• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.2
• /
• pp.305-314
• /
• 2017

Recently, user authentication through biometric traits such as fingerprint and iris raise more and more attention especially in mobile commerce and fin-tech fields. In particular, commercialized authentication methods using fingerprint recognition are widely utilized mainly because customers are more adopted and used to fingerprint recognition applications. In the meantime, the security issues caused by fingerprint falsification bring lots of attention. In this paper, we propose a new method to improve the performance of fake fingerprint detection using CNN(Convolutional Neural Network). It is common practice to increase the amount of learning data by using affine transformation or horizontal reflection to improve the detection rate in CNN characteristics that are influenced by learning data. However, in this paper we propose an effective data augmentation method based on the database difficulty level. The experimental results confirm the validity of proposed method.

• Journal of KIISE:Software and Applications
• /
• v.26 no.10
• /
• pp.1143-1155
• /
• 1999

셀룰러 방식의 이동통신 시스템에서 전파의 유효신호 도달범위를 예측하기 위해서는 전파전파 모델을 이용한 예측기법이 주로 사용된다. 그러나, 전파과정에서 주변 지형지물에 의해 발생하는 전파손실은 매우 복잡한 비선형적인 특성을 가지며 수식으로는 정확한 표현이 불가능하다. 본 논문에서는 신경회로망의 함수 근사화 능력을 이용하여 전파손실 예측모델을 생성하는 방법을 제안한다. 즉, 전파손실을 송수신 안테나간의 거리, 송신안테나의 특성, 장애물 투과영향, 회절특성, 도로, 수면에 의한 영향 등과 같은 전파환경 변수들의 함수로 가정하고, 신경회로망 학습을 통하여 함수를 근사화한다. 전파환경 변수들이 신경회로망 입력으로 사용되기 위해서는 3차원 지형도와 벡터지도를 이용하여 전파의 반사, 회절, 산란 등의 물리적인 특성이 고려된 특징 추출을 통해 정량적인 수치들을 계산한다. 이와 같이 얻어진 훈련데이타를 이용한 신경회로망 학습을 통해 전파손실 모델을 완성한다. 이 모델을 이용하여 서울 도심 지역의 실제 서비스 환경에 대한 타 모델과의 비교실험결과를 통해 제안하는 모델의 우수성을 보인다.Abstract In cellular mobile communication systems, wave propagation models are used in most cases to predict cell coverage. The amount of propagation loss induced by the obstacles in the propagation path, however, is a highly non-linear function, which cannot be easily represented mathematically. In this paper, we introduce the method of producing propagation loss prediction models by function approximation using neural networks. In this method, we assume the propagation loss is a function of the relevant parameters such as the distance from the base station antenna, the specification of the transmitter antenna, obstacle profile, diffraction effect, road, and water effect. The values of these parameters are produced from the field measurement data, 3D digital terrain maps, and vector maps as its inputs by a feature extraction process, which takes into account the physical characteristics of electromagnetic waves such as reflection, diffraction and scattering. The values produced are used as the input to the neural network, which are then trained to become the propagation loss prediction model. In the experimental study, we obtain a considerable amount of improvement over COST-231 model in the prediction accuracy using this model.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.5
• /
• pp.592-601
• /
• 2016

The recurrent neural network (RNN) is a deep learning model which is suitable to sequential or length-variable data. The Long Short-Term Memory (LSTM) mitigates the vanishing gradient problem of RNNs so that LSTM can maintain the long-term dependency among the constituents of the given input sequence. In this paper, we propose a LSTM based language model which can predict following words of a given incomplete sentence to generate a complete sentence. To evaluate our method, we trained our model using multiple Korean corpora then generated the incomplete part of Korean sentences. The result shows that our language model was able to generate the fluent Korean sentences. We also show that the word based model generated better sentences compared to the other settings.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.1
• /
• pp.9-16
• /
• 2020

The k-anonymity scheme has been widely used to protect private information when Big Data are distributed to a third party for research purposes. When the scheme is applied, an optimal k value determination is one of difficult problems to be resolved because many factors should be considered. Currently, the determination has been done almost manually by human experts with their intuition. This leads to degrade performance of the anonymization, and it takes much time and cost for them to do a task. To overcome this problem, a simple idea has been proposed that is based on machine learning. This paper describes implementations and experiments to realize the proposed idea. In thi work, a deep neural network (DNN) is implemented using tensorflow libraries, and it is trained and tested using input dataset. The experiment results show that a trend of training errors follows a typical pattern in DNN, but for validation errors, our model represents a different pattern from one shown in typical training process. The advantage of the proposed approach is that it can reduce time and cost for experts to determine k value because it can be done semi-automatically.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2006.05a
• /
• pp.437-438
• /
• 2006

Generally, the vertical roll process is used to achieve extensive width reduction in hot strip mill. However, it is difficult to avoid the defects such as dog-bone and seam-defect. The sizing press has been developed in response to the defects mentioned above. Especially, this study is carried out to investigate the deformation of slab by two-step sizing press. The deformation behavior in the width sizing process is more favorable than that in conventional vertical rolling edger. The objective of this study is to determine the optimal anvil shape parameters in the sizing press with two-step die from the viewpoint of edge-seam length. In general, the edge-seam defect occurs parallel to the rolling direction at both edges in horizontal rolling process after sizing press. The optimal combination of the parameters is determined by FE-simulation and Artificial Neural Network (ANN). The slab deformation in sizing press with convex anvil is analyzed by FE-simulation. The most suitable profile of the anvil is also discussed fur the improvement of trimming loss because of the side seam defect by FE-simulation and ANN.

• Journal of Preventive Medicine and Public Health
• /
• v.36 no.2
• /
• pp.147-152
• /
• 2003

Objectives : To develop a Diagnosis-Related Group (DRG) fraud candidate detection method, using data mining techniques, and to examine the efficiency of the developed method. Methods ; The Study included 79,790 DRGs and their related claims of 8 disease groups (Lens procedures, with or without, vitrectomy, tonsillectomy and/or adenoidectomy only, appendectomy, Cesarean section, vaginal delivery, anal and/or perianal procedures, inguinal and/or femoral hernia procedures, uterine and/or adnexa procedures for nonmalignancy), which were examined manually during a 32 months period. To construct an optimal prediction model, 38 variables were applied, and the correction rate and lift value of 3 models (decision tree, logistic regression, neural network) compared. The analyses were peformed separately by disease group. Results : The correction rates of the developed method, using data mining techniques, were 15.4 to 81.9%, according to disease groups, with an overall correction rate of 60.7%. The lift values were 1.9 to 7.3 according to disease groups, with an overall lift value of 4.1. Conclusions : The above findings suggested that the applying of data mining techniques is necessary to improve the efficiency of DRG fraud candidate detection.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.39 no.9
• /
• pp.839-847
• /
• 2011

This paper proposes a new guidance law for increasing the lethality of munitions. The well known PNG (Proportional Navigation Guidance) is inadequate for the munitions because of some weaknesses. Even if the munition does not have the impact point error, the acceleration command is non zero because the line-of-sight changes at all times in flight. Therefore, we use a difference between a target and an impact point. This proposed guidance law is similar to PNG in the form, but this guidance law concentrates a correction rate of flight path angle instead of the LOS (Line of Sight) rate. The correction of flight path angle is defined as the amount of impact point error. This impact point error can be calculated by neural networks rapidly. Finally, we show that the simulation results prove the suitability of this law.

• Journal of Korean Society for Atmospheric Environment
• /
• v.22 no.5
• /
• pp.614-626
• /
• 2006

Using the data of three environmental monitoring sites in Pohang area(KME112, KME113, and KME114), statistical forecasting models of the daily maximum and mean values of PM10 have been developed. Since the distributions of the daily maximum and mean PM10 values are skewed, which are similar to the Weibull distribution, these values were log-transformed to increase prediction accuracy by approximating the normal distribution. Three statistical forecasting models, which are regression, neural networks(NN) and support vector regression(SVR), were built using the log-transformed response variables, i.e., log(max(PM10)) or log(mean (PM10)). Also, the forecasting models were validated by the measure of RMSE, CORR, and IOA for the model comparison and accuracy. The improvement rate of IOA before and after the log-transformation in the daily maximum PM10 prediction was 12.7% for the regression and 22.5% for NN. In particular, 42.7% was improved for SVR method. In the case of the daily mean PM10 prediction, IOA value was improved by 5.1% for regression, 6.5% for NN, and 6.3% for SVR method. As a conclusion, SVR method was found to be performed better than the other methods in the point of the model accuracy and fitness views.

• Journal of Biosystems Engineering
• /
• v.19 no.2
• /
• pp.124-137
• /
• 1994

Visual features of a mushroom(Lentinus Edodes, L.) are critical in grading and sorting as most agricultural products are. Because of its complex and various visual features, grading and sorting of mushrooms have been done manually by the human expert. To realize the automatic handling and grading of mushrooms in real time, the computer vision system should be utilized and the efficient and robust processing of the camera captured visual information be provided. Since visual features of a mushroom are distributed over the front and back sides, recognizing sides and states of the stalk including the stalk orientation from the captured image is a prime process in the automatic task processing. In this paper, the efficient and robust recognition process identifying the front and back side and the state of the stalk was developed and its performance was compared with other recognition trials. First, recognition was tried based on the rule set up with some experimental heuristics using the quantitative features such as geometry and texture extracted from the segmented mushroom image. And the neural net based learning recognition was done without extracting quantitative features. For network inputs the segmented binary image obtained from the combined type automatic thresholding was tested first. And then the gray valued raw camera image was directly utilized. The state of the stalk seriously affects the measured size of the mushroom cap. When its effect is serious, the stalk should be excluded in mushroom cap sizing. In this paper, the stalk removal process followed by the boundary regeneration of the cap image was also presented. The neural net based gray valued raw image processing showed the successful results for our recognition task. The developed technology through this research may open the new way of the quality inspection and sorting especially for the agricultural products whose visual features are fuzzy and not uniquely defined.

• Journal of Internet Computing and Services
• /
• v.7 no.1
• /
• pp.75-89
• /
• 2006

This paper is concerned with analysing the bankruptcy prediction power of three models: Multivariate Discriminant Analysis(MDA), Logit Analysis, Neural Network. The research targeted the bankrupted companies after the foreign exchange crisis in 1997 to differentiate from previous research efforts, and all participating companies were randomly selected from the KSE listed companies belonging to manufacturing industry to improve prediction accuracy and validity of the model. In order to assure meaningful bankruptcy prediction, training data and testing data were not extracted within the corresponding period. The result is that prediction accuracy of neural networks is more excellent than that of logit analysis and MDA model when considering that execution of testing data was followed by execution of training data.