• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05a
• /
• pp.543-546
• /
• 2003

본 논문에서는 최근 영상 처리 및 검색 분야에서 많이 활용되고 있는 웨이브렛 변환과 원 영상의 영역 분류를 이용하여 밝기가 변화된 영상과 회전된 영상의 검색이 가능한 알고리즘을 제안하였다. 제안한 방식을 통해 영상 전제에 대해 검색이 수행되지 않고 영역 분류 결과인 블록맵과 변환 대역에서의 분산값을 이용함으로써 적은 양의 정보만을 저장하고, 이를 기반으로 영상 검색을 수행함으로써 검색속도의 향상과 효율적인 검색이 가능함을 실험을 통해 확인하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.6 no.8
• /
• pp.1139-1145
• /
• 2002

Wavelet transformed data can filter signal with each frequency band, because it includes detail information about original signal. Therefore, in this paper, important two noises were removed by wavelet. About AWGN environment UDWT(undecimated discrete wavelet transform), applying hard-threshold, was used and about impulse noise environment, it can be possible to recognize edge of original signal as well as superior denoising effect by using two methods, denoising by threshold and slope of signal by wavelet. SNR was used as a judgemental criterion of a denoising effect and Blocks and DTMF(dual tone multi frequency) were used as a test signal.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.3
• /
• pp.289-294
• /
• 2002

A valid speech-sound block can be classified to provide important information for speech recognition. The classification of the speech-sound block comes from the MRA(multi-resolution analysis) property of the DWT(discrete wavelet transform), which is used to reduce the computational time for the pre-processing of speech recognition. The merging algorithm is proposed to extract valid speech-sounds in terms of position and frequency range. It needs some numerical methods for an adaptive DWT implementation and performs unvoiced/voiced classification and denoising. Since the merging algorithm can decide the processing parameters relating to voices only and is independent of system noises, it is useful for extracting valid speech-sounds. The merging algorithm has an adaptive feature for arbitrary system noises and an excellent denoising SNR(signal-to-nolle ratio).

• 전자공학회논문지 IE
• /
• v.43 no.4
• /
• pp.65-75
• /
• 2006

This thesis deals with a wavelet-based method of denoising of infrared images contaminated with impulse noise and Gaussian noise, he method of thresholding the wavelet coefficients using derivatives and median absolute deviations of the wavelet coefficients of the detail subbands was proposed to effectively denoise infrared images with noises. Particularly, in order to eliminate the impulse noise the method of generating binary masks indicating locations of the impulse noise was selected. By this method, the threshold values dividing edges and noises were obtained more effectively proving the validity of the denoising method compared with the conventional wavelet shrinkage method.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.4
• /
• pp.76-85
• /
• 2011

In this paper, we propose a texture feature-based language identification using Gabor feature and wavelet-domain BDIP (block difference of inverse probabilities) and BVLC (block variance of local correlation coefficients) features. In the proposed method, Gabor and wavelet transforms are first applied to a test image. The wavelet subbands are next denoised by Donoho's soft-thresholding. The magnitude operator is then applied to the Gabor image and the BDIP and BVLC operators to the wavelet subbands. Moments for Gabor magnitude image and each subband of BDIP and BVLC are computed and fused into a feature vector. In classification, the WPCA (whitened principal component analysis) classifier, which is usually adopted in the face identification, searches the training feature vector most similar to the test feature vector. Experimental results show that the proposed method yields excellent language identification with rather low feature dimension for a document image DB.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.8
• /
• pp.1901-1906
• /
• 2010

This classification of walking patterns is important and many kinds of applications. Therefore, we attempted to classify walking on level ground from slow walking to fast walking using a waist acceleration signal. A tri-axial accelerometer was fixed to the subject's waist and the three acceleration signals were recorded by bluetooth module at a sampling rate of 100 Hz eleven healthy. The data were analyzed using discrete wavelet transform. Walking patterns were classified using two parameters; One was the ratio between the power of wavelet coefficients which were corresponded to locomotion and total power in the anteroposterior direction (RPA). The other was the ratio between root mean square of wavelet coefficients at the anteroposterior direction and that at the vertical direction(RAV). Slow walking could be distinguished by the smallest value in RPA from other walking pattern. Fast walking could be discriminated from level walking using RAV. It was possible to classify the walking pattern using acceleration signal in healthy people.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 1995.10a
• /
• pp.741-745
• /
• 1995

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 1995.10a
• /
• pp.205-208
• /
• 1995

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 1996.06a
• /
• pp.84-88
• /
• 1996

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 1995.06a
• /
• pp.416-420
• /
• 1995