• 대한전자공학회논문지SP
• /
• 제41권4호
• /
• pp.39-44
• /
• 2004

움직임 보상을 고려한 3차원 웨이블릿 변환은 공간적, 시간적인 상관관계에 중복된 정보를 효과적으로 제거한다. 그러나 웨이블릿 방식으로 압축된 영상이라 하더라도 압축률이 높은 경우 고주파 서브밴드의 변환계수가 상당수 손실되어 압축복원 시에 링 현상과 같은 왜곡이 생긴다. 본 논문에서는 이러한 3차원 웨이블릿의 압축왜곡을 줄이기 위하여 적응적 반복 복원 기법을 사용하는 새로운 알고리듬을 제안하였다. 제안된 적응적 기법에서는 에지의 방향에 따라 서로 다른 고역통과필터를 정칙화 복원에 사용하였다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.316-320
• /
• 2009

In this paper, a reference-free perceptual quality metric is proposed for image assessment. It measures the amount of overall blockiness and blurring in the image. And edge-oriented artifacts, such as ringing, mosaic and staircase noise are also considered. In order to give a single quality score, the individual artifact scores are adaptively combined according to the difference between the edge-oriented artifacts and other artifacts. The quality score obtained by the proposed algorithm shows strong correlation with the MOS values by VQEG.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 1997년도 Proceedings International Workshop on New Video Media Technology
• /
• pp.107-112
• /
• 1997

This paper proposes the regularized constrained iterative image restoration algorithms which apply new space-adaptive methods to degraded image signals, and analyzes the convergence condition of the proposed algorithm. First, we introduce space-adaptive regularization operators which change according to edge characteristics of local images in order to effectively prevent the restored edges and boundaries from reblurring. And, pseudo projection operator is used to reduce the ringing artifact which results from extensive amplification of noise components in the restoration process. The analysed algorithm is stable convergent to the fixed point. According to the experimental results for various signal-to-noise ratios(SNR) and blur models, the proposed algorithms other methods and is robust to noise effects and edge reblurring by regularization especially.

• 한국멀티미디어학회논문지
• /
• 제25권1호
• /
• pp.29-38
• /
• 2022

Although image compression is one of the essential technologies to transmit image data on a variety of surveillance and mobile healthcare applications, it causes unnecessary compression artifacts such as blocking and ringing artifacts by the lossy compression in the limited network bandwidth. Recently, image restoration methods using convolutional neural network (CNN) show the significant improvement of image quality from the compressed images. In this paper, we propose Image Denoising Convolutional Neural Networks (IDCNN) to reduce the compression artifacts for the purpose of improving the performance of object classification. In order to evaluate the classification accuracy, we used the ImageNet test dataset consisting of 50,000 natural images and measured the classification performance in terms of Top-1 and Top-5 accuracy. Experimental results show that the proposed IDCNN can improve Top-1 and Top-5 accuracy as high as 2.46% and 2.42%, respectively.

• 대한의용생체공학회:의공학회지
• /
• 제33권2호
• /
• pp.72-77
• /
• 2012

Phase Contrast MR Angiography(PC MRA) is excellent MRA technique for measuring the velocity of vessels in the human body. PC MRA need to at least four images for angiogram reconstruction and it caused longer scan time. Therefore, we used keyhole imaging combined PC MRA to reduce the scan time. However, keyhole imaging can lead the erroneous effects as loss of phase information or frequency discontinuous. In this study, we applied the keyhole imaging combined 2D PC MRA for improving the temporal resolution and also measured the velocity to evaluate the accuracy of phase information. We used 0.32T MRI scanner(Magfinder II, Scimedix, Korea). Using the 2D PC MRA pulse sequence, the vascular images for a human brain targeted on the Superior Sagittal Sinus(SSS) were obtained. We applied tukey window function for keyhole images to minimize the ringing artifact and erroneous factors that are induced frequency discontinuous and phase information loss. We also applied zero-padded algorithm to peripheral missing k-space lines to compare keyhole imaging results and the artifact power(AP) value was measured on the complex difference images to validate the image quality. Consider as based on our results, heavy image distortions and artifacts were shown until using at least 50% keyhole factor. Using above the 50% keyhole factors are shown well reconstructed and matched for magnitude images and velocity information measurements. In conclusion, we confirmed the image quality and velocity information of keyhole technique combined 2D PC MRA. Especially, measured velocity information through the keyhole imaging combination was similar to the velocity information of full sampled k-space image despite of frequency discontinuous and phase information loss in the keyhole imaging reconstruction process. Consequently, the keyhole imaging combined 2D PC MRA will give some clinical usefulness and advantages as improving the temporal resolution and measuring the velocity information via selecting the appropriate keyhole factor at low tesla MRI system.