• 제목/요약/키워드: post processing

검색결과 1,753건 처리시간 0.027초

An Echo Canceller Robust to Noise and Residual Echo

  • Kim, Hyun-Tae;Park, Jang-Sik
    • Journal of information and communication convergence engineering
    • /
    • 제8권6호
    • /
    • pp.640-644
    • /
    • 2010
  • When we talk with hands-free in a car or noisy lobby, the performance of the echo canceller degrade because background noise added to echo caused by the distance from mouth to microphone is relatively long. It gives a reason for necessity of noise-robust and high convergence speed adaptive algorithm. And if acoustic echo canceller operated not perfectly, residual signal going through the echo canceller to far-end speaker remains residual echo, which degrade quality of talk. To solve this problem, post-processing needed to remove residual echo ones more. In this paper, we propose a new acoustic echo canceller, which has noise robust and high convergence speed, linked with linear predictor as a post-processor. By computer simulation, it is confirmed that the proposed algorithm shows better performance from acoustic interference cancellation (AIC) viewpoint.

XML 기반의 지능형 무선인터넷 문서 변환기 (XML-based Intelligent Reformatter for Mobile Access Documents)

  • 홍순재;조성진;김호윤;함준호;이재원
    • 한국정보처리학회:학술대회논문집
    • /
    • 한국정보처리학회 2000년도 추계학술발표논문집 (하)
    • /
    • pp.1227-1230
    • /
    • 2000
  • 본 연구는 서로 다른 형식(format)으로 작성된 무선인터넷 문서들간의 상호 변환(reformatting)을 다루고 있다. 무선인터넷에 접속하는 단말기는 보통 하나의 형식만을 지원하는데 비해서, 무선인터넷 문서를 작성하는데 필요한 마크업 언어(markup language)로는 WML, HDML, mHTML 등 다양한 형식들이 존재하고 있다. 이에 본 연구에서 지식 기반 시스템(knowledge based system)과 XML(extensible Markup Language)을 이용하는 문서 형식 변환 기술을 제공하여, 무선인터넷 컨텐츠 구축시 특정 형식에 구애받지 않도록 하였다.

  • PDF

모바일 우편 서비스 모형 (Mobile Fostal Service Model)

  • 김동호;전성우;김진석
    • 한국정보처리학회:학술대회논문집
    • /
    • 한국정보처리학회 2004년도 추계학술발표논문집(상)
    • /
    • pp.665-668
    • /
    • 2004
  • 우편 서비스는 사람과 조직간을 이어주어 실물 배달과 의사 소통을 위한 중요 네트워크 중의 하나이다. 정보통신의 발달로 우편 서비스의 대표적인 형태인 서장이 전자메일로 대치되고 있으나, 전자상거래의 급속한 확장에 힘입어 소포 등과 같은 새로운 우편 서비스 영역의 중요성이 동반 확대되고 있는 실정이다. 2000년에 출범한 인터넷 기반의 우체국 서비스인 ePOST는 전통적인 우편서비스와 농수산물 중심의 쇼핑몰에서 전자고지납부 및 에스크로 서비스 등으로 확대되고 있다. 한편 우리나라의 휴대전화의 보급율은 2004년 현재 $75\%$에 이르고 있으며, 이제 ePOST도 휴대전화상에서 새로운 서비스를 제공하기 위한 준비가 필요하다. 따라서 이 논문에서는 대국민 서비스와 효과적인 비즈니스를 동시에 지원하기 위한 모바일 우편 서비스(mPOST) 방안을 제시하고자 한다.

  • PDF

HR-평가 문장 Multi-classification 및 Unlabeled data 를 활용한 Post-training 효과 분석 (HR-evaluation sentence multi-classification and Analysis post-training effect using unlabeled data)

  • 최철;임희석
    • 한국정보처리학회:학술대회논문집
    • /
    • 한국정보처리학회 2022년도 춘계학술발표대회
    • /
    • pp.424-427
    • /
    • 2022
  • 본 연구는 도메인 특성이 강한 HR 평가문장을 BERT PLM 모델을통해 4 가지 class 로 구분하는 문제를 다룬다. 다양한 PLM 모델 적용과 training data 수에 따른 모델 성능 비교를 통해 특정 도메인에 언어모델을 적용하기 위해서 필요한 기준을 확인하였다. 또한 Unlabeled 된 HR 분야 corpus 를 활용하여 BERT 모델을 post-training 한 HR-BERT 가 PLM 분석모델 정확도 향상에 미치는 결과를 탐구한다. 위와 같은 연구를 통해 HR 이 가지고 있는 가장 큰 text data 에 대한 활용 기반을 마련하고, 특수한 도메인 분야에 PLM 을 적용하기 위한 가이드를 제시하고자 한다

Post-processing of 3D Video Extension of H.264/AVC for a Quality Enhancement of Synthesized View Sequences

  • Bang, Gun;Hur, Namho;Lee, Seong-Whan
    • ETRI Journal
    • /
    • 제36권2호
    • /
    • pp.242-252
    • /
    • 2014
  • Since July of 2012, the 3D video extension of H.264/AVC has been under development to support the multi-view video plus depth format. In 3D video applications such as multi-view and free-view point applications, synthesized views are generated using coded texture video and coded depth video. Such synthesized views can be distorted by quantization noise and inaccuracy of 3D wrapping positions, thus it is important to improve their quality where possible. To achieve this, the relationship among the depth video, texture video, and synthesized view is investigated herein. Based on this investigation, an edge noise suppression filtering process to preserve the edges of the depth video and a method based on a total variation approach to maximum a posteriori probability estimates for reducing the quantization noise of the coded texture video. The experiment results show that the proposed methods improve the peak signal-to-noise ratio and visual quality of a synthesized view compared to a synthesized view without post processing methods.

인공신경망의 학습기능과 화성진행을 이용한 자동작곡 (Automatic Composition Using Training Capability of Artificial Neural Networks and Chord Progression)

  • 오진우;송정현;김경환;정성훈
    • 한국멀티미디어학회논문지
    • /
    • 제18권11호
    • /
    • pp.1358-1366
    • /
    • 2015
  • This paper proposes an automatic composition method using the training capability of artificial neural networks and chord progression rules that are widely used by human composers. After training a given song, the new melody is generated by the trained artificial neural networks through applying a different initial melody to the neural networks. The generated melody should be modified to fit the rhythm and chord progression rules for generating natural melody. In order to achieve this object, we devised a post-processing method such as chord candidate generation, chord progression, and melody correction. From some tests we could find that the melody after the post-processing was very improved from the melody generated by artificial neural networks. This enables our composition system to generate a melody which is similar to those generated by human composers.

Fine Blanking의 가장자리 Bridge 최소화 방법에 관한 연구 (A Study on Edge Bridge Minimization of Fine Blanking Process)

  • 김기태
    • 한국기계가공학회지
    • /
    • 제12권4호
    • /
    • pp.108-113
    • /
    • 2013
  • Industrialization and modernization of the beginning of the IT industry is growing very fast. Since telecommunications industry was developed rapidly, technologies about miniaturization and high-precision of parts have been actively developed to lead information revolution. generally, the entire shear surface of the product applying fine blanking technology must be very precise. Fine blanking is used to save cost by avoiding post-processing of the product. When using press blanking, it spends a lot of money on the production by using many post-processing. Fine blanking typically used in 0.5~18 mm thick steel plate. Because a lot of post-processing cost can be used to process, except for fine blanking. In order to develop components "CHANCE CONTENTS" in the fine blanking process, the purpose of this study is to minimize the edge of the bridge, secured 95% of the material thickness of the shear surface using the 1.6 mm thickness of the material SPCC. Blanking process by introducing after changing thickness through forging process, due to change in vee-rring force and counter force, the experimental amount of depressions and flatness and the shear surface were analyzed.

자동공구교환장치를 이용한 융합 FDM 공정 및 장치개발에 관한 연구 (Development of Hybrid-FDM Process Using Automatic Tool Changer for Multi-Material Production and Post-Processing)

  • 최성민;샤오젠;박인백;이석희
    • 한국정밀공학회지
    • /
    • 제33권3호
    • /
    • pp.235-242
    • /
    • 2016
  • The purpose of this study is an attempt to improve the functionality of a conventional Fused Deposition Modeling (FDM) process using the Automatic Tool Changer (ATC) to perform multimaterial production and post-processing. Hybrid-FDM means a fusion of an Additive Manufacturing process and grinding process using the ATC system. In order to enhance the potentiality of production capacity for multi-material fabrication and surface roughness improvement, two extrusion tools and one grinding tool system are suggested. A pneumatic chuck is attached on a moving platform in the XY axes plane and an extrusion head and grinding head are placed in a docking station, allowing for a quick changeover with each other. Therefore, the manufacturing lead time can be reduced efficiently for the fabrication of a product.

모바일 기기를 위한 음성인식의 사용자 적응형 후처리 (User Adaptive Post-Processing in Speech Recognition for Mobile Devices)

  • 김영진;김은주;김명원
    • 한국정보과학회논문지:컴퓨팅의 실제 및 레터
    • /
    • 제13권5호
    • /
    • pp.338-342
    • /
    • 2007
  • 본 논문에서는 모바일 환경에서 고립단어 음성인식을 할 경우 화자종속 방법을 이용하여 성능을 높이는 사용자 적응형 후처리 방법을 제안한다. 이 방법은 인식기의 정확한 인식 결과를 위한 추가적인 처리들로 구성된다. 즉 인식기의 출력과 정확한 최종 결과들 간의 관계를 학습하여 이를 잘못된 인식기의 출력을 수정하는 데에 사용한다. 학습에는 패턴인식에 강인한 다층 퍼셉트론을 사용하며 학습 시간을 고려하여 모델을 세분화하고 동적으로 동작할 수 있도록 구현한다. 이 결과 인식기의 오류에 대해 41%를 수정하는 성과(오류 수정률: 41%)를 보였다.

Changes in Cathepsin Activity during Low-Temperature Storage and Sous Vide Processing of Beef Brisket

  • Kaur, Lovedeep;Hui, Seah Xin;Boland, Mike
    • 한국축산식품학회지
    • /
    • 제40권3호
    • /
    • pp.415-425
    • /
    • 2020
  • It is believed that two main proteolytic systems are involved in the tenderization of meat: the cathepsins and the calpains. Many researchers consider the calpain system to be the major contributor to meat tenderness during post-mortem storage. However, the role and activity of cathepsins during post-mortem storage or low temperature meat processing is unclear, particularly for the tough meat cuts like brisket. Thus, the study was designed to investigate the effects of cold (refrigerated and frozen) storage and sous vide processing on the activities of cathepsin B, H, and L in beef brisket. There were no significant changes in pH and cathepsin H activity throughout the 18 d of storage at both temperatures. However, an increase in cathepsin B activity was observed during the first 4 d at both storage temperatures, but subsequently the activity remained unchanged. Cathepsins B and L were found to be more heat stable at sous vide temperatures (50℃ for 24 h, 55℃ for 5 h and at 60℃ and 70℃ for 1 h) compared to cathepsin H. Cathepsin B+L activity was found to increase after sous vide cooking at 50℃ for 1 h but decreased to about 47% relative to the uncooked control after 24 h of cooking. These results suggest that cathepsins B and L may contribute to the improved meat tenderness usually seen in sous vide cooked brisket meat.