• 전자공학회논문지 IE
• /
• v.49 no.1
• /
• pp.24-29
• /
• 2012

In this paper, we propose a progressive filter for impulse noise reduction. The proposed method uses non-linear filter and linear filter progressively. Non-linear filter reduces abrupt noise pattern. Also, linear filter adjusts filtering direction according to an edge in the image which is filtered by non-linear filter. Thus, our proposed method not only preserves edge, but also reduces noise in uniform region. Experimental results show that our proposed method has better quality than those by existing non-linear and linear progressive filtering methods.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.43 no.1 s.307
• /
• pp.39-44
• /
• 2006

Pattern noise in the Delta-Sigma modulator is a well Known phenomenon that intrigued many circuit designers. These noise appear as the modulator output falls into a cyclic mode of operation. This paper addresses the dependence of these tone signal upon the system topologies. Among the four well known single-stage DSM topologies, namely Cascade of Integrators with Feedback Form(CIFB), Cascade of Integrators with Feedforward Form(CIFF), Cascade of Resonators with Feedback Form(CRFB), and Cascade of Resonators with Feedforward Form(CRFF), resonator type DSMs turn out to be more susceptible to the pattern noise than the integrator type. Noise transfer functions of the investigated topologies are also presented.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.342-344
• /
• 1999

신경트리(evolutionary neural trees)는 트리 구조의 신경망 모델로서 진화 알고리즘으로 학습하기에 적합한 구조이다. 본 연구에서는 진화 신경트리를 시계열 예측에 적용하였다. 시계열 데이터는 대개 잡음이 포함되어 있으며 동역학적인 특성을 지닌다. 본 논문에서는 견고한 예측 결과를 획득하기 위해 한 개의 신경트리가 아닌 여러개의 신경트리를 결합하여 예측 모델을 구성하는 committee machine을 소개한다. 출력 패턴가에 correlation이 최소가 되도록 상이한 신경트리를 선택하여 결합함으로써 모델 결합 효과를 최대화하는 방법을 사용하였다. 인공적인 잡음을 포함한 시계열 예측 문제와 실세계 데이터에 대한 실험에서 예측에 대한 정확도가 단일 모델을 사용한 경우 보다 향상되었다.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 1983.10a
• /
• pp.102-104
• /
• 1983

This paper analyzes the patterns of Toneburst Waveform that generated Volumetarily in Broadband, as various Parameters, and applicate in the case that it is mixed Random Noise. As a result, it prooves that auto correlation function is Optimal parameter in analysis of Tone Burst wave form but reference signal.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.8
• /
• pp.11-19
• /
• 2023

In this paper, it is confirmed that the metrics used to evaluate image quality can be applied to low-light images. Due to the nature of low-illumination images, factors related to light create various noise patterns, and the smaller the amount of light, the more severe the noise. Therefore, in situations where it is difficult to obtain a clean image without noise, the quality of a low-illuminance image from which noise has been removed is often judged by the human eye. In this paper, noise in low-illuminance images for which ground truth cannot be obtained is removed using Noise2Noise, and spatial resolution and radial resolution are evaluated using ISO 12233 charts and colorchecker as metrics such as MTF and SNR. It can be shown that the quality of the low-illuminance image, which has been evaluated mainly for qualitative evaluation, can also be evaluated quantitatively.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.10
• /
• pp.72-82
• /
• 1999

The error diffusion algorithm is excellent for reproducing continuous gray-scale images to bianry images and also has good edge characteristics. However, it is well known that artifacts with objectionable patterns can occur in the halftoned images. On the other hand, a halftone algorithm using blue noise mask has been proposed. where the halftoning is achieved by a pixelwise comparison of gray-scale image with an array, the blue noise mask. It doesn't have pattern artifacts, but the halftoned image looks unclear because the quantization errors are not feedbacked compared to the error diffusion. In this paper, edge enhanced error diffusion which dithers the threshold with the blue noise mask is proposed. We show that the proposed algorithm can produce unstructured and edge enhanced halftone images. This algorithm is analyzed by the concept of an equivalent input image. The performace of the proposed algorithm is compared with that of the conventional halftoning by measuring the radially averaged power spectrum and edge correlation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.2
• /
• pp.214-220
• /
• 2021

In modern society, various digital communication equipments are being used due to the influence of the 4th industrial revolution, and accordingly, interest in removing noise generated in the data transmission process is increasing. In this paper, we propose a filtering algorithm to remove AWGN generated during digital image transmission. The proposed algorithm removes noise based on mask matching to preserve information such as the boundary of an image, and uses pixel values with similar patterns according to the pattern of the input pixel value and the surrounding pixels for output calculation. To evaluate the proposed algorithm, we simulated with existing AWGN removal algorithms, and analyzed using enlarged image and PSNR comparison. The proposed algorithm has superior AWGN removal performance compared to the existing method, and is particularly effective in images with strong noise intensity of AWGN.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.459-462
• /
• 2001

본 논문에서는 하이브리드 패턴벡터를 이용하여 자동차의 고유 마크와 차량 번호를 실시간으로 인식하는 알고리즘을 제안하였다. 제안한 알고리즘에서는 차량 입력 영상에서 차량의 마크와 번호판의 수평 및 수직 명암값 빈도수 변화를 이용해 마크와 번호판 영역을 추출한다. 또한, 추출된 알고리즘으로부터 수평 수직 패턴을 적용해 자동차의 마크를 인식하고 하이브리드 패턴벡터를 이용하여 번호판의 문자 및 숫자를 인식하도록 하였다. 제안한 자동차 마크 및 번호판 추출 과정에서는 마크와 번호판 영역의 문자와 배경이 뚜렷하게 구별되는 상대적인 크기의 특성과 수평 및 수직 빈도수와 패턴 벡터를 사용하여 마크 및 번호판 영역을 추출, 인식하도록하였다. 제안한 방법들을 적용한 결과, 차량 번호판의 크기에 관계없이 잡음에 영향을 받지 않고 차량의 종류와 번호를 실시간으로 처리할 수 있으며 차량번호판 추출 및 인식뿐 아니라 차량의 마크 추출 가능성을 제시하였다.

• The KIPS Transactions:PartB
• /
• v.13B no.7 s.110
• /
• pp.689-698
• /
• 2006

Pattern recognition is the study of how machines can observe the environment, learn to distinguish patterns of interest from their background, and make sound and reasonable decisions about the categories of the patterns. Color image consists of various color patterns. And most pattern recognition methods use the information of color which has been trained and extract the feature of the color. This thesis extracts adaptively specific color feature from images with several limited colors. Because the number of the color patterns is limited, the distribution of the color in the image is similar. But, when there are some noises and distortions in the image, its distribution can be various. Therefore we cannot extract specific color regions in the standard image that is well expressed in special color patterns to extract, and special color regions of the image to test. We suggest new method to reduce the error of recognition by extracting the specific color feature adaptively for images with the low distortion, and six test images with some degree of noises and distortion. We consequently found that proposed method shouws more accurate results than those of statistical pattern recognition.

• The KIPS Transactions:PartB
• /
• v.9B no.3
• /
• pp.369-374
• /
• 2002

In this paper, we propose the react-time car license plate recognition algorithm using intensity variation and geometric pattern vector. Generally, difference of car license plate region between character and background is more noticeable than other regions. And also, car license plate region usually shows high density values as well as constant intensity variations. Based on these characteristics, we first extract car license plate region using intensity variations. Secondly, lightness compensation process is performed on the considerably dark and brightness input images to acquire constant extraction efficiency. In the proposed recognition step, we first pre-process noise reduction and thinning steps. And also, we use geometric pattern vector to extract features which independent on the size, translation, and rotation of input values. In the experimental results, the proposed method shows better computation times than conventional circular pattern vector and better extraction results regardless of irregular environment lighting conditions as well as noise, size, and location of plate.