• 제목/요약/키워드: power normalization

검색결과 105건 처리시간 0.027초

Robustness of Face Recognition to Variations of Illumination on Mobile Devices Based on SVM

  • Nam, Gi-Pyo;Kang, Byung-Jun;Park, Kang-Ryoung
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제4권1호
    • /
    • pp.25-44
    • /
    • 2010
  • With the increasing popularity of mobile devices, it has become necessary to protect private information and content in these devices. Face recognition has been favored over conventional passwords or security keys, because it can be easily implemented using a built-in camera, while providing user convenience. However, because mobile devices can be used both indoors and outdoors, there can be many illumination changes, which can reduce the accuracy of face recognition. Therefore, we propose a new face recognition method on a mobile device robust to illumination variations. This research makes the following four original contributions. First, we compared the performance of face recognition with illumination variations on mobile devices for several illumination normalization procedures suitable for mobile devices with low processing power. These include the Retinex filter, histogram equalization and histogram stretching. Second, we compared the performance for global and local methods of face recognition such as PCA (Principal Component Analysis), LNMF (Local Non-negative Matrix Factorization) and LBP (Local Binary Pattern) using an integer-based kernel suitable for mobile devices having low processing power. Third, the characteristics of each method according to the illumination va iations are analyzed. Fourth, we use two matching scores for several methods of illumination normalization, Retinex and histogram stretching, which show the best and $2^{nd}$ best performances, respectively. These are used as the inputs of an SVM (Support Vector Machine) classifier, which can increase the accuracy of face recognition. Experimental results with two databases (data collected by a mobile device and the AR database) showed that the accuracy of face recognition achieved by the proposed method was superior to that of other methods.

연속음성 인식기를 위한 벡터양자화기 기반의 화자정규화 (Vector Quantizer Based Speaker Normalization for Continuos Speech Recognition)

  • 신옥근
    • 한국음향학회지
    • /
    • 제23권8호
    • /
    • pp.583-589
    • /
    • 2004
  • 포만트 등의 음향학적인 정보를 이용하지 않는 연속음성인식 (CSR)을 위한 벡터 양자화기 기반의 화자 정규화 방법을 제안한다. 이 방법은 앞서 제안한 간단한 숫자음 인식기를 위한 화자정규화 방법을 개선한 것으로, 코드북의 크기를 증가시켜 가면서 벡터양자화기를 반복적으로 학습시킴으로써 정규화된 코드북을 구한 다음, 치를 이용하여 시험용화자의 워핑계수를 추정한다. 코드북 생성과 워핑계수 추정을 위해 모음 음소의 집합과 자음과 모음을 포함한 모든 음소의 집합 등 두 가지 음소집합을 이용i,겨 실험하였으며, 추정한 워핑계수에 상응하는 구간선형 워핑함수를 이용하여 인식기의 학습과 시험에 사용될 특징벡터를 워핑하였다. TIMIT 코퍼스와 HTK toolkit을 이용한 음소인식 실험을 수행하여 제안하는 방법의 성능을 조사한 결과, 포만트를 이용한 워핑 방법과 비슷한 성능을 가짐을 확인하였다.

영오차 확률 기반 알고리즘의 입력 정력 정규화 (Input Power Normalization of Zero-Error Probability based Algorithms)

  • 김종일;김남용
    • 한국통신학회논문지
    • /
    • 제42권1호
    • /
    • pp.1-7
    • /
    • 2017
  • 충격성 잡음 환경에서 최대 영확률 (MZEP) 알고리듬은 최소자승오차 (MSE) 기반의 알고리듬 보다 우수한 성능을 지닌다. 그리고 알고리듬 자체에 내재한 크기 조절 입력 (MCI)가 MZEP 알고리듬을 충격성 잡음으로부터 알고리듬을 안정되게 유지하는 역할을 하는 것으로 알려져 있다. 이 논문에서는 MCI 입력의 평균전력으로 MZEP 알고리듬의 스텝 사이즈를 정규화하는 방식을 제안하였다. 충격파 발생률이 0.03인 충격성 잡음하의 시뮬레이션에서 정상상태 MSE 성능 비교에서 기존 MZEP에 비해 제안한 방식이 약 2dB 정도 향상된 특성을 보인다.

역기전력 정규화에 의한 PMSM의 센서리스 제어 (PMSM sensorless control by back emf normalization)

  • 이정준;박성준;김철우
    • 전력전자학회:학술대회논문집
    • /
    • 전력전자학회 2002년도 전력전자학술대회 논문집
    • /
    • pp.300-303
    • /
    • 2002
  • With increase of servo motor In industrial and home application, a number of papers related to PMSM control have been researched. Among them, sensorless control schemes are especially concerned in the view point of its cost reduction. In the conventional approach, a rotor position is generally estimated by the integration of estimated rotor speed. In this method, because of their tight relationship between the amplitude of back-emf and rotor position. it is somewhat difficult to find two parameters at the same time. To solve this problem, a novel sensorless control scheme is proposed. It utilizes a back-emf normalization, so it does not requires the variables related with the amplitude of back-emf. The validity of the proposed control scheme was verified through experimental results.

  • PDF

과변조 영역에서의 전류제어를 가지는 영구자석형 동기 전동기의 속도제어 (Speed Control of PMSM with Current Controller in Overmodulation Region)

  • 박진식;김학원;윤명중
    • 전력전자학회:학술대회논문집
    • /
    • 전력전자학회 2005년도 전력전자학술대회 논문집
    • /
    • pp.251-253
    • /
    • 2005
  • This paper presents a study on a current control of a PM synchronous motor which is fed by two-level voltage source inverter in the high modulation index area. In order to increase dynamic response and extend speed limit caused by voltage limit, overmodulation method is generally used. The current control in overmodulation have voltage limit. So larger voltage command than voltage limit should be diminished. In this paper, the problem conventional normalization method is analyzed and New normalization method in overmodulation region is proposed.

  • PDF

CV케이블의 부분방전 신호를 통한 열화과정의 정량적 진단 (Normalization Diagnosis of Aging Process on Partial Discharge Signals of CV Cable)

  • 소순열;임장섭;김진사;이준웅;김태성
    • 한국전기전자재료학회:학술대회논문집
    • /
    • 한국전기전자재료학회 1997년도 추계학술대회 논문집
    • /
    • pp.451-455
    • /
    • 1997
  • The partial discharge has been blown as the chief breakdown of power equipments. The analysis and the recognition is much difficult because the partial discharge signal is very small and has complex aging pattern. Recently, insulation aging diagnosis based on pattern of phase(Ф), partial discharge magnitude(q), number(n) has been very important. Owing to depreciate the reappearance of aging progress at the electrical tree pattern and to be difficult to analyze visually, the study on partial discharge pattern is suggested to normalizing analysis method of partial discharge signals. This parer is purposed on prediction of life-time measurement of cv-cable, on decision of risk degree with normalization and real-time measurement of partial discharge signals for aging diagnosis of cv-cable. As normalizing the aging signals of electrical tree in cv-cable, it is able to confirm risk degree of insulation material with the distribution of Ф-q-n and recognize the process of aging pattern using neural network.

  • PDF

3MW 풍력발전시스템 출력 성능시험 및 불확도 분석 (Power Performance Testing and Uncertainty Analysis for a 3MW Wind Turbine)

  • 김건훈;현승건
    • 한국태양에너지학회 논문집
    • /
    • 제30권6호
    • /
    • pp.10-15
    • /
    • 2010
  • The installed capacity of wind turbines in KOREA are growing and enlarging by the central government's support program. Thus, the importance of power performance verification and its uncertainty analysis are recognizing rapidly. This paper described the power testing results of a 3MW wind turbine and analysed an uncertainty level of measurements. The measured power curves are very closely coincide with the calculated one and the annual power production under the given Rayleigh wind speed distribution are estimated with the 3.6~12.7% of uncertainty but, in the dominant wind speed region as 7~8m/s, the uncertainty are stably decreased to 6.3~5.3%.

3.0MW 풍력발전기 출력 성능 및 불확실성 분석 (ANALYSIS OF POWER PERFORMANCE AND UNCERTAINTY FOR A 3.0MW WIND TURBINE)

  • 허수영;김경보;허종철
    • 한국전산유체공학회:학술대회논문집
    • /
    • 한국전산유체공학회 2010년 춘계학술대회논문집
    • /
    • pp.28-31
    • /
    • 2010
  • In order to clarify the characteristics of power performance and uncertainty of a wind turbine, an investigation was performed in Hangyeong wind farm, Jeju island, Korea. Data were collected for 12 months from Feb. 2, 2008 to Jan. 1, 2009. This study was conducted on the base of the International standard, and observed the methods of mesurement and evaluation form IEC 61400-12. As a result, power performance curve was calculated by measured data and compared with the sixth unit of VESTAS V90-3.0MW in Hangyeong wind farms. In consequence of this paper, uncertainty was estimated from 7% to 14% on the range of the average wind speed from 4m/s to 11m/s.

  • PDF

실시간 COFDM시스템을 위한 효율적인 구조를 갖는 비터비 디코더 설계 (The viterbi decoder implementation with efficient structure for real-time Coded Orthogonal Frequency Division Multiplexing)

  • 황종희;이승열;김동순;정덕진
    • 대한전자공학회논문지TC
    • /
    • 제42권2호
    • /
    • pp.61-74
    • /
    • 2005
  • 디지털 멀티미디어 방송(DMB)은 대용량의 멀티미디어 정보를 무선환경의 이동체에 전송하기 위해 제안된 방식이다. 이러한 멀티미디어 서비스를 제공하기 위해 DM시스템은 COFDM 변조방식을 사용하여 다중 경로 페이딩 현상을 극복하고, 동시에 강력한 채널오류 정정 능력을 필요로 한다. DMB 수신기를 위한 비터비 디코더(구속장 7, code rate 1/4)는 가변 부호화된 데이터의 복호화를 수행해야 하고, 방송시스템이므로 실시간으로 동작하기 위해서 효율적인 구조를 가져야 한다. 따라서 DMB 시스템을 위한 비터비 디코더를 구현하기 위해서는 복호화 과정을 고속으로 수행할 수 있는 별도의 전용 하드웨어 모듈을 설계하는 것이 바람직하다. 본 논문에서는 많은 연산량을 효율적으로 줄일 수 있는 결합된 Add-Compare-Select(ACS)와 Path Metric Normalization(PMN)구조를 새롭게 제안하고자 한다. PMN구조에서의 단점인 comparison tree에 의한 임계 경로(critical path)의 문제를 고정치(fixed value)에 의한 선택 알고리즘을 적용함으로써 고속 동작이 가능하게 하였고, ACS구조에서는 분할 기법(decomposition method)과 선계산(pre-computation)을 이용하여 덧셈기, 비교기, 표준화기의 복잡도를 줄일 수 있도록 하였다. 시뮬레이션 결과 펑처드 비터비 디코더는 일반적인 구조를 적용했을 때 보다 면적 $3.78\%$, 전력소모 $12.22\%$, 최대 게이트 지연 $23.80\%$의 감소율을 보였다.

Towards Low Complexity Model for Audio Event Detection

  • Saleem, Muhammad;Shah, Syed Muhammad Shehram;Saba, Erum;Pirzada, Nasrullah;Ahmed, Masood
    • International Journal of Computer Science & Network Security
    • /
    • 제22권9호
    • /
    • pp.175-182
    • /
    • 2022
  • In our daily life, we come across different types of information, for example in the format of multimedia and text. We all need different types of information for our common routines as watching/reading the news, listening to the radio, and watching different types of videos. However, sometimes we could run into problems when a certain type of information is required. For example, someone is listening to the radio and wants to listen to jazz, and unfortunately, all the radio channels play pop music mixed with advertisements. The listener gets stuck with pop music and gives up searching for jazz. So, the above example can be solved with an automatic audio classification system. Deep Learning (DL) models could make human life easy by using audio classifications, but it is expensive and difficult to deploy such models at edge devices like nano BLE sense raspberry pi, because these models require huge computational power like graphics processing unit (G.P.U), to solve the problem, we proposed DL model. In our proposed work, we had gone for a low complexity model for Audio Event Detection (AED), we extracted Mel-spectrograms of dimension 128×431×1 from audio signals and applied normalization. A total of 3 data augmentation methods were applied as follows: frequency masking, time masking, and mixup. In addition, we designed Convolutional Neural Network (CNN) with spatial dropout, batch normalization, and separable 2D inspired by VGGnet [1]. In addition, we reduced the model size by using model quantization of float16 to the trained model. Experiments were conducted on the updated dataset provided by the Detection and Classification of Acoustic Events and Scenes (DCASE) 2020 challenge. We confirm that our model achieved a val_loss of 0.33 and an accuracy of 90.34% within the 132.50KB model size.