• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.11
• /
• pp.114-122
• /
• 2016

EIS (Electro interstitial scan) is a non-invasive and simple method to find the physio-pathological information inferred by electric current response with respect to low direct current applied between remote sites of the body. Although a few EIS-based devices for diagnosing diabetes were commercialized, they were not successful in offering clinical validity nor in confirming diagnostic principle. In this study, we measured the voltage responses of diabetic patients and normal subjects with a commercialized EIS device to test the usefulness of EIS in screening diabetes. For this purpose, voltage was measured between pairs of electrodes contacted at both palm, both soles of the feet and left and right forehead above both eyes. After feature extraction of voltage signals, the AUC (area under the curve) between the two groups was calculated and we found that seven variables were appropriately shown above 60% of accuracy. In addition, we applied the k-NN (k-nearest neighbors) method and found that the accuracy of classification between the two groups reached the accuracy of 76.2%. This result implies that the voltage response analysis based on EIS has potential as a diabetics screening method.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.4
• /
• pp.392-397
• /
• 2014

Recently, there are many researches on active safety system of intelligent vehicle. To reduce the probability of collision caused by driver's inattention and mistakes, the active safety system gives warning or controls the vehicle toward avoiding collision. For the purpose, it is necessary to recognize and analyze circumstances around. In this paper, we will treat the problem about collision risk assessment. In general, it is difficult to calculate the collision risk before it happens. To consider the uncertainty of the situation, Monte Carlo simulation can be employed. However it takes long computation time and is not suitable for practice. In this paper, we apply neural networks to solve this problem. It efficiently computes the unseen data by training the results of Monte Carlo simulation. Furthermore, we propose the features affects the performance of the assessment. The proposed algorithm is verified by applications in various crash scenarios.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.6
• /
• pp.1506-1514
• /
• 2015

In this paper, we proposed a forest fires prediction and detection system. It could provide a situation of fire prediction and detection methods using context awareness sensor. A fire occurs wide range of sensing a fire in a single camera sensor, it is difficult to detect the occurrence of a fire. In this paper, we propose an algorithm for real-time by using a temperature sensor, humidity, Co2, the flame presence information acquired and comparing the data based on multiple conditions, analyze and determine the weighting according to fire in complex situations. In addition, it is possible to differential management of intensive fire detection and prediction for required dividing the state of fire zone. Therefore we propose an algorithm to determine the prediction and detection from the fire parameters as an temperature, humidity, Co2 and the flame in real-time by using a context awareness sensor and also suggest algorithm that provide the path of fire diffusion and service the secure safety zone prediction.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.30 no.12 s.159
• /
• pp.1768-1777
• /
• 2006

As the fashion industry comes under the influence of globalization throughout all fields of industry, the globalization and the market entry strategies are required for Korean fashion firms. This study attempted to analyze the factors influencing foreign entry mode of Korean fashion business based on Eclectic Theory. Data collection has been carried out from November 25 until December 25, 2005. The questionnaires were sent through e-mail or fax to 622 trading companies. 67 questionnaires were returned for a response rate of 10.7%. Of these returns, 61 usable questionnaires were employed for data analyses. Descriptive analysis, factor analysis, discriminant analysis, and t-test were used for data analysis. First, the most important venture motivation was price competitiveness and many firms were engaged in both production and sales in their target countries, which were mainly in Southeast Asia. Second, the firm's ability and experience were found out as ownership advantage factor, investment stability and market potential as location advantage factor, and contract stability as internalization advantage factor. Third, the result of discriminant analysis showed that location advantage factor was a significant factor in predicting the entry of fashion firms into foreign countries.

• Science of Emotion and Sensibility
• /
• v.14 no.3
• /
• pp.347-360
• /
• 2011

To find out characteristics of TV home shopping's virtual space deepness styling, this study had analysis characteristics of space deepness which is showed on flat TV screen with actually aired 6 dinnerware sales case, at $C^*$ home shopping, March.2005~November.2010, and survey consumer's emotional verbal image according to space styling character to get space deepness, which were shown on flat TV screen with camera angle, and research mutual relation with consumer preference. Also consumer's typical emotional verbal images for each space styling images for brand dinner ware sales had been extracted with reliability analysis, factor analysis, and multi dimensional scaling MDS using SPSS. Styling characteristics of space deepness were contrast of size, layering, vertical arrangement, and perspective arrangement, and used camera angles were bird's eye view, hi angle, and eye level. Result from the research is, highly marked consumer preferred styling material had a deep corelation to material's main factor and perceived emotional verbal images. Therefore this research could bring forward to new consumer preferred styling characteristics according to camera angle. Furthermore, it will be possible to make a study of preferred styling material through evaluation of quantitative spectators of in this area.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.2
• /
• pp.189-196
• /
• 2020

In general, fragmented files without signatures and file meta-information are difficult to recover. Multimedia files, in particular, are highly fragmented and have high entropy, making it almost impossible to recover with signature-based carving at present. To solve this problem, research on fragmented files is underway, but research on multimedia files is lacking. This paper is a study that classifies the types of fragmented multimedia files without signature and file meta-information. Extracts the characteristic values of each file type through the frequency differences of specific byte values according to the file type, and presents a method of designing the corresponding Gray-Scale table and classifying the file types of a total of four multimedia types, JPG, PNG, H.264 and WAV, using the CNN (Convolutional Natural Networks) model. It is expected that this paper will promote the study of classification of fragmented file types without signature and file meta-information, thereby increasing the possibility of recovery of various files.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.2
• /
• pp.25-31
• /
• 2011

Principal component analysis (PCA) is a well-known method for dimensionality reduction and feature extraction. Although PCA has been applied in many areas successfully, it is sensitive to outliers due to the use of sum-square-error. Several variants of PCA have been proposed to resolve the noise sensitivity and, among the variants, improved robust fuzzy PCA (RF-PCA2) demonstrated promising results. RF-PCA2, however, still can fall into a local optimum due to equal initial membership values for all data points. Another reason comes from the fact that RF-PCA2 is based on sum-square-error although fuzzy memberships are incorporated. In this paper, a variant of RF-PCA2 called RF-PCA3 is proposed. The proposed algorithm is based on the objective function of RF-PCA2. RF-PCA3 augments RF-PCA2 with the objective function of PCA and initial membership calculation using data distribution, which make RF-PCA3 to have more chance to converge on a better solution than that of RF-PCA2. RF-PCA3 outperforms RF-PCA2, which is demonstrated by experimental results.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2014.02a
• /
• pp.422.2-422.2
• /
• 2014

가스장 이온원(GFIS: Gas Field Ionization Source)은 전자현미경보다 분해능이 향상된 이온현미경의 광원으로 사용하기 위하여 연구되고 있고, 큰 각전류 밀도, 작은 크기의 가상 이온원 그리고 좁은 에너지 퍼짐을 특징으로 한다. 여러 가지 장점을 가지고 있는 GFIS을 개발하기 위해서는 GFIS에서 발생된 이온빔의 형상을 관찰 것이 매우 중요하며, 이러한 관찰을 위한 시스템에는 주로 마이크로 채널 플레이트 (MCP: Micro Channel Plate)가 사용된다. MCP는 채널내부에 입사한 입자의 에너지에 의해서 생성된 이차전자를 수 천 배에서 수 백 만 배 이상 증폭시켜 형광판에 조사하고 발광시키는 방법으로 작은 신호를 영상으로 관찰 할 수 있도록 한다. MCP의 큰 증폭비는 작은 크기의 신호를 큰 신호로 증폭하여 관찰하는데 용이하여, GFIS 방법으로 생성된 이온빔(이온빔 전류 값은 pA 수준)을 관찰하기에 적합하다. 그러나 MCP를 이용하여도 증폭된 이온빔의 세기가 매우 작기때문에 생성된 이온빔 형상을 정확하게 관찰하기 위해서는 MCP의 형광판을 촬영하는 카메라 노출시간을 길게하여 데이터 수집 시간을 늘려야 하는 문제가 있다. 본 발표에서는 이온빔 형상 관찰에 소요되는 시간을 단축하기 위하여 MCP의 잡음이 GFIS의 이온빔 이미지 관찰에 미치는 영향을 분석하고 이를 제거 방법을 소개한다. 본 연구에서는 GFIS 방출 이온빔의 이미지에 포함된 MCP 잡음 특성을 장(전계)이온현미경 (Field Ion Microscope)실험을 통하여 분석하였고, 디지털 이미지 처리 방법을 이용하여 방출 이온빔 이미지에서 MCP 잡음을 제거하여 방출 이온빔 이미지만 추출할 수 있었다. 본 연구에서 제안한 방법을 GFIS 방출 이온빔 관찰시스템에 적용함으로써 기존 방법에 비해 노출시간을 단축하여 방출 이온빔을 관찰 할 수 있었으며, 노이즈 제거 효과로 향상된 이온빔 형상을 얻을 수 있었다. 본 연구결과의 관찰시간 단축과 향상된 이온빔 형상 획득은 이온현미경 개발에 필수적인 단원자 이온빔을 보다 효율적으로 개발할 수 있으며 디지털 이미지 처리로 GFIS 이온빔 생성을 자동화하는데 응용할 수 있다. 더불어 기존방법에 비해 이미지 획득을 위한 MCP의 노출시간을 단축할 수 있으므로 실험장비 수명 단축 방지 및 관리에 큰 장점이 있다.

• The KIPS Transactions:PartB
• /
• v.14B no.4
• /
• pp.249-254
• /
• 2007

A mobile robot with various types of sensors and wireless camera is introduced. We show this mobile robot can detect objects well by combining the results of active sensors and image processing algorithm. First, to detect objects, active sensors such as infrared rays sensors and supersonic waves sensors are employed together and calculates the distance in real time between the object and the robot using sensor's output. The difference between the measured value and calculated value is less than 5%. We focus on how to detect a object region well using image processing algorithm because it gives robots the ability of working for human. This paper suggests effective visual detecting system for moving objects with specified color and motion information. The proposed method includes the object extraction and definition process which uses color transformation and AWUPC computation to decide the existence of moving object. Shape information and signature algorithm are used to segment the objects from background regardless of shape changes. We add weighing values to each results from sensors and the camera. Final results are combined to only one value which represents the probability of an object in the limited distance. Sensor fusion technique improves the detection rate at least 7% higher than the technique using individual sensor.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_1
• /
• pp.959-971
• /
• 2019

CNN (Convolutional Neural Network) is one representative deep learning algorithm, which can extract high-level spatial and spectral features, and has been applied for hyperspectral image classification. However, one significant drawback behind the application of CNNs in hyperspectral images is the high dimensionality of the data, which increases the training time and processing complexity. To address this problem, several CNN based hyperspectral image classification studies have exploited PCA (Principal Component Analysis) for dimensionality reduction. One limitation to this is that the spectral information of the original image can be lost through PCA. Although it is clear that the use of PCA affects the accuracy and the CNN training time, the impact of PCA for CNN based hyperspectral image classification has been understudied. The purpose of this study is to analyze the quantitative effect of PCA in CNN for hyperspectral image classification. The hyperspectral images were first transformed through PCA and applied into the CNN model by varying the size of the reduced dimensionality. In addition, 2D-CNN and 3D-CNN frameworks were applied to analyze the sensitivity of the PCA with respect to the convolution kernel in the model. Experimental results were evaluated based on classification accuracy, learning time, variance ratio, and training process. The size of the reduced dimensionality was the most efficient when the explained variance ratio recorded 99.7%~99.8%. Since the 3D kernel had higher classification accuracy in the original-CNN than the PCA-CNN in comparison to the 2D-CNN, the results revealed that the dimensionality reduction was relatively less effective in 3D kernel.