http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : 金聖珞 (검색결과 11 건)
- 무료
- 기관 내 무료
- 유료
-
金聖珞 관동대학교 1996 關大論文集 Vol.24 No.1
In this paper, two methods are proposed in order to get more accusation letters. The first proposal is that the difference of shading is to be clear after series of a processing on a display screen. Considering with the application pixel, decision will be made existing a pixel value is either to be letters or to be background value as comparing with both side. The second proposal is that the original computer screen treats a rationalization as one bundle of the process after series of the processing (a letter recognition). Also after considering with around values, an application pixel value is defined. Therefore, these methods were proved that both the reduction of a processing time and the performance of a picking wanted letters are superior. To show a superiority of the proposal method, the result was made comparing both of existing method and the proposed method in this paper.
-
Analog형 PLL을 사용한 Pull-In Range와 주파수 체배 회로에 관한 연구
李俊模,金聖珞 관동대학교 1984 關大論文集 Vol.12 No.2
In this paper, Lock range, Capture range, frequency stability of the VCO by source voltage variation of the frequency multiplier are experimented and frequency multiplier is composed by the PLL and the frequency divider. Because multiplication rate passes the reference output to the VCO through R and C, it is possible to multiply about ten times the maximum free running frequency of VCO, 500KHz. The capture range has no concern with the multiplication constant and it has about 70KHz to 80KHz bandwidth when the input voltage is 50mV/div but lock range is decreased as much the frequency divider rate as multiplication constant. Though the frequency variation of the VCO has no concern with source voltage, a amplitude varition happened.
-
金聖珞 관동대학교 1998 關大論文集 Vol.26 No.2
This dissertation proposes a method for pattern recognition using spectrum analyzer and fuzzy neural network. Contour sequences obtained from 2-D planar images represent the Euclidean distance between the centroid and all boundary pixels of the shape. and are related to the overall shape of images. The Fourier transform of contour sequence and spectrum analyzer are used as a means of feature selection and data reduction. The four dimensional spectral feature vectors are extracted by spectrum analyzer from the energy spectrum. These Spectral feature vectors are invariant to shape translation, rotation, and scale transformations. The fuzzy neural network which is combined with two fuzzy ART modules is trained and tested with these spectral feature vectors. The experiments including 6 aircrafts recognition process are presented to illustrate the high performance of this proposed method in the recognition problems of noisy planar shapes.
-
申美英,李建翊,金聖珞 관동대학교 2000 關大論文集 Vol.28 No.2
In this paper fingerprint thinning method and feature extraction is studied. 256×256 gray level fingerprint images is partitioned into the same size block. The same size blocks are converted into the binary image. The binary images are converted into binary thinned images using Hildith thinning algorithm. From these binary thinned images we extract the ending points and the bifuration points. which are the most useful critical feature points in the fingerprint images. using 3×3 MASK. From these extracted feature points we eliminated error feature points using vertical horizonal number value. The extracted feature points is used to classification and verification.
-
마이크로프로세서를 이용한 교차로에서의 교통신호등 제어에 관한 연구
金聖珞,林海鎭 관동대학교 1988 關大論文集 Vol.16 No.2
The traffic-light controller uaing microcomputer is suggested. The basic algorithm is developed in consideration of program size, execution times of critical routines and real-time interactions with peripheral devices. The suggested algorithm can be easily adapted to any specific microcomputer. The goal is to feed a maximum number of cars through the intersection in any time period and the maximum duration of a red light for each route represents the primary constraint.
-
김성낙,金興逸 관동대학교 1995 關大論文集 Vol.23 No.1
In this paper, the potential of neural networks for the recognition of the printed Korean characters is examined. We obtained 105 printed Korean characters image as an experimental object through scanner with 300 DPI(Dot per inch) resolution. The preprocessing for feature extraction of Korean character is the segmentation of individual character and noise elimination. To reduce the input character image data, black pixel components are extracted from each input images. As this type of feature extraction method do not need the thinning process. A multilayer perceptron with one hidden layer was lesrning with a EBP(Error back propagation) learning algorithm. In a experiment with the most frequently used 105 printed Korean characters, the suggesed 96.7% of character recognition rate.
-
Basic Data-Flow Languane를 위한 Basic Data-Flow Processor 설계
金聖珞,朴贊政 관동대학교 1989 關大論文集 Vol.17 No.1
Data Flow 형태로 나타난 프로그램의 고 병렬 실행을 이룰 수 있는 프로세서에 대하여 서술하며, FORTRAN과 같은 일반적인 형태의 언어에 관하여 실체적인 데이타 처리 프로세서를 구성하여 Data-flow 모델에 대한 기초를 제시한다.
-
-
한국어 음성 인식을 위한 알고리즘의 설계와 하드웨어 구성
李相範,金聖珞,南侍秉 단국대학교 1990 論文集 Vol.24 No.-
Recently Korean speech recognition has been studied. Several methods used to designed the speech recognition system among the LPC, PARCOR, zero crossing, formant and sonagram. This paper presents a method to implement the system for phonemes of Korean speech recognition through the analysis of voice spectrum between the phonemes in the frequency domain by using filter bank. Frequency filter bank extracts characteristic parameters from voice spectrums separated by each frequency band. The strong point of this method lies in simplicity of hardware implementation and capability of real time processing. This results proves that simples data compression and real time voice recognition are possible by exploiting voice spectrum of 15 channels.
-
李相範,金聖珞,安昌 단국대학교 1989 論文集 Vol.23 No.-
Nowadays, speed recognition has based on speaker, vocabulary and so on. Many data, however, were required for speech analysis and synthesis. Thus it is necessary for the development of an algorithm to solve the problem. Since Korean syllables are divided into a consonant and vowel, a speech recognition algorithm using phonemes is effective. In this paper, 256 data samples and 12 predictor coefficients are used for the recognition. An effective voiced/unvoiced decision and pitch detection is performed by cepstrum. Also speech parameters are extracted by linear prediction and formant frequency analysis. Using these parameters, a speech recognition algorithm has been developed and make an experiment in some Korean speech samples. This experiment is performed for a speaker, and using linear prediction analysis and formant frequency extraction the acquired results are 87.14% and 88.09% respectively.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
