• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.6
• /
• pp.773-778
• /
• 2009

This paper studied the training methods less affected by the emotional variation for the development of the robust speech recognition system. For this purpose, the effect of emotional variations on the speech signal were studied using speech database containing various emotions. The performance of the speech recognition system trained by using the speech signal containing no emotion is deteriorated if the test speech signal contains the emotions because of the emotional difference between the test and training data. In this study, it is observed that vocal tract length of the speaker is affected by the emotional variation and this effect is one of the reasons that makes the performance of the speech recognition system worse. In this paper, vocal tract normalization method is used to develop the robust speech recognition system for emotional variations. Experimental results from the isolated word recognition using HMM showed that the vocal tract normalization method reduced the error rate of the conventional recognition system by 41.9% when emotional test data was used.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.11a
• /
• pp.161-164
• /
• 2011

영상 매칭은 컴퓨터 비전에서 기초적인 기술로써 영상 추적, 물체인식 등 다양한 분양에서 많이 사용되고 있다. 하지만 스케일, 시점변화, 조명 변화에 강인한 매칭점을 찾는 것은 어려운 일이다. 이러한 문제점을 보완하기 위해 SURF(Scale Invariant Feature Transform), SIFT(Speed up Robust Features) 등의 알고리즘이 제안 되었지만, 여전히 조명변화에 불안정하고 정확하지 못한 성능을 보인다. 본 논문에서는 이러한 조명변화에 대한 문제점을 해결하기 위해 히스토그램 평활화를 이용하여 영상을 보정 후, SURF를 통한 영상 매칭을 하였다. 열악한 조명환경 내에서 촬영된 영상에서 SURF를 이용하여 표현자(Descriptor)를 생성 할 때 특징점이 잘 추출되지 않는 문제점을 해결하기 위하여 히스토그램 평활화를 이용하였고, 보정 후 특징점 개수가 많이 증가하는 것을 보여 확인하였다. 기존의 SURF와 개량된 SURF를 조명이 서로 다른 영상간의 매칭 성능을 비교함으로써 제안한 알고리즘의 우수성을 확인하였다

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.01a
• /
• pp.65-66
• /
• 2020

이 논문에서는 NTFS 파일시스템에서 여러 개의 클러스터에 걸쳐서 슬랙 영역에 데이터 숨기기를 수행하는 데이터 숨기기 방법의 강인성을 조사하기 위하여 디스크 조각모음 실험을 수행할 때 필요한 방법을 제안한다. 디스크 조각모음의 실행으로 클러스터의 위치가 변동되기 위해서는 여러 클러스터에 걸친 디스크 조각의 단편화가 일어날 수 있도록 파일을 생성하고 삭제하는 작업을 수행한다. 그 후에 용량이 큰 파일을 복사하여 여러 클러스터에 걸친 단편화된 파일을 작위적으로 생성한다. 단편화된 파일만을 남기고 그 이외의 다른 파일들을 삭제한 후에 디스크조각 모음을 수행한다. 이 실험에 필요한 준비과정과 실험과정을 제안한다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.11C
• /
• pp.1140-1151
• /
• 2003

This paper proposes a face recognition method based on modified Otsu's binarization and Hu moment. Proposed method is robust to brightness, contrast, scale, rotation, and translation changes. As the proposed modified Otsu's binarization computes other thresholds from conventional Otsu's binarization, namely we create two binary images, we can extract higher dimensional feature vector. Here the feature vector has properties of robustness to brightness and contrast changes because the proposed method is based on Otsu's binarization. And our face recognition system is robust to scale, rotation, and translation changes because of using Hu moment. In the perspective of brightness, contrast, scale, rotation, and translation changes, experimental results with Olivetti Research Laboratory (ORL) database and the AR database showed that average recognition rates of conventional well-known principal component analysis (PCA) are 93.2% and 81.4%, respectively. Meanwhile, the proposed method for the same databases has superior performance of the average recognition rates of 93.2% and 81.4%, respectively.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.4 s.304
• /
• pp.59-68
• /
• 2005

This paper proposes a face recognition under varying poses using local area obtained by side-view pose normalization. General normalization methods for face recognition under varying pose have a problem with the information about invisible area of face. Generally this problem is solved by compensation, but there are many cases where the image is distorted or features lost due to compensation .To solve this problem we normalize the face pose in side-view to reduce distortion that happens mainly in areas that have large depth variation. We only use undistorted area, removing the area that has been distorted by normalization. We consider two cases of yaw pose variation and pitch pose variation, and by experiments, we confirm the improvement of recognition performance.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.2
• /
• pp.71-79
• /
• 1997

In this paper, a codeword-dependent distance normalization(CDDN) and an instar-formed fuzzy smoothing of output distribution are proposed for robust estimation of output probabilities in the FVQ(fuzzy vector quantization)-DHMM(discrete hidden Markov model). The FVQ-DHMM is a variant of DHMM in which the state output probability is estimated by the sum oft he product of the output probability and its weighting factor for each codeword on an input vector. As the performance of the FVQ-DHMM is influenced by weighting factor and output distribution from a state, it is required to get a method to get robust estimation of weighting factors and output distribution for each state. From experimental results, the proposed CDDN method has reduced 24% of error rate over the conventional FVQ-DHMM, and also reduced 79% of error rate when the smoothing of output distribution is also applied to the computation of an output probability. These results indicate that the use of CDDN and the fuzzy smoothing of output distribution to the FVQ-DHMM lead to improved recognition, and therefore it may be used as an alternative to the robust estimation of output probabilities for HMMs.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2006.06a
• /
• pp.730-736
• /
• 2006

In this paper, we propose a rotation resistant robust logo embedding watermarking technique. Geometric manipulations make the detection process very complex and difficult. Watermark embedding ill the normalized image directly suffers from smoothing effect due to the interpolation during the image normalization. This can be avoided by estimating the transform parameters using image normalization technique, instead of embedding in the normalized image. Conventional rotation resistant schemes that use full frame transform. In this paper, we adopt $8{\times}8$ block DCT and calculate masking using a spatio-frequency localization of the $8{\times}8$ block DCT coefficients. Experimental results show that the proposed algorithm is robust against rotation process.

• Proceedings of the Korean Environmental Sciences Society Conference
• /
• 1993.10a
• /
• pp.25-25
• /
• 1993

• The Bimonthly Magazine for Agrochemicals and Plant Protection
• /
• v.12 no.1 s.100
• /
• pp.15-21
• /
• 1991

• Journal of the Korean Society for Railway
• /
• v.9 no.3 s.32
• /
• pp.43-47
• /
• 2006