• Title/Summary/Keyword: 반올림

Search Result 37, Processing Time 0.023 seconds

Hole Filling Method for Natural Eye Gaze Correction (자연스러운 눈맞춤 영상을 위한 홀 채움 방법)

  • Ko, Eunsang;Ho, Yo-Sung
    • Proceedings of the Korean Society of Broadcast Engineers Conference
    • /
    • 2015.11a
    • /
    • pp.169-172
    • /
    • 2015
  • 영상회의 장치에서 눈맞춤 영상을 생성하기 위해 어파인 변환(affine transformation)을 이용하면 반올림 오차(round-off error) 때문에 홀이 발생한다. 이러한 홀을 채우려면 홀 영역을 가리키는 홀 채움 마스크가 필요하다. 홀 채움은 보통 홀 채움 마스크를 참조하여 홀이 아닌 이웃 화소값들을 기반으로 손상된 영상을 복원하는 작업이다. 따라서 홀 영역을 정확히 검출하고 적당한 개수의 이웃 화소값을 참조해야 자연스러운 홀 채움 영상을 생성할 수 있다. 한편, 눈맞춤 영상을 생성할 때 얼굴 특징점을 이용해 얼굴 변환 마스크를 만들고, 얼굴 변환 마스크에만 어파인 변환을 수행한다. 이 논문에서는 얼굴 특징점에도 어파인 변환을 수행하고 수정된 얼굴 변환 마스크를 획득하여 정확한 홀 채움 마스크를 구한다. 또한, 홀 채움 마스크에서 레이블링을 수행하여 큰 홀 영역을 제거한다. 마지막으로, 어파인 변환을 수행할 때 기존 영상의 좌표값을 이용하여 자연스러운 홀 채움 영상을 생성한다. 제안하는 방법으로 홀 채움을 수행한 결과, 연속적인 눈맞춤 동영상에서 이웃값들을 참조하여 홀 채움을 수행한 영상보다 자연스러움을 확인했다.

  • PDF

On Factorizing the Discrete Cosine Transform Matrix (DCT 행렬 분해에 관한 연구)

  • 최태영
    • The Journal of Korean Institute of Communications and Information Sciences
    • /
    • v.16 no.12
    • /
    • pp.1236-1248
    • /
    • 1991
  • A new fast algorithm for computing the discrete cosine transform(DCT) Is developed decomposing N-point DCT into an N /2-point DCT and two N /4 point transforms(transpose of an N /4-point DCT. TN/t'and)It has an important characteristic that in this method, the roundoff noise power for a fixed point arithmetic can be reduced significantly with respect to the wellknown fast algorithms of Lee and Chen. since most coefficients for multiplication are distributed at the nodes close to the output and far from the input in the signal flow graph In addition, it also shows three other versions of factorization of DCT matrix with the same number of operations but with the different distributions of multiplication coefficients.

  • PDF

The Media Coverage of Industrial Disaster in Korea A Case of Samsung Workers' leukemia (언론이 산업재해를 보도하는 방식에 관한 연구 삼성 백혈병 사태의 경우)

  • Pang, Huikyong;Won, Yongjin
    • Korean journal of communication and information
    • /
    • v.79
    • /
    • pp.40-69
    • /
    • 2016
  • This study examines how the Korean media has covered the leukemia case of Samsung workers. The news articles on the cases, published in the three terrestrial broadcasting stations, the five major newspapers, progressive Internet news media, conservative Internet news media, and business newspapers between 2007 and 2015, were retrieved and analyzed. Four characteristics of news reports were discovered as following: First, while conservative news media and business news media have exnominated the first phase of the case in which a civil organization Sharp claimed that Samsung workers had leukemia from their workplace; they have nominated the second phase of the case in which Samsung attempted to solve the problem by compensating the victims for their disease. Second, the media generally have displayed more interests in the results than in the causes of the case. Third, the conservative press and business newspapers have reported the case in favor of Samsung rather than laborer victims. Fourth, the conservative press have deployed market-friendly discourses, rather than problematizing the state and civil society. Through an abduktive analysis, this study argues that except the Hangyore and Kyunghyang Shinmun, most of the Korean media, assuming pro-business attitude, do not question or clarify where the responsibility of the industrial disaster lies by displaying their interests in the phase of solution, not in the phase of cause.

  • PDF

Legal and Institutional Outcomes from the 10-year Struggle against Occupational Diseases of Semiconductor workers (반도체 직업병 10년 투쟁의 법·제도적 성과와 과제)

  • Lim, Jawoon
    • Journal of Science and Technology Studies
    • /
    • v.18 no.1
    • /
    • pp.5-62
    • /
    • 2018
  • Over the last 10 years, the fight against occupational diseases of semiconductor workers led by SHARPS(the Supporters for the Health And Rights of People in the Semiconductor industry, NGO) has accomplished considerable achievements, especially in the legal and institutional aspects. First, the court and the government accepted the claims that 24 injured workers respectively filed, recognizing their 10 types of diseases as occupational illness. The court not only expanded the list of work places and diseases that it recognized, but also presented more progressive logic of recognition. The most remarkable achievement among them is the case ruled by the Supreme court in July, 2017. In terms of 'worker's right to know', which is the most important factor in preventing occupational diseases, there have been significant legislative bills, court rulings and government guidelines. The revised bill of the Industrial Safety and Health Act to strengthen workers' rights to know and to introduce the pre-review system on trade secret is currently under review by the National Assembly. The court recently ruled that the government should disclose its inspection results on safety and health management at semiconductor factories. The ministry of labor has drawn up internal guidelines to more actively open its safety and health data to public. This study looks over recent developments in such rulings, bills and guidelines and then, analyzes their implications, laying the groundwork for future actions for worker health in the electronic industry.

An Implementation of Pan-So-Ri Classification Program Using Naive Bayesian Classifier (나이브 베이지안 분류기를 이용한 판소리 분류 프로그램 구현)

  • Kim, Won-Jong;Lee, Kang-Bok;Kim, Myung-Gwan
    • The Journal of the Institute of Internet, Broadcasting and Communication
    • /
    • v.11 no.3
    • /
    • pp.153-159
    • /
    • 2011
  • Pan-So-Ri singing a story as song is one of Korea traditional musics. it divide into two sect(east-sect, west-sect), and it is hard to classify two sect without knowledge about Pan-So-Ri. In this paper, we have propose a Pan-So-Ri classification program using PCD(Pitch Class Distribution) and Naive Bayesian Classifier. Attribute value of classifier is each appearance frequency of pitch. Experiment is conducted two time with different rounding off location of probability value. Better one show correct classification with east-sect 80%, west-sect 97%, and total accuracy of 88%. this result is used our program.

A Design of Radix-2 SRT Floating-Point Divider Unit using ]Redundant Binary Number System (Redundant Binary 수치계를 이용한 radix-2 SRT부동 소수점 제산기 유닛 설계)

  • 이종남;신경욱
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.5 no.3
    • /
    • pp.517-524
    • /
    • 2001
  • This paper describes a design of radix-2 SRT divider unit, which supports IEEE-754 floating-point standard, using redundant binary number system (RBNS). With the RBNS, the partial quotient decision logic can operate about 20-% faster, as well as can be implemented with a simple hardware when compared to the conventional methods based on two's complement arithmetic. By using a new redundant binary adder proposed in this paper, the mantissa divider is efficiently implemented, thus resulting in about 20% smaller area than other works. The divider unit supports double precision format, five exceptions and four rounding modes. It was verified with Verilog HDL and Verilog-XL.

  • PDF

Deal price model in Deal-or-No-Deal game (딜또는노딜 게임에서 딜금액 결정 모형)

  • Song, Seolhee;Ahn, Soohan
    • Journal of the Korean Data and Information Science Society
    • /
    • v.25 no.4
    • /
    • pp.697-703
    • /
    • 2014
  • Deal-or-No-Deal game is a famous TV show program of NBC, USA, which is composed of 10 stages at most. At each stage from the first and the ninth, a banker suggests a deal price to participants. In this paper, we intend to reveal the banker's deal price model using a constrained linear model and quadratic program. As results, we provide a linear model in relation to the deal price at each stage and then show using simulation data that the deal price is equal to the nearest integer of the value to be obtained by the provided linear model.

Efficient Design of BCD-EXCESS 3 Code Converter Using Quantum-Dot Cellular Automata (QCA를 이용한 효율적인 BCD-3초과 코드 변환기 설계)

  • You, Young-Won;Jeon, Jun-Cheol
    • Journal of Advanced Navigation Technology
    • /
    • v.17 no.6
    • /
    • pp.700-704
    • /
    • 2013
  • Quantum-dot cellular automata(QCA) is a new technology and it is an one of the alternative high performance over existing complementary metal-oxide semi-conductor(CMOS). QCA is nanoscale device and ultra-low power consumption compared with transistor-based technologies, and various circuits using QCA technology have been proposed. Binary-coded decimal(BCD), which represents decimal digits in binary, is mainly used in electronic circuits and Microprocessor, and it is comfortable in conversion operation but many data loss. In this paper, we present an BCD-EXCESS 3 Code converter which can be efficiently used for subtraction and half adjust. The proposed scheme has efficiently designed considering space and time complexities and minimization of noise, and it has been simulated and confirmed.

Sensorless speed control of permanent magnet synchronous motor using square-root extended kalman filter (제곱근 확장 칼만 필터에 의한 영구자석 동기전동기의 센서리스 속도제어)

  • Moon, Cheol;Kwon, Young-Ahn
    • Journal of Advanced Marine Engineering and Technology
    • /
    • v.40 no.3
    • /
    • pp.217-222
    • /
    • 2016
  • This study investigates the design, analysis, and implementation of the square-root extended Kalman filter by using an algorithm derived by combining the Potter or Carlson algorithm with the modified Gram-Schmidt algorithm, for sensorless speed control of a permanent-magnet synchronous motor. The sensitivity of the Kalman filter to round-off errors is a well-known problem. A possible way to address this limitation is by combining the square-root concept and Kalman filter that can improve the numerical performance and solve instability-related problems such as divergence. This paper presents the design and analysis of the implementation of such a square-root extended Kalman filter. To demonstrate the performance of the proposed filter, experimental results under several operating conditions, such as high and low speeds, reversal rotation, detuned parameters and load test, have been analyzed. Further, code sizes and operation times have been compared. Experimental results establish the performance of the proposed square-root extended Kalman filter-based estimation technique for sensorless speed control of a permanent-magnet synchronous motor.

Design of High Performance Dual Channel Pipelined Interpolators for H.264 Decoder (이중 채널 파이프라인 구조의 H.264용 고성능 보간 연산기 설계)

  • Lee, Chan-Ho
    • Journal of IKEEE
    • /
    • v.13 no.4
    • /
    • pp.110-115
    • /
    • 2009
  • The motion compensation is the most time-consuming and complex unit in the H.264 decoder. The performance of the motion compensation is determined by the calculation of pixel interpolation. The quarter-pixel interpolation is achieved using 6-tap horizontal or vertical FIR filters for luminance data and bilinear FIR filters for chroma data. We propose the architecture for interpolation of luminance and chroma data in H.264 decoders. It is composed of dual-channel pipelined processing elements and can interpolate integer-, half- and quarter-pixel data. The number of the processing cycles is different depending on the position. The processing elements are composed of adders and shifters to reduce the complexity while the accuracy of the pixel data are maintained. We design interpolators for luminance and chroma data using Verilog-HDL and verify the function and performance by implementing using an FPGA.

  • PDF