• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.28 no.4
• /
• pp.847-857
• /
• 2018

As the number of malicious code increases steeply, cyber attack victims targeting corporations, public institutions, financial institutions, hospitals are also increasing. Accordingly, academia and security industry are conducting various researches on malicious code detection. In recent years, there have been a lot of researches using machine learning techniques including deep learning. In the case of research using Convolutional Neural Network, ResNet, etc. for classification of malicious code, it can be confirmed that the performance improvement is higher than the existing classification method. However, one of the characteristics of the target attack is that it is custom malicious code that makes it operate only for a specific company, so it is not a form spreading widely to a large number of users. Since there are not many malicious codes of this kind, it is difficult to apply the previously studied machine learning or deep learning techniques. In this paper, we propose a method to classify malicious codes when the amount of samples is insufficient such as targeting type malicious code. As a result of the study, we confirmed that the accuracy of 97% can be achieved even with a small amount of data by applying the Memory Augmented Neural Networks model.

• Journal of Intelligence and Information Systems
• /
• v.20 no.1
• /
• pp.121-131
• /
• 2014

Traffic speed is an important measure in transportation. It can be employed for various purposes, including traffic congestion detection, travel time estimation, and road design. Consequently, accurate speed prediction is essential in the development of intelligent transportation systems. In this paper, we present an analysis and speed prediction of a certain road section in Busan, South Korea. In previous works, only historical data of the target link are used for prediction. Here, we extract features from real traffic data by considering the neighboring links. After obtaining the candidate features, linear regression, model tree, and k-nearest neighbor (k-NN) are employed for both feature selection and speed prediction. The experiment results show that k-NN outperforms model tree and linear regression for the given dataset. Compared to the other predictors, k-NN significantly reduces the error measures that we use, including mean absolute percentage error (MAPE) and root mean square error (RMSE).

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.1
• /
• pp.73-79
• /
• 2020

Modern radar system requires high spectral purity and low phase noise characteristics for very low RCS target detection and high resolution SAR (Synthetic Aperture Radar) image. This paper presents a new X-band high stable frequency synthesizer for high performance radar system, which combines DAS (Direct Analog Synthesizer) and DDS (Direct Digital Synthesizer) techniques, in order to cope with very low phase noise and high frequency agility requirements. This synthesizer offers more than 10% operating bandwidth in X-band frequency and fast agile time lower than 1 usec. Also, the phase noise at 10kHz offset is lower than -136dBc/Hz, which shows an improvement of more than 10dB compared to the current state of art frequency synthesizer. This architecture can be applied to L-band and C-band application as well. This frequency synthesizer is able to used in modern AESA (Active Electronically Scanned Array) radar system and high resolution SAR application.

• Journal of Veterinary Clinics
• /
• v.33 no.2
• /
• pp.107-112
• /
• 2016

The objective of this study was to determine potential contributing factors associated with biosecurity level of farrow-to-finish pig farms and to develop a classification tree model to explore how these factors related to each other based on prediction model. To this end, the author analyzed data (n = 193) extracted from a cross-sectional study of 344 farrow-to-finish farms which was conducted between March and September 2014 aimed to explore swine disease status at farm level. Standardized questionnaires with information about basic demographical data and management practices were collected in each farm by on-site visit of trained veterinarians. For the classification of the data sets regarding biosecurity level as a dependent variable and predictor variables, Chi-squared Automatic Interaction Detection (CHAID) algorithm was applied for modeling classification tree. The statistics of misclassification risk was used to evaluate the fitness of the model in terms of prediction results. Categorical multivariate input data (40 variables) was used to construct a classification tree, and the target variable was biosecurity level dichotomized into low versus high. In general, the level of biosecurity was lower in the majority of farms studied, mainly due to the limited implementation of on-farm basic biosecurity measures aimed at controlling the potential introduction and transmission of swine diseases. The CHAID model illustrated the relative importance of significant predictors in explaining the level of biosecurity; maintenance of medical records of treatment and vaccination, use of dedicated clothing to enter the farm, installing fence surrounding the farm perimeter, and periodic monitoring of the herd using written biosecurity plan in place. The misclassification risk estimate of the prediction model was 0.145 with the standard error of 0.025, indicating that 85.5% of the cases could be classified correctly by using the decision rule based on the current tree. Although CHAID approach could provide detailed information and insight about interactions among factors associated with biosecurity level, further evaluation of potential bias intervened in the course of data collection should be included in future studies. In addition, there is still need to validate findings through the external dataset with larger sample size to improve the external validity of the current model.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.36 no.3
• /
• pp.195-201
• /
• 2016

Eddy current testing (ECT) is commonly applied for the inspection of automated production lines of metallic products, because it has a high inspection speed and a reasonable price. When ECT is applied for the inspection of a metallic object having an uneven target surface, such as the spline gear of a spline shaft, it is difficult to distinguish between the original signal obtained from the sensor and the signal generated by a defect because of the relatively large surface signals having similar frequency distributions. To facilitate the detection of defect signals from the spline gear, implementation of high-order filters is essential, so that the fault signals can be distinguished from the surrounding noise signals, and simultaneously, the pass-band of the filter can be adjusted according to the status of each production line and the object to be inspected. We will examine the infinite impulse filters (IIR filters) available for implementing an advanced filter for ECT, and attempt to detect the flaw signals through optimization of system design parameters for detecting the signals at the system level.

• Convergence Security Journal
• /
• v.18 no.5_1
• /
• pp.113-120
• /
• 2018

Today, the development of information and communication technology is rapidly increasing the number of users of network-based service, and enables real-time information sharing among users on the Internet. There are various methods in the information sharing process, and information sharing based on portal service is generally used. However, the process of information sharing serves as a cause of illegal activities in order to amplify the social interest of the relevant stakeholders. Public opinion attack using macro function can distort normal public opinion, so security measures are urgent. Therefore, security measures are urgently needed. Macro attacks are generally defined as attacks in which illegal users acquire multiple IP or ID to manipulate public opinion on the content of a particular web page. In this paper, we analyze network path information based on traceback for macro attack of a specific user, and then detect multiple access of the user. This is a macro attack when the access path information for a specific web page and the user information are matched more than once. In addition, when multiple ID is accessed for a specific web page in the same region, it is not possible to distort the overall public opinion on a specific web page by analyzing the threshold count value.

• Journal of the Korea Convergence Society
• /
• v.10 no.1
• /
• pp.103-113
• /
• 2019

License plate recognition camera is dedicated device designed for acquiring images of the target vehicle for recognizing letters and numbers in a license plate. Mostly, it is used as a part of the system combined with server and image analysis module rather than as a single use. However, building a system for vehicle license plate recognition is costly because it is required to construct a facility with a server providing the management and analysis of the captured images and an image analysis module providing the extraction of numbers and characters and recognition of the vehicle's plate. In this study, we would like to develop an embedded type convergent camera (Edge Base) which can expand the function of the camera to not only the license plate recognition but also the security CCTV function together and to perform two functions within the camera. This embedded type convergence camera equipped with a high resolution 4K IP camera for clear image acquisition and fast data transmission extracted license plate area by applying YOLO, a deep learning software for multi object recognition based on open source neural network algorithm and detected number and characters of the plate and verified the detection accuracy and recognition accuracy and confirmed that this camera can perform CCTV security function and vehicle number plate recognition function successfully.

• Applied Chemistry for Engineering
• /
• v.30 no.4
• /
• pp.429-434
• /
• 2019

A lateral flow immunoassay platform utilizing antibody functionalized water soluble CdSe/ZnS semiconductor quantum dots (QDs) was developed for the analysis of human serum amyloid A-1 (hSAA1) in a buffer solution. hSAA1 was chosen as a target protein because it is regarded as a potential biomarker associated with early diagnosis and prognosis in patients of lung cancer. The immunoassay strip on a nitrocellulose membrane was fabricated by spraying two lines composed of a test line with a monoclonal antibody against hSAA1 (10G1) (anti hSAA1) and a control line of anti-chicken IgY. While the CdSe/ZnS QDs synthesized in an organic phase were transferred to a water phase by ligand exchange using carboxylic acid modified alkane thiol. The QDs was then conjugated to monoclonal antibody against hSAA1 (14F8) [anti hSAA1 (14F8)] and used as a fluorescent detection probe. The sequential lateral flow of hSAA1 in different concentration and QDs-anti hSAA1 (14F8) complex allowed to form the surface sandwich complex of anti hSAA1 (10G1)/hSAA1/QD-anti hSAA1 (14F8), which was then analyzed using fluorescence microscope. A 100 nM concentration of hSAA1 protein can be detected by naked eyes under an optimized lateral flow buffer condition with a sensing time of 5 mins.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.4
• /
• pp.35-42
• /
• 2019

Nut fasteners that require ultra-precise control are required in the overall manufacturing industry including electronic products that are currently developing with the automobile industry. Important performance factors when tightening nuts include loosening due to insufficient fastening force, breakage due to excessive fastening, Tightening torque and angle are required to maintain and improve the assembling quality and ensure the life of the product. Nut fasteners, which are now marketed under the name Nut Runner, require high torque and precision torque control, precision angle control, and high speed operation for increased production, and are required for sophisticated torque control dedicated to high output BLDC motors and nut fasteners. It is demanded to develop a high-precision torque control driver and a high-speed, low-speed, high-response precision speed control system, but it does not satisfy the high precision, high torque and high speed operation characteristics required by customers. Therefore, in this paper, we propose a control technique of BLDC motor variable speed control and nut runner based on vector control and torque control based on coordinate transformation of d axis and q axis that can realize low vibration and low noise even at accurate tightening torque and high speed rotation. The performance results were analyzed to confirm that the proposed control satisfies the nut runner performance. In addition, it is confirmed that the pattern is programmed by One-Stage operation clamping method and it is tightened to the target torque exactly after 10,000 [rpm] high speed operation. The problem of tightening torque detection by torque ripple is also solved by using disturbance observer Respectively.

• Analytical Science and Technology
• /
• v.32 no.2
• /
• pp.35-47
• /
• 2019

In this study, high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) was employed to detect 26 antidiabetic compounds in adulterated dietary supplements using a simple, selective method. The work presented herein may help prevent incidents related to food adulteration and restrict the illegal food market. The best separation was obtained on a Shiseido Capcell Pak(R) C18 MG-II ($2.0mm{\times}100mm$, $3{\mu}m$), which improved the peak shape and MS detection sensitivity of the target compounds. A gradient elution system composed of 0.1 % (v/v) formic acid in distilled water and methanol at a flow rate of 0.3 mL/min for 18 min was utilized. A triple quadrupole mass spectrometer with an electrospray ionization source operated in the positive or negative mode was employed as the detector. The developed method was validated as follows: specificity was confirmed in the multiple reaction monitoring mode using the precursor and product ion pairs. For solid samples, LOD ranged from 0.16 to 20.00 ng/mL and LOQ ranged from 0.50 to 60.00 ng/mL, and for liquid samples, LOD ranged from 0.16 to 20.00 ng/mL and LOQ ranged from 0.50 to 60.00 ng/mL. Satisfactory linearity was obtained from calibration curves, with $R^2$ > 0.99. Both intra and inter-day precision were less than 13.19 %. Accuracies ranged from 80.69 to 118.81 % (intra/inter-day), with a stability of less than 14.88 %. Mean recovery was found to be 80.6-119.0 % and less than 13.4 % RSD. Using the validated method, glibenclamide and pioglitazone were simultaneously determined in one capsule at concentrations of 1.52 and 0.53 mg (per capsule), respectively.