Large size ships have a very flexible construction resulting in low resonance frequencies of the structural eigen-modes. This feature increases the dynamic response of the structure on short period waves (springing) and on impulsive wave loads (whipping). This dynamic response in its turn increases both the fatigue damage and the ultimate load on the structure; these aspects illustrate the importance of including the dynamic response into the design loads for these ship types. Experiments have been carried out using a segmented scaled model of a container ship in a Seakeeping Basin. This paper describes the development of the model for these experiments; the choice was made to divide the hull into six rigid segments connected with a flexible beam. In order to model the typical feature of the open structure of the containership that the shear center is well below the keel line of the vessel, the beam was built into the model as low as possible. The model was instrumented with accelerometers and rotation rate gyroscopes on each segment, relative wave height meters and pressure gauges in the bow area. The beam was instrumented with strain gauges to measure the internal loads at the position of each of the cuts. Experiments have been carried out in regular waves at different amplitudes for the same wave period and in long crested irregular waves for a matrix of wave heights and periods. The results of the experiments are compared to results of calculations with a linear model based on potential flow theory that includes the effects of the flexural modes. Some of the tests were repeated with additional links between the segments to increase the model rigidity by several orders of magnitude, in order to compare the loads between a rigid and a flexible model.

• International Journal of High-Rise Buildings
• /
• v.6 no.2
• /
• pp.165-175
• /
• 2017

The importance of designing urban building complexes so that they obtain 'urban' power, rather than become isolated from the surrounding urban context, has been well recognized by both researchers and practitioners. Nevertheless, most current discussions are made from architects' personal experiences and intuition, and lack a quantitative understanding, to which obstacles include an in-depth exploration of the 'urban' power between building complexes and the urban environment. This paper attempts to measure this feature of 'urban', i.e., 'urbanity,' through a new analytical approach derived from the opendata environment. Three measurements that can be easily collected though the Google Maps API and Open Street Map are applied herein to evaluate high or low values of urbanity. Specifically, these are 'metric depth', i.e., the scale of extended public space, 'development density', i.e., density and distribution of point of interests (POIs), and 'type diversity', i.e., diversity of different commercial types. Six cases located in Japan, China and Hong Kong respectively are ranked based on this analytical approach and compared with each other. It shows that Japanese cases, i.e., Osaka Station City and Namba Parks, Osaka, obtained clearly higher values than cases in Shanghai and Hong Kong. On one hand, the insight generated from measuring and explaining 'urban' power would help to assist better implementation of this feature in the design of urban building complexes. On the other hand, this analytical approach can be easily extended to achieve a large-scale measurement and comparison among different urban building complexes, which is also helpful for design practitioners.

• The KIPS Transactions:PartB
• /
• v.16B no.6
• /
• pp.471-478
• /
• 2009

The lung diseases classifying as one of the six incurable diseases in modern days are caused mostly by smoking and air pollution. Such causes the lung function damages, and results in malfunction of the exchange of carbon dioxide and oxygen in an alveolus, which the interest is augment with risk diseases of life prolongation. With this in the paper, we proposed a diagnosis method of lung diseases by applying parameters of voice analysis aiming at the getting the voice feature extraction. Firstly, we sampled the voice data from patients and normal persons in the same age and sex, and made two sample groups from them. Also, we conducted an analysis by applying the various parameters of voice analysis through the collected voice data. The relational significance between the patient and normal groups can be evaluated in terms of speech rates and intensity as a part of analized parameters. In conclusion, the patient group has shown slower speech rates and bigger intensity than the normal group. With this, we propose the method of voice feature extraction for lung diseases.

• Journal of the Korean Wood Science and Technology
• /
• v.37 no.6
• /
• pp.556-565
• /
• 2009

Automated visual grading system of lumber needs correct input image. In order to create a correct image of domestic red pine lumber 3.6 m long feeding on a conveyer, part images were captured using area sensor and template matching algorithm was applied to merge part images. Two kinds of template matching algorithms and six kinds of template sizes were adopted in this operation. Feature extracted method appeared to have more excellent image merging performance than fixed template method. Error length was attributed to a decline of similarity related by difference of partial brightness on a part image, specific pattern and template size. The mismatch part was repetitively generated at the long grain. The best size of template for image merging was $100{\times}100$ pixels. In a further study, assignment of exact template size, preprocessing of image merging for reduction of brightness difference will be needed to improve image merging.

• The Journal of the Convergence on Culture Technology
• /
• v.5 no.4
• /
• pp.413-420
• /
• 2019

The purpose of the present study is to develop a model for predicting hypercholesterolemia using an integrated set of body fat mass variables based on machine learning techniques, beyond the study of the association between body fat mass and hypercholesterolemia. For this study, a total of six models were created using two variable subset selection methods and machine learning algorithms based on the Korea National Health and Nutrition Examination Survey (KNHANES) data. Among the various body fat mass variables, we found that trunk fat mass was the best variable for predicting hypercholesterolemia. Furthermore, we obtained the area under the receiver operating characteristic curve value of 0.739 and the Matthews correlation coefficient value of 0.36 in the model using the correlation-based feature subset selection and naive Bayes algorithm. Our findings are expected to be used as important information in the field of disease prediction in large-scale screening and public health research.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.4
• /
• pp.1-11
• /
• 2021

Recently, the importance of preventive maintenance has been emerging since failures in a complex system are automatically detected due to the development of artificial intelligence techniques and sensor technology. Therefore, prognostic and health management (PHM) is being actively studied, and prediction of the remaining useful life (RUL) of the system is being one of the most important tasks. A lot of researches has been conducted to predict the RUL. Deep learning models have been developed to improve prediction performance, but studies on identifying the importance of features are not carried out. It is very meaningful to extract and interpret features that affect failures while improving the predictive accuracy of RUL is important. In this paper, a total of six popular deep learning models were employed to predict the RUL, and identified important variables for each model through SHAP (Shapley Additive explanations) that one of the explainable artificial intelligence (XAI). Moreover, the fluctuations and trends of prediction performance according to the number of variables were identified. This paper can suggest the possibility of explainability of various deep learning models, and the application of XAI can be demonstrated. Also, through this proposed method, it is expected that the possibility of utilizing SHAP as a feature selection method.