• The KIPS Transactions:PartD
• /
• v.14D no.7
• /
• pp.801-808
• /
• 2007

Video coding standards such as MPEG-2, MPEG-4, H.263, and H.264 transmit a compressed video data using wired/wireless communication line with limited bandwidth. Because highly compressed bit-streams is likely to fragile to error from channel noise, video is damaged by error. There have been many research works on error concealment techniques, which recover transmission errors at decoder side [1, 2]. We designed an error concealment technique for lost motion vectors of H.264 video coding. In this paper, we propose a Kalman filter based motion vector recovery scheme, and experimented with standard video sequences. The experimental results show that our scheme restores original motion vector with more precision of 0.91 - 1.12 on average over conventional H.264 decoding with no error recovery.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.12
• /
• pp.2844-2851
• /
• 2015

How to recover from the multiple 2D images into 3D object has been widely studied in the field of computer vision. In order to improve the accuracy of the recovered 3D shape, it is more important that noise must be minimized and the number of image frames must be guaranteed. However, potential noise is implied when tracking feature points. And the number of image frames which is consisted of an observation matrix usually decrease because of tracking failure, occlusions, or low image resolution, and so on. Therefore, it is obviously essential that the number of image frames must be secured by recovering the missing feature points under noise. Thus, we propose the analytic approach which can control directly the error distance and orientation of missing feature point by the geometrical properties under noise distribution. The superiority of proposed method is demonstrated through experimental results for synthetic and real object.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.2
• /
• pp.376-381
• /
• 2009

The image data compression, recognition, restoration, etc. are parts of the digital image processing technology. In the process by various devices, noises would be made. Because the noise could damage the image, we use the image filter to preserve the original image from the noise. The image filter used in digital image process basically has a two-dimensional structure. There an two methods of creating the filter - One is reiterating one dimension and the other is using an indivisible two-dimension image filter. The image filter is being widely used along with one-dimension filter according to each noise, and various median filters are being used to remove the impulse noise. In this paper, I suggested a powerful modified median filter, and compared with conventional filters for objective verification.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.10
• /
• pp.964-971
• /
• 1991

In DPCM coder which has flat notse distributions, reconstruction noises might be shaped or transferred to hign frequency portion using the relevant noise reduction loop under the constraints of image compression. Noise shaping filter directly in loop can be used, but we shows the complete characteristics of signal blurred DPCM coder in view of data compression and noise processing. Equi weighting in bluming procedure of prefilter is introduced and the blurred signal is restored by debiurning postfilter. Noise shaping tegion on two dimensional frequency domain depends upon the size of masking filter. In spite of its noise increments by the signal blurring, the algonthm is effective for the visually relevant coding which has low pass freqency properties.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.477-481
• /
• 2009

In this paper the process of transmitting images signal restore to image corrupted by impulse noise proposed switching mean filter. these filter is differential size using the two state noise detection distinguishes noise pixel and noise free pixel. Follow the detected impulse noise density in the image remove the impulse noise using switching mean filter these substituted pixel in order to non-recursive and recursive form from control process of the next pixel comes to be used with neighbor pixel process. Through the simulation, we compared with the existing methods and capabilties.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.05a
• /
• pp.373-376
• /
• 2009

영상처리 기술은 인간의 시각에 기반을 둔 영상정보와 관련된 분야에서 중요한 기반 기술로써 현재 여러 분야에서 연구가 활발하게 진행 중이다. 여러 응용 분야에서 사용되는 영상처리의 세부 기술범위는 영상 변환, 영상 개선, 영상 복원, 영상 압축등과 같이 다양하며, 이런 영상처리 기술의 중요한 연구 목표 중의 하나는 정확한 정보 추출을 위한 영상정보의 개선에 있다. 영상정보의 개선은 영상의 해석과 인식을 위한 기본적인 과제이며, 영상에서 나타날 수 있는 잡음을 제거하는 영상처리 기술이 영상정보 개선의 한 분야라고 할 수 있다. 영상정보 개선을 위한 기존의 필터링 알고리즘은 잡음제거율이 높은 만큼 경계선의 보존이 어렵다는 단점이 있으며, 이를 보완하기 위해 다른 영상처리 알고리즘을 함께 응용하여 처리함으로써 처리시간이 증가되고 원 영상의 중요한 정보를 훼손할 가능성이 존재한다. 따라서 본 논문에서는 기존의 필터링 알고리즘의 문제점을 개선하는 동시에 잡음 제거율을 높일 수 있는 Fuzzy Mask Filter 알고리즘을 제안한다. Fuzzy Mask Filter 알고리즘은 마스크에서 얻은 정보를 Fuzzy Logic에 적용하여 임계값을 구하며, 구해진 임계값을 기준으로 출력영상의 화소값을 결정하는 알고리즘이다. 본 논문에서 제안한 알고리즘의 효율성을 검증하기 위해 Impulse 잡음과 Salt pepper 잡음을 임의로 생성하여 기존의 알고리즘과 비교한 결과, 제안된 방법이 잡음 영상에 존재하는 픽셀 정보를 훼손하지 않고 잡음을 효과적으로 제거한 것을 확인할 수 있었다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.795-797
• /
• 2014

Image treatment has grown in its necessity in many fields due to the development of display technology used in smart phones and tablet computers. In digital imaging technology, noise is created by many causes during the process of acquiring, transmitting and treating image data. Therefore, this paper suggests a median filter that is more competent in removing noise by taking into account the direction when restoring noisy images that have been damaged by impulse noise. In order to verify the noise removal characteristics, PSNR(peak signal to noise ratio) was used for comparison against existing methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.3
• /
• pp.254-260
• /
• 2019

Recently, the importance of signal processing has become gradually significant, as the frequency of video media increases in various fields. However, numerous kinds of noise generated in the transmission and reception processes can possibly affect the signal information, and the noise removal is for that reason essential as a preprocessing step. In this paper, we propose an algorithm to remove the mixed noise which is composed of impulse noise and AWGN. This algorithm is used for image restoration by noise judgment for efficient noise removal in a complex noise environment, and the noise is removed by considering spatial characteristics and pixel variations. Simulation results show that unlike existing methods, the algorithm has excellent noise cancellation characteristics by minimizing both noise effects and consequently eliminating the mixed noise; for objective judgment, we compared and analyzed the data using PSNR and profile.

• Journal of Korea Multimedia Society
• /
• v.3 no.4
• /
• pp.365-371
• /
• 2000

To restore image degraded by out-of-focus blur and additive noise, an iterative restoration is used. Acceleration parameter is usually applied equally to all over the image without considering the local characteristics of degraded images. As a result, the conventional methods are not effective in restoring severely degraded edge region and shows slow convergence rate. To solve this problem we propose an adaptive iterative restoration according to local degradation, in which the acceleration parameter has low value in flat region that is less degraded and high value in edge region that is more degraded. Through experiments, we verified that the proposed method showed better results with fast convergence rate, showed Visually better image in edge region and lower MSE than the conventional methods.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.472-474
• /
• 2001

본 논문은 wavelet 기반의 watershed를 이용한 효율적인 영상 분할을 기법을 제안한다. 영상 분할을 위해 입력 영상을 wavelet transform을 사용하여 low-resolution 영상을 생성한 후 watershed 알고리즘을 이용해 분할하고, 이를 Inverse wavelet transform함으로써 원 영상으로 복원한다. 복원된 영상을 의미 있는 영역들로 분할하기 위해 wavelet 특징값의 유사성을 두 인접한 영역에 비교하여 병합한다. 실험 결과 제안한 방법은 영상의 잡음에 대한 강인함과 영상의 과분할 문제를 해결할 수 있다.