• The Journal of the Acoustical Society of Korea
• /
• v.24 no.4
• /
• pp.224-229
• /
• 2005

In this paper, we Propose a novel cepstrum normalization method which normalizes the probability density function (pdf) of cepstrum for robust speech recognition in additive noise environments. While the conventional methods normalize the first- and/or second-order statistics such as the mean and/or variance of the cepstrum. the proposed method fully normalizes the statistics of cepstrum by making the pdfs of clean and noisy cepstrum identical to each other For the target Pdf, the generalized Gaussian distribution is selected to consider various densities. In recognition phase, we devise a table lookup method to save computational costs. From the speaker-independent isolated-word recognition experiments, we show that the Proposed method gives improved Performance compared with that of the conventional methods, especially in heavy noise environments.

• KIPS Transactions on Computer and Communication Systems
• /
• v.6 no.1
• /
• pp.25-30
• /
• 2017

The influence of the data deletion method which is one of anti-forensic techniques is substantial in terms of forensic analysis compared to its simplicity of the act. In academic world, recovery techniques on deleted files have been continuously studied in response to the data deletion method and representatively, the file system-based file recovery technique and file format based recovery technique exist. If there's metadata of deleted file in file system, the file can be easily recovered by using it, but if there's no metadata, the file is recovered by using the signature-based carving technique or the file format based recovery technique has to be applied. At this time, in the file format based recovery technique, the file structure analysis and possible recovery technique should be provided. This paper proposes the page recovery technique on deleted PDF file based on the structural characteristics of PDF file. This technique uses the tag value of page object which constitutes one page of PDF file. Object is extracted by utilizing each tag value as a kind of signature and by analyzing extracted object, the metadata of PDF file is recombined and then it's reconfigured page by page. Recovering by page means that even if deleted PDF file is damaged, even some pages consisting of PDF file can be recovered. Generally, if the file system based file is not recoverable, deleted file is recovered by applying the signature based carving technique. The technique which we proposed in this paper can recover PDF files that are damaged. In the digital forensic perspective, it can be utilized to recover more data than previously.

• International Journal of Knowledge Content Development & Technology
• /
• v.9 no.3
• /
• pp.23-41
• /
• 2019

This study combines the concept of degree centrality in complex network with the Term Frequency $^*$ Proportional Document Frequency ($TF^*PDF$) algorithm; the combined method, called PMCN (PDF-Modified similarity and Complex Network), constructs relationship networks among sentences for writing news summaries. The PMCN method is a multi-document summarization extension of the ideas of Bun and Ishizuka (2002), who first published the $TF^*PDF$ algorithm for detecting hot topics. In their $TF^*PDF$ algorithm, Bun and Ishizuka defined the publisher of a news item as its channel. If the PDF weight of a term is higher than the weights of other terms, then the term is hotter than the other terms. However, this study attempts to develop summaries for news items. Because the $TF^*PDF$ algorithm summarizes daily news, PMCN replaces the concept of "channel" with "the date of the news event", and uses the resulting chronicle ordering for a multi-document summarization algorithm, of which the F-measure scores were 0.042 and 0.051 higher than LexRank for the famous d30001t and d30003t tasks, respectively.

• Journal of Mechanical Science and Technology
• /
• v.18 no.8
• /
• pp.1470-1478
• /
• 2004

In this paper, a new 'presumed' Probability Density Function (PDF) approach coupled with a Lagrangian tracking method is proposed for turbulent combustion modeling. The test and the investigation of the model are conducted by comparing the model results with DNS data for a premixed flame subjected in a decaying turbulent field. The newly constructed PDF, which incorporates the instantaneous chemical reaction term, demonstrates consistent improvement over conventional assumed PDF models. It has been found that the time evolution of the mean scalar, the variance and the mean reaction rate are strongly influenced by a parameter deduced by a Lagrangian equation which takes into account explicitly the local reaction rate. Tests have been performed for a moderate Damkohler number, and it is expected the model may cover a broader range of Damkohler number. The comparison with the DNS data demonstrates that the proposed model may be promising and affordable for implementation in a moment-equation solver.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.13 no.1
• /
• pp.12-18
• /
• 1988

It is found that the estimation pdf of the real intraframe DPCM prediction error by $x^3$ goodness-of-fit test for the images is nearer gamma distribution that laplacian. Also the new pdf estimation method by NMAE is proposed and applied to the pdf adaptive DPCM system.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.16 no.3
• /
• pp.301-309
• /
• 2003

The non linear control algorithm with actuator saturation for a randomly excited oscillator has been widely explored and has shown promising results, but the probabilistic analysis of the algorithm has been rarely made due to its non-linear nature and the fact that the analytical solution of probability density function (PDF) for controlled responses does not exist. In this paper, a method for the probabilistic analysis on the non linear control algorithm with actuator saturation is proposed based on the equivalent non linear system method. Numerical examples are given to verify the approximation solution of PDF comparing to a statistically obtained PDF using a Gaussian white noise and a Kanai - Tagimi filtered Gaussian white noise.

• Journal of the Korean Society of Combustion
• /
• v.14 no.2
• /
• pp.32-41
• /
• 2009

The transported probability density function model combined with the consistent finite volume (FV) method has been applied to simulate the turbulent bluff-body reacting flows. To realistically account for the non-isotropic turbulence effects on the turbulent bluff-body reacting flows, the present PDF transport approach is based on the joint velocity- turbulent frequency-composition PDF formulation. The evolution of the fluctuating velocity of a particle is modeled by a simplified Langevin equation and the particle turbulence frequency is represented by the modified Jayesh - Pope model. Effects of molecular diffusion are represented by the interaction by exchange with the mean (IEM) mixing model. To validate this hybrid FV/PDF transport model, the numerical results are compared with experimental data for the turbulent bluff-body reacting flows.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2009.09a
• /
• pp.504-510
• /
• 2009

Dredging and Reclamation which have been conducted steadily for creation of new coastal area have demerit that takes a long time. Hence, a lot of researches on acceleration of self-weight consolidation is proceeding continuously. In this paper, 30 cases of laboratory self-weight consolidation tests were conducted to understand application of PDF method, one of the self-weight consolidation acceleration methods, to domestic dredged soils. Acceleration effect of self-weight consolidation was confirmed through comparison and analysis on completion time and settlement of self-weight consolidation for none installed case and 4 kinds of common used PBD installed cases.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.05a
• /
• pp.233-235
• /
• 2011

Because of various style-sheet and Korean-Fonts, converting from WYSIWYG types of HTML data to forms of PDF has many limitations. In print-work areas, for printing specific fonts, it has to use CMYK colors instead of RGB colors. This paper describes about the process to make web font to printing PDF files, which is high resolution image captured document based on HTML, by using COM component feature of hardware.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.317-320
• /
• 2002

In this paper, an automatic histogram specification method is proposed for image enhancement, Fuzzy membership value is adopted for the representation of image histogram. The desired PDF is automatically constructed by the fuzzy membership value. Fuzzy membership value is extracted from dark membership, bright membership function and original histogram. The effectual results are demonstrated by desired PDF which meet the image enhancement requirements. The performance and effectiveness are shown by the analysis and the resultant image in comparison with histogram equalization method.