• 전기전자학회논문지
• /
• 제22권4호
• /
• pp.1242-1245
• /
• 2018

본 논문에서, Gaussian noise를 제거할 때 발생하는 over blurring 현상을 감소시키는 network를 구현하였다. 기존 filtering 방식은 원 영상을 blurring하여 noise를 제거함으로써, edge나 corner 같은 high frequency 성분도 함께 지워지는 것을 확인할 수 있다. CNN (Convolutional Neural Network)기반 denoiser의 경우도 사소한 edge, keypoint를 noise로 인식하여 이러한 정보를 잃게 된다. 우리는 CNN을 기반으로 denoising된 high frequency 성분만을 획득하여 기존 denoiser에 추가함으로써 denoising 성능을 유지하면서 over blurring을 완화하는 network 제안한다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅳ
• /
• pp.1707-1710
• /
• 2003

In this paper, we propose a method of decision on blurring for business card images using block classification. In the proposed method, an input image is partitioned into 8${\times}$8 blocks and each block is classified into character block or background block using a block energy calculated in DCT domain. Whether the input image is blurring or non-blurring is determined using a ratio of low frequency energy and high frequency energy in DCT domain. Experimental results show that the proposed block classification classifies block well and the proposed decision on blurring decides well for various business card images.

• Journal of the Optical Society of Korea
• /
• 제18권5호
• /
• pp.472-476
• /
• 2014

The Joint Transform Correlator (JTC) has been the most suitable technique for real time optical pattern recognition and target tracking applications. This paper proposes a new application of the JTC system for an analysis of the blurring effect of the optical images caused by a defocused lens. We present the relation between the correlation peak, optical transfer function (OTF), and the amount of blurring caused by focusing error. Moreover, we show a possibility of calibrating the blurred image by simply measuring the correlation peak.

• 전기전자학회논문지
• /
• 제21권1호
• /
• pp.46-65
• /
• 2017

This paper proposes a fast patch-based de-blurring algorithm including kernel estimation based on the angle between the edge and the blur direction. For de-blurring, image patches from the most informative edges in the blurry image are used to estimate a kernel with low computational cost. Moreover, the kernels of each patch are estimated based on the correlation between the edge direction and the blur direction. This makes the final kernel more reliable and creates an accurate latent image from the blurry image. The combination of directionally oriented kernel estimation and patch-based de-blurring is faster and more accurate than existing state-of-the art methods. Experimental results using various test images show that the proposed method achieves its objectives: speed and accuracy.

• 대한디지털의료영상학회논문지
• /
• 제15권2호
• /
• pp.55-61
• /
• 2013

The purpose of this study is to evaluate image blurring according to variation of the ETL and propose the clinically appropriate ETL range. SIEMENS MAGNETOM Skyra 3.0T and 20 channel head coil were used for the study. MRI phantom was kept the lines horizontally to three direction(X,Y,Z) of the coil and T1, T2 weighted images that used the fast spin echo technique acquired. The ETL with increase of 10 was applied from 10 to 80. In addition, the ETL with increase of 1 was applied in the interval statistically significant differences occurred. And T1, T2 weighted images that used the conventional spin echo technique acquired to compare image blurring of the images that used the fast spin echo technique. The slope of lattice in the images was measured using Image J 1.47v program to evaluate image blurring. And image blurring was determined by the degree of the slope. The statistical significance of both techniques was evaluated by the Kruskal-Wallis test of the SPSS 17.0v. And the correlation of the ETL and image blurring was evaluated quantitatively by regression analysis. The slope of the T1, T2 weighted images that used fast spin echo technique decreased as contrasted with conventional spin echo technique. In the result of the Kruskal-Wallis test, the T1, T2 weighted images that used fast spin echo technique made a significant difference with conventional spin echo technique. Particularly, in the Tomhane' T2 test, the T1, T2 weighted images made a significant difference from ETL 22 and 31 respectively. In the result of the regression analysis, the R-squared of the T1, T2 weighted images are 0.762 and 0.793. It is difficult to apply the long ETL in the T1 weighted image caused by the short TR and multi-slices study. Therefore, clinical impact according to variation of the ETL is very slight in the T1 weighted images. But the application of the proper ETL is demanded in T2 weighted images using the fast spin echo technique in order to prevent image blurring.

• 대한전자공학회논문지
• /
• 제20권6호
• /
• pp.75-80
• /
• 1983

스므싱 필터를 사용하여 spot noise를 갖는 협상을 스므싱할 때에, 잡음은 거의 제거되지만 영상의 blurring을 초래한다. 본 논문에서는 이러한 영상 blurring을 동시에 개선하기 위하여 메디안 필터를 이용한 한 알고리즘을 제안하였다. 본 논문에서 제시한 방법은 잡음협상과 그 잡음영상을 메디안 필터링한 결과와의 차를 이용하였으며, 잡음제거와 동시에 blurring의 감소를 효과적으로 이룩할 수 있음을 보여 주었다. 본 논문에서는 또한 제안한 알고리즘을 여러가지 기존 스므싱 방법과 비교하여 그 특징을 제시하였다.

• 한국통신학회논문지
• /
• 제35권4C호
• /
• pp.408-414
• /
• 2010

CCTV이미지에서 교통정보를 수집하려고 할 때 관측자가 카메라를 움직이면 설치된 검지영역이 손실되고 복원하려고 해도 기계적 오차로인해 어려움을 겪게 된다. 그래서 어안이나 거울을 이용하면 카메라를 움직이지 않아도 된다. 하지만 이러한 상황하에서 가장 큰 문제점은 영상의 왜곡이다. 본 논문에서는 이러한 왜곡을 극복하기 위해 비선형 확산 방정식을 이용한 분할과 이 영역에디포메이션을 적용하여 디블러링 한 후 차량정보 수집을 시도한 결과 이전 결과보다 5% 향상된 결과를 얻었다.

• Journal of Biosystems Engineering
• /
• 제17권4호
• /
• pp.363-369
• /
• 1992

The aim of this study was to examine the blurring effects on performance of the experimental fruit grader in grading Fuji apples by size and coloration of the whole surface of individual apples. The grader consisted of a black/white image prcessing system, one camera, and utilized the algorithm developed for high speed sorting in the previous study. The results are summarized as follows : 1. With the algorithm developed in the previous study, it took 0.27~0.33 second in analyzing the size and coloration of an apple, and relative errors were within 3% for size and 1.3% for coloration. 2. The effect of blurring increased linearly with the conveying speed of apple and showed more significant effect on detection of coloration than on determining of size. 3. Considering the blurring effect, capacity of the experimental fruit grader was estimated to 7,500 apples per hour.

• 반도체디스플레이기술학회지
• /
• 제22권1호
• /
• pp.39-42
• /
• 2023

Recently, with the explosive increase of video streaming services, real-time live broadcasting has also increased, which leads to an infringement problem for user privacy. So, to solve such problems, we proposed the real image blurring system using dlib face-recognition library. 68 face landmarks are extracted and convert into 128 vector values. After that the proposed system tries to compare this value with the image in the database, and if it is over 0.45, it is considered as different person and image blurring processing is performed. With the proposed system, it is possible to solve the problem of user privacy infringement, and also to be utilized to detect the specific person. Through experimental results, the proposed system has an accuracy of more than 90% in terms of face recognition.

• 한국생산제조학회지
• /
• 제19권4호
• /
• pp.562-569
• /
• 2010

This paper analyzes the problems that occurred in the magnification process for a fine input image and investigates a method to improve the blurring of magnified image. This paper applies a curve interpolation algorithm in CAD/CAM for the same test images with the existing image algorithm in order to improve the blurring of magnified image. As a result, the nearest neighbor interpolation, which is the most frequently applied algorithm for the existing image interpolation algorithm, shows that the identification of a magnified image is not possible. Therefore, this study examines an interpolation of gray-level data by applying a low-pass spatial filter and verifies that a bilinear interpolation presents a lack of property that accentuates the boundary of the image where the image is largely changed. The periodic B-spline interpolation algorithm used for curve interpolation in CAD/CAM can remove the blurring but shows a problem of obscuration, and the Ferguson' curve interpolation algorithm shows a more sharpened image than that of the periodic B-spline algorithm. For the future study, hereafter, this study will develop an interpolation algorithm that has an excellent improvement for the boundary of the image and continuous and flexible property by using the NURBS, Ferguson' complex surface, and Bezier surface used in CAD/CAM engineering based on the results of this study.