• Title/Summary/Keyword: Log-Ratio 방법

Search Result 107, Processing Time 0.024 seconds

Efficient Speech Enhancement based on left-right HMM with State Sequence Decision Using LRT (좌-우향 은닉 마코프 모델에서 상태결정을 이용한 음질향상)

  • 이기용
    • The Journal of the Acoustical Society of Korea
    • /
    • v.23 no.1
    • /
    • pp.47-53
    • /
    • 2004
  • We propose a new speech enhancement algorithm based on left-right Hidden Markov Model (HMM) with state decision using Log-likelihood Ratio Test (LRT). Since the conventional HMM-based speech enhancement methods try to improve speech quality for all states, they introduce huge computational loads inappropriate to real-time implementation. In the left-right HMM, only the current and the next state are considered for a possible state transition so to reduce the computational complexity. In this paper, we propose a method to decide the current state by using the LRT on the previous state. Experimental results show that the proposed method improves the speed up to 60% with 0.2∼0.4 dB degradation of speech quality compared to the conventional method.

A Fast Fractal Image Compression Using The Normalized Variance (정규화된 분산을 이용한 프랙탈 압축방법)

  • Kim, Jong-Koo;Hamn, Do-Yong;Wee, Young-Cheul;Kimn, Ha-Jine
    • The KIPS Transactions:PartA
    • /
    • v.8A no.4
    • /
    • pp.499-502
    • /
    • 2001
  • Fractal image coding suffers from the long search time of domain pool although it provides many properties including the high compression ratio. We find that the normalized variance of a block is independent of contrast, brightness. Using this observation, we introduce a self similar block searching method employing the d-dimensional nearest neighbor searching. This method takes Ο(log/N) time for searching the self similar domain blocks for each range block where N is the number of domain blocks. PSNR (Peak Signal Noise Ratio) of this method is similar to that of the full search method that requires Ο(N) time for each range block. Moreover, the image quality of this method is independent of the number of edges in the image.

  • PDF

A Statistical Image Segmentation Method in the Hierarchical Image Structure (계층적 영상구조에서 통계적 방법에 의한 영상분할)

  • 최성진
    • Journal of Broadcast Engineering
    • /
    • v.1 no.2
    • /
    • pp.165-175
    • /
    • 1996
  • In this paper, the image segmentation method based on the hierarchical pyramid image structure of reduced resolution versions of the image for solving the problems in the conventional methods is presented. This method is described the object detection and delineation by statistical approach. In the object detection method, IFSVR( Inverse-father-son variance ratio) method and FSVR(father-son variance ratio ) method are proposed for solving clustering validity problem occurred In the hierarchical pyramid image structure. An optimal object pixel Is detected at some level by this method. In the object delineation method, the iterative algorithm by top-down traversing method is proposed for moving the optimal object pixel to levels of higher resolution. Using the computer simulation, the results by the proposed statistical methods and object traversing method are investigated for the binary Image and the real image. At the results of computer simulation, the proposed methods of image segmentation based on the hierarchical pyramid Image structure seem to have useful properties and deserve consideration as a possible alternative to existing methods of image segmentation. The computation for the proposed method is required 0(log n) for n${\times}$n input image.

  • PDF

The Study on CMA using field terminator (Field terminator를 이용한 CMA 제작에 관한 연구)

  • 이충만;성면창;권순남;정광호
    • Journal of the Korean Vacuum Society
    • /
    • v.5 no.4
    • /
    • pp.278-283
    • /
    • 1996
  • We constructed a single path cylindrical mirror analyser(CMA) using field terminator methods. With computer simulation, the best fit voltage ratio and position of the field terminaor was determined with maintain log-scale equipotential line near both end of the CMA. Then we construct field terminator with voltage divider of metal-oxide resisters which reduces the fringe field effect. The resolving power of the CMA was better than $\Delta$E/E=0.4%.

  • PDF

Application of LLR on Cooperative Communications for Wireless Relay Networks (무선 중계 네트워크의 협력 통신 방법에 대한 LLR 적용 연구)

  • Khuong Ho Van;Kong Hyung-Yun;Lee Dong-Un
    • The KIPS Transactions:PartC
    • /
    • v.13C no.5 s.108
    • /
    • pp.601-606
    • /
    • 2006
  • Decode-and-forward cooperative communications protocol (DFP) allows single-antenna users in wireless medium to obtain the powerful benefits of multi-antenna systems without physical antenna arrays. For this protocol, so far the relays have used SNR to evaluate the reliability of the received signal before deciding whether to forward the decoded data so as to prevent their unsuccessful detection. However, SNR only characterizes the long-term statistic of Gaussian noise and thus leading to inaccurate assesment. Therefore, we propose using log-likelihood ratio (LLR) which accounts for the instantaneous noise in the received signal as an alternative to SNR. A variety of simulation results reveal the significant superiority of the SNR-based DFP to the SNR-based DFP regardless of threshold level and relay position under the flat Rayleigh fading channel plus AWGN (Additive White Gaussian Noise).

Cepstral Distance and Log-Energy Based Silence Feature Normalization for Robust Speech Recognition (강인한 음성인식을 위한 켑스트럼 거리와 로그 에너지 기반 묵음 특징 정규화)

  • Shen, Guang-Hu;Chung, Hyun-Yeol
    • The Journal of the Acoustical Society of Korea
    • /
    • v.29 no.4
    • /
    • pp.278-285
    • /
    • 2010
  • The difference between training and test environments is one of the major performance degradation factors in noisy speech recognition and many silence feature normalization methods were proposed to solve this inconsistency. Conventional silence feature normalization method represents higher classification performance in higher SNR, but it has a problem of performance degradation in low SNR due to the low accuracy of speech/silence classification. On the other hand, cepstral distance represents well the characteristic distribution of speech/silence (or noise) in low SNR. In this paper, we propose a Cepstral distance and Log-energy based Silence Feature Normalization (CLSFN) method which uses both log-energy and cepstral euclidean distance to classify speech/silence for better performance. Because the proposed method reflects both the merit of log energy being less affected with noise in high SNR and the merit of cepstral distance having high discrimination accuracy for speech/silence classification in low SNR, the classification accuracy will be considered to be improved. The experimental results showed that our proposed CLSFN presented the improved recognition performances comparing with the conventional SFN-I/II and CSFN methods in all kinds of noisy environments.

Text Categorization Based on the Maximum Entropy Principle (최대 엔트로피 기반 문서 분류기의 학습)

  • 장정호;장병탁;김영택
    • Proceedings of the Korean Information Science Society Conference
    • /
    • 1999.10b
    • /
    • pp.57-59
    • /
    • 1999
  • 본 논문에서는 최대 엔트로피 원리에 기반한 문서 분류기의 학습을 제안한다. 최대 엔트로피 기법은 자연언어 처리에서 언어 모델링(Language Modeling), 품사 태깅 (Part-of-Speech Tagging) 등에 널리 사용되는 방법중의 하나이다. 최대 엔트로피 모델의 효율성을 위해서는 자질 선정이 중요한데, 본 논문에서는 자질 집합의 선택을 위한 기준으로 chi-square test, log-likelihood ratio, information gain, mutual information 등의 방법을 이용하여 실험하고, 전체 후보 자질에 대한 실험 결과와 비교해 보았다. 데이터 집합으로는 Reuters-21578을 사용하였으며, 각 클래스에 대한 이진 분류 실험을 수행하였다.

  • PDF

Linear Regression-based 1D Invariant Image for Shadow Detection and Removal in Single Natural Image (단일 자연 영상에서 그림자 검출 및 제거를 위한 선형 회귀 기반의 1D 불변 영상)

  • Park, Ki-Hong
    • Journal of Digital Contents Society
    • /
    • v.19 no.9
    • /
    • pp.1787-1793
    • /
    • 2018
  • Shadow is a common phenomenon observed in natural scenes, but it has a negative influence on image analysis such as object recognition, feature detection and scene analysis. Therefore, the process of detecting and removing shadows included in digital images must be considered as a pre-processing process of image analysis. In this paper, the existing methods for acquiring 1D invariant images, one of the feature elements for detecting and removing shadows contained in a single natural image, are described, and a method for obtaining 1D invariant images based on linear regression has been proposed. The proposed method calculates the log of the band-ratio between each channel of the RGB color image, and obtains the grayscale image line by linear regression. The final 1D invariant images were obtained by projecting the log image of the band-ratio onto the estimated grayscale image line. Experimental results show that the proposed method has lower computational complexity than the existing projection method using entropy minimization, and shadow detection and removal based on 1D invariant images are performed effectively.

Identification Characteristics of Irradiated Dried Red Pepper during Storage by Analysis of Thermoluminescence, DNA Comet, and DEFT/APC (Thermoluminescence, DNA Comet 및 DEFT/APC 분석에 의한 방사선처리 건고추의 저장 중 검지 특성)

  • Kim, Byeong-Keun;Kwon, Joong-Ho
    • Korean Journal of Food Science and Technology
    • /
    • v.36 no.6
    • /
    • pp.851-856
    • /
    • 2004
  • Minerals separated from irradiated dried red pepper (whole) at 2.5 kGy or higher showed typical thermoluminescence (TL) glow curves ($TL_1$) at around $150^{\circ}C$, which increased with irradiation dose. The TL ratio ($TL_1/TL_2$) through re-irradiation step at 1 kGy enhanced reliability of TL identification results. DNA comet assay indicated that the intact cell was observed in non-irradiated pepper (seed), while some long tails were found in irradiated ones, showing relationship between irradiation dose and tail length. Log DEPT/APC values increased in proportion to irradiation doses in powdered and whole peppers. Based on overall results, irradiated dried red peppers could be screened using DNA comet assay or log DEFT/APC, and moreover the identification results were verified by TL analysis.

Reducing peak-to-average-power-ratio for OFDM system by using complementary code scrambling (Complementary Code Scrambling을 이용한 OFDM 시스템의 Peak-to-Average-Power-Ratio 감소 방안)

  • 권기범;오성근
    • Proceedings of the IEEK Conference
    • /
    • 1998.10a
    • /
    • pp.191-194
    • /
    • 1998
  • 본 논문에서는 OFDM 시스템에서 사용되는 부 반송파의 수와 동일한 길이의 complementary code를 사용하여 각 부 반송파 데이터를 frequency 영역에서 scrambling하는 PAPR 감소 방안을 제안한다. 이 방법은 부 반송파의 수와 동일한 길이의 complementary code 집합을 이용하여 주어진 데이터에 부과해서 PAPR 값이 가장 적은 complementary code를 사용하여 scrambled된 데이터를 송신함으로서 PAPR 값을 감소시키는 방법이다. 수신기에서의 복호를 위하여 Nc개의 complementary code에 해당하는 log2Nc 비트의 code 인덱스를 데이터와 함께 전송한다. 이때 code 인덱스 비트가 PAPR 값에 영향을 주지 않도록 하기 위하여 CODE인덱스와 데이터가 함께 complementary code로 scrambling되어진다. 제안된 이 방법을 사용하면 주파수 대역효율을 높이면서 PAPR 값을 크게 줄일 수 있다.

  • PDF