• Journal of Internet Computing and Services
• /
• v.8 no.6
• /
• pp.43-53
• /
• 2007

Intrusion detection system(IDS) has recently evolved while combining signature-based detection approach with anomaly detection approach. Although signature-based IDS tools have been commonly used by utilizing machine learning algorithms, they only detect network intrusions with already known patterns, Ideal IDS tools should always keep the signature database of your detection system up-to-date. The system needs to generate the signatures to detect new possible attacks while monitoring and analyzing incoming network data. In this paper, we propose a new outlier cluster detection algorithm with density (or influence) function, Our method assumes that an outlier is a kind of cluster with similar instances instead of a single object in the context of network intrusion, Through extensive experiments using KDD 1999 Cup Intrusion Detection dataset. we show that the proposed method outperform the conventional outlier detection method using Euclidean distance function, specially when attacks occurs frequently.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.2
• /
• pp.186-190
• /
• 2015

Intelligent vehicle plans motion and navigate itself based on the surrounding environment perception. Hence, the precise environment recognition is an essential part of self-driving vehicle. There exist many vulnerable road users (e.g. vehicle, pedestrians) on vehicular driving environment, the vehicle must percept all the dynamic obstacles accurately for safety. In this paper, we propose an multiple vehicle tracking algorithm using microwave radar. Our proposed system includes various special features. First, exceptional radar measurement model for vehicle, concentrated on the corner, is described by mixture density network (MDN), and applied to particle filter weighting. Also, to conquer the curse of dimensionality of particle filter and estimate the time-varying number of multi-target states, reversible jump markov chain monte carlo (RJMCMC) is used to sampling step of the proposed algorithm. The robustness of the proposed algorithm is demonstrated through several computer simulations.

• Journal of the Korea Convergence Society
• /
• v.9 no.11
• /
• pp.15-21
• /
• 2018

In this study, we try to detect anomalies on the network intrusion detection system by learning only one class. We use KDD CUP 1999 dataset, an intrusion detection dataset, which is used to evaluate classification performance. One class classification is one of unsupervised learning methods that classifies attack class by learning only normal class. When using unsupervised learning, it difficult to achieve relatively high classification efficiency because it does not use negative instances for learning. However, unsupervised learning has the advantage for classifying unlabeled data. In this study, we use one class classifiers based on support vector machines and density estimation to detect new unknown attacks. The test using the classifier based on density estimation has shown relatively better performance and has a detection rate of about 96% while maintaining a low FPR for the new attacks.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.1
• /
• pp.1-11
• /
• 2023

Deep-learning-based image segmentation is one of the most widely employed lane detection approaches, and it requires a post-process for extracting the key points on the lanes. A general approach for key-point extraction is using a fixed threshold defined by a user. However, finding the best threshold is a manual process requiring much effort, and the best one can differ depending on the target data set (or an image). We propose a novel key-point extraction algorithm that automatically adapts to the target image without any manual threshold setting. In our adaptive key-point extraction algorithm, we propose a line-level normalization method to distinguish the lane region from the background clearly. Then, we extract a representative key point for each lane at a line (row of an image) using a kernel density estimation. To check the benefits of our approach, we applied our method to two lane-detection data sets, including TuSimple and CULane. As a result, our method achieved up to 1.80%p and 17.27% better results than using a fixed threshold in the perspectives of accuracy and distance error between the ground truth key-point and the predicted point.

• Land and Housing Review
• /
• v.14 no.2
• /
• pp.35-50
• /
• 2023

The purpose of this study is to identify areas in need of urban renewal by utilizing spatial data and analyzing their types and characteristics. For this, this research employed a kernel density function and K-means cluster analysis with spatial data, through which it sought ways to identify high-demand areas for urban renewal projects. The key findings and implications of the research are summarized as follows. Firstly, this research classified 587 target sites in Seoul based on development density (ratios) and an indicator for aged buildings. Approximately half of these areas were consistent with leading pilot project sites and Accelerated Integration Sites. Secondly, it was observed that residential environments in the designated leading pilot project sites, as decided by public sectors, were relatively poor compared to other areas. Lastly, the target areas for urban renewal were not clearly categorized through statistical analysis. Instead, it was found that categorization should be made depending on the requirements of each project.

• 한국지구물리탐사학회:학술대회논문집
• /
• 2005.05a
• /
• pp.9-14
• /
• 2005

For the understanding the locus of the Quinling-Dabie-Sulu continental collision's boundary and the underground structure of the sedimentray basin in the Yellow Sea, three dimensional density modelling is carrid out by using gravity dataset (Free Air Anomaly), which is measured by Tamhae 2, KIGAM in a period 2000 - 2002. The measured gravity anomaly in the investigations area is mainly responsed by depth distribution of the sedimentary basin. After comparing the sea-measured gravity data to CHAMP-GRACE satellite gravity data, I suggested that the high density model bodies extend mainly from the southern part of China to the middle-western part of the Korean Peninsula., which might be emplaced along the continental collision's boundary. The total volume of very low density bodies modified by modelling might be about $20000\;km^3$.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.37 no.2
• /
• pp.55-64
• /
• 2000

In this paper, we propose an efficient datapath placement algorithm to minimize track density. Here, we consider each datapath element as a cluster, and merge the most strongly connected two clusters to a new cluster until only one cluster remains. As nodes in the two clusters to be merged are already linearly ordered respectively, we can merge two clusters with connecting them. The proposed algorithm produces circular linear ordering by connecting starting point and end point of the final cluster, and n different linear ordering by cutting between two contiguous elements of the circular linear ordering. Among the n different linear ordering, the linear ordering to minimize track density is final solution. In this paper, we show and utilize that if two clusters are strongly connected in a graph, the inner product of the corresponding vectors mapped in d-dimensional space using spectral method is maximum. Compared with previous datapath placement algorithm GA/S $A^{［2］}$, the proposed algorithm gives similar results with much less computation time.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.15 no.2
• /
• pp.207-213
• /
• 1997

The purpose of this study is to aim at presenting efficient technique of 3-D Terrain Modelling through multilateral approach methods and to compare with raw data, using low-densed randomly located point data. The subject religion of this study are selected two sites and take into consideration for degree of freedom about low-densed randomly located point data. The result of this study by precision analysis of digital cartographic map-ping using low-densed randomly located point data bave shown that . First, making digital cartographic map, the technique of making it using low-desned randomly located point data by TIN-based results to good and fast run-time in A and B sites all together. Second, the visualization analysis results of digital cartographic map using TIN and GRID-based terrain modeling techniqus similar exacts A and B sites, but the terrain modeling techniqus by TIN-based are small data size than GRID-based with the data with the data size of saving with DXF files. Third, making digital catographic map using terrain modeling techniques by Grid-based, the standard errors of low-densed randomly located point data and interpolated data using gridding method have more good results by radial basis function interpolation techniques at A and B sites all together.

• Journal of the Korean Geophysical Society
• /
• v.8 no.2
• /
• pp.89-92
• /
• 2005

For the understanding the locus of the Quinling-Dabie-Sulu continental collision’s boundary and the underground structure of the sedimentray basin in the Yellow Sea, three dimensional density modelling is carrid out by using gravity dataset (Free Air Anomaly), which is measured by Tamhae 2, GIGAM in a period 2000-2002. The measured gravity anomaly in the investigations area is mainly responsed by depth distribution of the sedimentary basin. After comparing the sea-measured gravity data to CHAMP-GRACE satellite gravity data, I suggested that the high density model bodies extend mainly from the southern part of China to the middle-western part of the Korean Peninsula, which might be emplaced along the continental collision’s boundary. The total volume of very low density bodies modified by modelling might be about 20 000 km3.

• Aerospace Engineering and Technology
• /
• v.11 no.2
• /
• pp.87-95
• /
• 2012

This technical paper describes the analysis results of telemetry data for the initial activation of star trackers for an agile high accuracy low earth orbit satellite. The satellite was recently launched and is in the Launch and Early Operation Phases. It uses two SED36 star trackers manufactured by SODERN. The star tracker is separated by three parts, an optical head, an electronics box, and a baffle with maintaining optical head base plate temperature 20 degC in order to achieve the better performance in low frequency error. This paper presents the initial activation status, requirements and performance, anomaly occurrence, and noise equivalent angle performance analysis during the mission mode by processing the telemetry data.