• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.317-326
• /
• 2024

Biometric recognition is a technology that determines whether a person is identified by extracting information on a person's biometric and behavioral characteristics with a specific device. Cyber threats such as forgery, duplication, and hacking of biometric characteristics are increasing in the field of biometrics. In response, the security system is strengthened and complex, and it is becoming difficult for individuals to use. To this end, multiple biometric models are being studied. Existing studies have suggested feature fusion methods, but comparisons between feature fusion methods are insufficient. Therefore, in this paper, we compared and evaluated the fusion method of multiple biometric models using fingerprint, face, and iris images. VGG-16, ResNet-50, EfficientNet-B1, EfficientNet-B4, EfficientNet-B7, and Inception-v3 were used for feature extraction, and the fusion methods of 'Sensor-Level', 'Feature-Level', 'Score-Level', and 'Rank-Level' were compared and evaluated for feature fusion. As a result of the comparative evaluation, the EfficientNet-B7 model showed 98.51% accuracy and high stability in the 'Feature-Level' fusion method. However, because the EfficietnNet-B7 model is large in size, model lightweight studies are needed for biocharacteristic fusion.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.1
• /
• pp.115-127
• /
• 2024

The recent increase in digital audio media has greatly expanded the size and diversity of sound data, which has increased the importance of sound data analysis in the digital forensics process. However, the lack of standardized procedures and guidelines for sound data analysis has caused problems with the consistency and reliability of analysis results. The digital environment includes a wide variety of audio formats and recording conditions, but current audio forensic methodologies do not adequately reflect this diversity. Therefore, this study identifies Life-Cycle-based sound data elemental technologies and provides overall guidelines for sound data analysis so that effective analysis can be performed in all situations. Furthermore, the identified elemental technologies were analyzed for use in the development of digital forensic techniques for sound data. To demonstrate the effectiveness of the life-cycle-based sound data elemental technology identification system presented in this study, a case study on the process of developing an emergency retrieval technology based on sound data is presented. Through this case study, we confirmed that the elemental technologies identified based on the Life-Cycle in the process of developing digital forensic technology for sound data ensure the quality and consistency of data analysis and enable efficient sound data analysis.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.1
• /
• pp.129-141
• /
• 2024

Amidst the acceleration of digital transformation across various sectors, the financial market is increasingly focusing on the development of digital and electronic payment methods, including currency. Among these, Central Bank Digital Currencies (CBDC) are emerging as future digital currencies that could replace physical cash. They are stable, not subject to value fluctuation, and can be exchanged one-to-one with existing physical currencies. Recently, both domestic and international efforts are underway in researching and developing CBDCs. However, current CBDC systems face scalability issues such as delays in processing large transactions, response times, and network congestion. To build a universal CBDC system, it is crucial to resolve these scalability issues, including the low throughput and network overload problems inherent in existing blockchain technologies. Therefore, this study proposes a solution based on reinforcement learning for handling large-scale data in a CBDC environment, aiming to improve throughput and reduce network congestion. The proposed technology can increase throughput by more than 64 times and reduce network congestion by over 20% compared to existing systems.

• International Journal of Computer Science & Network Security
• /
• v.24 no.9
• /
• pp.30-40
• /
• 2024

Building an accurate 3-D spatial road network model has become an active area of research now-a-days that profess to be a new paradigm in developing Smart roads and intelligent transportation system (ITS) which will help the public and private road impresario for better road mobility and eco-routing so that better road traffic, less carbon emission and road safety may be ensured. Dealing with such a large scale 3-D road network data poses challenges in getting accurate elevation information of a road network to better estimate the CO₂ emission and accurate routing for the vehicles in Internet of Vehicle (IoV) scenario. Clustering and regression techniques are found suitable in discovering the missing elevation information in 3-D spatial road network dataset for some points in the road network which is envisaged of helping the public a better eco-routing experience. Further, recently Explainable Artificial Intelligence (xAI) draws attention of the researchers to better interprete, transparent and comprehensible, thus enabling to design efficient choice based models choices depending upon users requirements. The 3-D road network dataset, comprising of spatial attributes (longitude, latitude, altitude) of North Jutland, Denmark, collected from publicly available UCI repositories is preprocessed through feature engineering and scaling to ensure optimal accuracy for clustering and regression tasks. K-Means clustering and regression using Support Vector Machine (SVM) with radial basis function (RBF) kernel are employed for 3-D road network analysis. Silhouette scores and number of clusters are chosen for measuring cluster quality whereas error metric such as MAE ( Mean Absolute Error) and RMSE (Root Mean Square Error) are considered for evaluating the regression method. To have better interpretability of the Clustering and regression models, SHAP (Shapley Additive Explanations), a powerful xAI technique is employed in this research. From extensive experiments , it is observed that SHAP analysis validated the importance of latitude and altitude in predicting longitude, particularly in the four-cluster setup, providing critical insights into model behavior and feature contributions SHAP analysis validated the importance of latitude and altitude in predicting longitude, particularly in the four-cluster setup, providing critical insights into model behavior and feature contributions with an accuracy of 97.22% and strong performance metrics across all classes having MAE of 0.0346, and MSE of 0.0018. On the other hand, the ten-cluster setup, while faster in SHAP analysis, presented challenges in interpretability due to increased clustering complexity. Hence, K-Means clustering with K=4 and SVM hybrid models demonstrated superior performance and interpretability, highlighting the importance of careful cluster selection to balance model complexity and predictive accuracy.

• Journal of the Korea Society for Simulation
• /
• v.32 no.3
• /
• pp.43-53
• /
• 2023

In this detailed and comprehensive study, our primary focus has been placed on accurately gauging the number of visitors and their real-time locations in commercial spaces. Particularly, in a real cafe, using security cameras, we have developed a system that can offer live updates on available seating and predict future congestion levels. By employing YOLO, a real-time object detection and tracking algorithm, the number of visitors and their respective locations in real-time are also monitored. This information is then used to update a cafe's indoor map, thereby enabling users to easily identify available seating. Moreover, we developed a model that predicts the congestion of a cafe in real time. The sophisticated model, designed to learn visitor count and movement patterns over diverse time intervals, is based on Long Short Term Memory (LSTM) to address the vanishing gradient problem and Sequence-to-Sequence (Seq2Seq) for processing data with temporal relationships. This innovative system has the potential to significantly improve cafe management efficiency and customer satisfaction by delivering reliable predictions of cafe congestion to all users. Our groundbreaking research not only demonstrates the effectiveness and utility of indoor location tracking technology implemented through security cameras but also proposes potential applications in other commercial spaces.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.12
• /
• pp.1959-1970
• /
• 2013

Attempts were made in this paper to compare the practice of protective actions for information of patients' medical treatment between hospital administrators who do not make direct contact with patients and occupational therapists who usually do. The comparison between jobs in charge showed that occupational therapists did not practice much protective actions for information of patients' medical treatment ($3.52{\pm}.809$) compared to hospital administrators ($3.92{\pm}.724$), even though the former had received regular education about protection, management and supervision of patients' medical information more often ($3.17{\pm}1.129$) than the latter ($3.16{\pm}1.037$). In spite of the fact that occupational therapists were exposed frequently to the danger of revealing medical information in the process of their job performance through talks and communications with patients, they displayed relatively little concern for and awareness of keeping information of medical treatment from being leaked by them. It is thus suggested to promote awareness of medical staff to protect medical information by means of flexible educational system for each occupational group, periodical monitoring, continuing public relation, training and quality control for protection of medical information, as well as routine self-examination of such practice.

• Korean Journal of Remote Sensing
• /
• v.39 no.1
• /
• pp.77-86
• /
• 2023

Ship detection is widely used in areas such as maritime security, maritime traffic, fisheries management, illegal fishing, and border control, and ship detection is important for rapid response and damage minimization as ship accident rates increase due to recent increases in international maritime traffic. Currently, according to a number of global and national regulations, ships must be equipped with automatic identification system (AIS), which provide information such as the location and speed of the ship periodically at regular intervals. However, most small vessels (less than 300 tons) are not obligated to install the transponder and may not be transmitted intentionally or accidentally. There is even a case of misuse of the ship'slocation information. Therefore, in this study, ship detection was performed using high-resolution optical satellite images that can periodically remotely detect a wide range and detectsmallships. However, optical images can cause false-alarm due to noise on the surface of the sea, such as waves, or factors indicating ship-like brightness, such as clouds and wakes. So, it is important to remove these factors to improve the accuracy of ship detection. In this study, false alarm wasreduced, and the accuracy ofship detection wasimproved by removing wake.As a ship detection method, ship detection was performed using machine learning-based random forest (RF), and convolutional neural network (CNN) techniquesthat have been widely used in object detection fieldsrecently, and ship detection results by the model were compared and analyzed. In addition, in this study, the results of RF and CNN were combined to improve the phenomenon of ship disconnection and the phenomenon of small detection. The ship detection results of thisstudy are significant in that they improved the limitations of each model while maintaining accuracy. In addition, if satellite images with improved spatial resolution are utilized in the future, it is expected that ship and wake simultaneous detection with higher accuracy will be performed.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39B no.4
• /
• pp.207-215
• /
• 2014

Password based authentication systems are most widely used for user convenience in web services. However such authentication systems are known to be vulnerable to various attacks such as password guessing attack, dictionary attack and key logging attack. Besides, many of the web systems just provide user authentication in a one-way fashion such that web clients cannot verify the authenticity of the web server to which they set access and give passwords. Therefore, it is too difficult to protect against DNS spoofing, phishing and pharming attacks. To cope with the security threats, web system adopts several enhanced schemes utilizing one time password (OTP) or long and strong passwords including special characters. However there are still practical issues. Users are required to buy OTP devices and strong passwords are less convenient to use. Above all, one-way authentication schemes generate several vulnerabilities. To solve the problems, we propose a multi-channel, multi-factor authentication scheme by utilizing QR-Code. The proposed scheme supports both user and server authentications mutually, thereby protecting against attacks such as phishing and pharming attacks. Also, the proposed scheme makes use of a portable smart device as a OTP generator so that the system is convenient and secure against traditional password attacks.

• International Commerce and Information Review
• /
• v.16 no.1
• /
• pp.45-68
• /
• 2014

Since the introduction of AEO certification in South Korea in 2008, 516 areas have been AEO certified based on a report dated December 31st, 2013. The number is expected to increase drastically as more and more companies strive to obtain the AEO certification. In this regards, this study is designed to contribute to the further invigoration of the AEO certification system as well as to propose directions to improve companies' performance in a practical point of view. To do so, the writer reviewed the past studies done in the relevant subjects in order to 1) grasp the principal subject, 2) catch the areas with insufficient information/lacking points, and to 3) get a focus on future directions on how additional research should be conducted. An analysis of the past studies proved that prior to the introduction of the AEO certification, the focus of the studies were to present the reasons as to why AEO should be introduced, the benefits of AEO, and a comparative analysis of countries already in wide usage of the AEO system. Studies done after the introduction of AEO (that is, after 2008) focused mainly on invigorating the system into the market. Compared to the previous studies, this study will present a distinct conclusion by breaking down the major points of each study on a year-by-year basis and studying them in total in order to provide an even more practical direction in the future of AEO in South Korea.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.7
• /
• pp.141-149
• /
• 2014

This paper describes a method to insert and detect watermark or fingerprint to/from videos in low-computing powered system as Android platforms. Fingerprint, which is a kind of watermark, has features such as patterns that contain information. The inserting frame numbers in video-stream and the positions in a picture were chosen from the encrypted user ID to insert the watermarks. The used encrypt algorithm is the HIGHT algorithm which was developed for low-computing powered systems by KISA(Korean Internet & Security Agency). Subtracting an inferred picture from the previous picture was used to extract a candidate feature. Median filtering was used to get rid of noise and stabilize the candidate feature. New algorithm that reduces calculating steps of the median filtering was developed and applied for low-specification systems. The stabilized features were accumulated over 150 times and calculated by correlation coefficient method to recognize the patterns. We examined 22 videos and successfully detected the patterns from 21 videos. The correlation coefficient r values that we examined through this study exceeded over 0.79 more than the threshold (0.7).