• The Transactions of the Korea Information Processing Society
• /
• v.2 no.3
• /
• pp.382-389
• /
• 1995

In this paper, we propose an efficient transmission schedule which can be used in indoor wireless LAN. It reduces considerably the time delay and increases the throughput by reusing the bandwidth. We describe the architecture of the wireless LAN, the algorithm of step-by-step allocation of time slot reusing the resource and the results of the computer simulation.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2003.06a
• /
• pp.62-65
• /
• 2003

We propose an algorithm for almost unsupervised detection of linear structures, in particular, axes in road network and river, as seen in synthetics aperture radar (SAR) images. The first is local step and used to extract linear features from the speckle radar image, which are treated as road segment candidates. We present two local line detectors as well as a method for fusing information from these detectors. The second is global step, we identify the real roads among the segment candidates by defining a Markov random field (MRF) on a set of segments, which introduces contextual knowledge about the shape of road objects.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.12a
• /
• pp.121-124
• /
• 2000

We propose an algorithm for almost unsupervised detection of linear structures, in particular, axes in road network and river, as seen in synthetics aperture radar (SAR) images. The first is local step and used to extract linear features from the speckle radar image, which are treated as road segment candidates. We present two local line detectors as well as a method for fusing Information from these detectors. The second is hybrid step, we Identify the real roads among the segment candidates by defining a Markov random field (MRF) on a set of segments, which introduces contextual knowledge about the shape of road objects.

• Journal of Information Processing Systems
• /
• v.14 no.1
• /
• pp.150-161
• /
• 2018

We propose a novel deep neural network model for detecting human activities in untrimmed videos. The process of human activity detection in a video involves two steps: a step to extract features that are effective in recognizing human activities in a long untrimmed video, followed by a step to detect human activities from those extracted features. To extract the rich features from video segments that could express unique patterns for each activity, we employ two different convolutional neural network models, C3D and I-ResNet. For detecting human activities from the sequence of extracted feature vectors, we use BLSTM, a bi-directional recurrent neural network model. By conducting experiments with ActivityNet 200, a large-scale benchmark dataset, we show the high performance of the proposed DeepAct model.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.2
• /
• pp.91-95
• /
• 2011

In this paper, we present an improved multi-scale gradient algorithm. The proposed algorithm works the effectively handling of both step and blurred edges. In the proposed algorithm, the image sharpening operator is sharpening the edges and contours of the objects. This operation gives an opportunity to get noise reduced image and step edged image. After that, multi-scale gradient operator works on noise reduced image in order to get a gradient image. The gradient image is segmented by watershed transform. The approach of region merging is used after watershed transform. The region merging is carried out according to the region area and region homogeneity. The region number of the proposed algorithm is 36% shorter than that of the existing algorithm because the proposed algorithm produces a few irrelevant regions. Moreover, the computational time of the proposed algorithm is relatively fast in comparison with the existing one.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.08a
• /
• pp.221-224
• /
• 2000

We propose an algorithm for almost unsupervised detection of linear structures, in particular, axes in road network and river, as seen in synthetics aperture radar (SAR) images. The first is local step and used to extract linear features from the speckle radar image, which are treated as road segment candidates. We present two local line detectors as well as a method for fusing information from these detectors. The second is global step, we identify the real roads among the segment candidates by defining a Markov random field (MRF) on a set of segments, which introduces contextual knowledge about the shape of road objects.

• Journal of Information Processing Systems
• /
• v.17 no.4
• /
• pp.772-786
• /
• 2021

The process of tracking suspicious behavior manually on a system and gathering evidence are labor-intensive, variable, and experience-dependent. The system logs are the most important sources for evidences in this process. However, in the Microsoft Windows operating system, the action events are irregular and the log structure is difficult to audit. In this paper, we propose a model that overcomes these problems and efficiently analyzes Microsoft Windows logs. The proposed model extracts lists of both common and key events from the Microsoft Windows logs to determine detailed actions. In addition, we show an approach based on the proposed model applied to track illegal file access. The proposed approach employs three-step tracking templates using Elastic Stack as well as key-event, common-event lists and identify event lists, which enables visualization of the data for analysis. Using the three-step model, analysts can adjust the depth of their analysis.

• Annual Conference of KIPS
• /
• 2013.05a
• /
• pp.474-475
• /
• 2013

In healthcare systems, the accuracy of a classifier for classifying medical diseases depends on a reduced dataset. Key to achieve true classification results is the reduction of data to a set of optimal number of significant features. The initial step towards data reduction is the integration of heterogeneous data sources to a unified reduced dataset which is further reduced by considering the range of values of all the attributes and then finally filtering and dropping out the least significant features from the dataset. This paper proposes a three step data reduction model which plays a vital role in the classification process.

• Annual Conference of KIPS
• /
• 2008.11a
• /
• pp.691-694
• /
• 2008

인간의 지식을 컴퓨터에 저장하기 위한 연구 중 하나로 온톨로지가 등장하게 되었다. 온톨로지는 최근 주목 받고 있는 시맨틱 웹을 구현하기 위해 필수적으로 필요할 뿐만 아니라 다른 형식으로 저장된 정보 사이의 공유를 위해서도 많은 연구가 진행되고 있다. 특히 온톨로지의 초기 설계 후 변경을 하기 위해서는 매우 큰 비용이 소모된다는 점에서 온톨로지 설계에 대한 연구 역시 진행되고 있다. 본 논문에서는 온톨로지 설계를 위한 단계적인 온톨로지 정규화 기법을 제안한다. 제시된 온톨로지 정규화는 기존의 제시된 데이터베이스 정규화 기법을 활용하여, 단계적으로 더 엄격한 정규화가 가능하도록 설계되어 있다. 이러한 정규화 기법의 적용을 통해서 비효율적인 온톨로지 설계를 수정하고 짜임새 있는 설계가 가능하도록 온톨로지를 수정하는 기법을 제안한다.

• Journal of KIISE:Databases
• /
• v.30 no.4
• /
• pp.381-396
• /
• 2003

Subsequence matching is an operation that finds subsequences whose changing patterns are similar to a given query sequence from time-series databases. This paper points out the performance bottleneck in subsequence matching, and then proposes an effective method that improves the performance of entire subsequence matching significantly by resolving the performance bottleneck. First, we analyze the disk access and CPU processing times required during the index searching and post processing steps through preliminary experiments. Based on their results, we show that the post processing step is the main performance bottleneck in subsequence matching, and them claim that its optimization is a crucial issue overlooked in previous approaches. In order to resolve the performance bottleneck, we propose a simple but quite effective method that processes the post processing step in the optimal way. By rearranging the order of candidate subsequences to be compared with a query sequence, our method completely eliminates the redundancy of disk accesses and CPU processing occurred in the post processing step. We formally prove that our method is optimal and also does not incur any false dismissal. We show the effectiveness of our method by extensive experiments. The results show that our method achieves significant speed-up in the post processing step 3.91 to 9.42 times when using a data set of real-world stock sequences and 4.97 to 5.61 times when using data sets of a large volume of synthetic sequences. Also, the results show that our method reduces the weight of the post processing step in entire subsequence matching from about 90% to less than 70%. This implies that our method successfully resolves th performance bottleneck in subsequence matching. As a result, our method provides excellent performance in entire subsequence matching. The experimental results reveal that it is 3.05 to 5.60 times faster when using a data set of real-world stock sequences and 3.68 to 4.21 times faster when using data sets of a large volume of synthetic sequences compared with the previous one.