• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.7
• /
• pp.495-502
• /
• 2013

Motion estimation (ME) that limits the performance of image quality and encoding speed has been developed to reduce temporal redundancy in video sequences and plays an important role in digital video compression. But it is computational demanding part of the encoder. Multi-view video is obtained by capturing one three-dimensional scene with many cameras at different positions. ME for Multi-view video requires high computational complexity. To reduce computational complexity and maintain the image quality, a fast motion estimation method is proposed in this paper. The proposed method uses a hierarchical search strategy. This strategy method consists of modified diamond search patten, multi gird diamond search pattern, and raster search pattern. These search patterns place search points symmetrically and evenly that can cover the overall search area not to fall into the local minimum or exploits the characteristics of the distribution of motion vectors to place the search points. Experiment results show that the speedup improvement of the proposed method over TZ search method (JMVC) can be up to 1.2 ~3 times faster while maintaining similar video quality and bit rates.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2004.05a
• /
• pp.673-676
• /
• 2004

Fast and high-quality document clustering algorithms play an important role in providing data exploration by organizing large amounts of information into a small number of meaningful clusters. In particular, hierarchical clustering provide a view of the data at different levels, making the large document collections are adapted to people's instinctive and interested requires. Many papers have shown that the hierarchical clustering method takes good-performance, but is limited because of its quadratic time complexity. In contrast, K-means has a time complexity that is linear in the number of documents, but is thought to produce inferior clusters. Think of the factor of simpleness, high-quality and high-efficiency, we combine the two approaches providing a new system named CONDOR system [10] with hierarchical structure based on document clustering using K-means algorithm to "get the best of both worlds". The performance of CONDOR system is compared with the VIVISIMO hierarchical clustering system [9], and performance is analyzed on feature words selection of specific topics and the optimum hierarchy depth.

• KSBB Journal
• /
• v.16 no.4
• /
• pp.381-388
• /
• 2001

A program for integrated gene expression profile analysis such as hierarchical clustering, K-means, fuzzy c-means, self-organizing map(SOM), principal component analysis(PCA), and singular value decomposition(SVD) was made for DNA chip data anlysis by using Matlab. It also contained the normalization method of gene expression input data. The integrated data anlysis program could be effectively used in DNA chip data analysis and help researchers to get more comprehensive analysis view on gene expression data of their own.

• Journal of Information Processing Systems
• /
• v.16 no.4
• /
• pp.820-831
• /
• 2020

Extractive document summarization aims to select a few sentences while preserving its main information on a given document, but the current extractive methods do not consider the sentence-information repeat problem especially for news document summarization. In view of the importance and redundancy of news text information, in this paper, we propose a neural extractive summarization approach with joint sentence semantic clipping and selection, which can effectively solve the problem of news text summary sentence repetition. Specifically, a hierarchical selective encoding network is constructed for both sentence-level and document-level document representations, and data containing important information is extracted on news text; a sentence extractor strategy is then adopted for joint scoring and redundant information clipping. This way, our model strikes a balance between important information extraction and redundant information filtering. Experimental results on both CNN/Daily Mail dataset and Court Public Opinion News dataset we built are presented to show the effectiveness of our proposed approach in terms of ROUGE metrics, especially for redundant information filtering.

• Korea Information Processing Society Review
• /
• v.11 no.1
• /
• pp.8-19
• /
• 2004

Many Researchers and practitioners have expressed the view that artificial intelligence(AI) may have significant application to solution of manufacturing problems. Expert systems have been developed for solving problem areas.(omitted)

• The Journal of the Korea institute of electronic communication sciences
• /
• v.10 no.9
• /
• pp.1059-1064
• /
• 2015

To write a medical record using SNOMED CT standard clinical terminologies, it is necessary to find and select an appropriate terminology from the huge volume of terminologies within short time. Using previous SNOMED CT search browsers, it is very difficult to select appropriate one from search results since they provide a simple list-up of similar candidate terminologies. This paper proposes a novel search browser which supports effect searching of clinical terminology by utilizing characteristics of SNOMED CT. The proposed system provides a simplified tree-view representing hierarchical structures of search results which enables fast selection of appropriate terminology from the search results. Design and Implementation of the system proves effectiveness of the proposed approach.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1999.06a
• /
• pp.49-56
• /
• 1999

In this paper, the general formula of disparity estimation based on Bayesian Maximum A Posteriori (MAP) algorithm is derived. The generalized formula is implemented with the plane configuration model and applied to multiple baselined stereograms. The probabilistic plane configuration model consists of independence and similarity among the neighboring disparities in the configuration. The independence probabilistic model reduces the computation and guarantees the discontinuity at the object boundary region. The similarity model preserves the continuity or the high correlation of disparity distribution. In addition, we propose a hierarchical scheme of disparity compensation in the application to multiple-view stereo images. According to the experiments, the derived formula and the proposed estimation algorithm outperformed other ones. The proposed probabilistic models are reasonable and approximate the pure joint probability distribution very well with decreasing the computations to O(n(D)) from O(n(D)4) of the generalized formula. And, the hierarchical scheme of disparity compensation with multiple-view stereos improves the performance without any additional overhead to the decoder.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.9
• /
• pp.2550-2572
• /
• 2023

Epidemiological survey is an important means for the prevention and control of infectious diseases. Due to the particularity of the epidemic survey, 1) epidemiological survey in epidemic prevention and control has a wide range of people involved, a large number of data collected, strong requirements for information disclosure and high timeliness of data processing; 2) the epidemiological survey data need to be disclosed at different institutions and the use of data has different permission requirements. As a result, it easily causes personal privacy disclosure. Therefore, traditional access control technologies are unsuitable for the privacy protection of epidemiological survey data. In view of these situations, we propose a black box-assisted fine-grained hierarchical access control scheme for epidemiological survey data. Firstly, a black box-assisted multi-attribute authority management mechanism without a trusted center is established to avoid authority deception. Meanwhile, the establishment of a master key-free system not only reduces the storage load but also prevents the risk of master key disclosure. Secondly, a sensitivity classification method is proposed according to the confidentiality degree of the institution to which the data belong and the importance of the data properties to set fine-grained access permission. Thirdly, a hierarchical authorization algorithm combined with data sensitivity and hierarchical attribute-based encryption (ABE) technology is proposed to achieve hierarchical access control of epidemiological survey data. Efficiency analysis and experiments show that the scheme meets the security requirements of privacy protection and key management in epidemiological survey.

• Korean Journal of Labor Studies
• /
• v.19 no.1
• /
• pp.65-104
• /
• 2013

This study aims to examine the intermediary roles of employee performance between high performance work system (HPWS) and its operational performance on the resource based view. Taking into account the unit of analysis, this study used a hierarchical linear modeling analysis in order to test rigorously the association between HPWS at the organisational level and employee performance at the individual level. For this empirical test, Human Capital Corporate Panel (HCCP) data including 316 firms and 7,872 respondents (including 923 team leaders) were used. To meet the unit of analysis and test the mediation effect, data at the individual and team level were aggregated into the organisational level. The empirical results show that HPWS have a positive impact on operational performance as well as employee performance such as job satisfaction, organisational commitment and organisational trust. Regarding the mediation effect, job satisfaction and organisational trust mediate between HPWS and operational performance. Theoretical implications are discussed in conclusion.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.11
• /
• pp.2720-2736
• /
• 2013

Multi-view face detection has become an active area for research in the last few years. In this paper, a novel multi-view human face detection algorithm based on improved real Adaboost is presented. Real Adaboost algorithm is improved by weighted combination of weak classifiers and the approximately best combination coefficients are obtained. After that, we proved that the function of sample weight adjusting method and weak classifier training method is to guarantee the independence of weak classifiers. A coarse-to-fine hierarchical face detector combining the high efficiency of Haar feature with pose estimation phase based on our real Adaboost algorithm is proposed. This algorithm reduces training time cost greatly compared with classical real Adaboost algorithm. In addition, it speeds up strong classifier converging and reduces the number of weak classifiers. For frontal face detection, the experiments on MIT+CMU frontal face test set result a 96.4% correct rate with 528 false alarms; for multi-view face in real time test set result a 94.7 % correct rate. The experimental results verified the effectiveness of the proposed approach.