• 한국멀티미디어학회논문지
• /
• 제23권3호
• /
• pp.412-420
• /
• 2020

This paper proposes a performance-improving face recognition system based on a super resolution method for low-resolution images. The conventional face recognition algorithm has a rapidly decreased accuracy rate due to small image resolution by a distance. To solve the previously mentioned problem, this paper generates a super resolution images based o deep learning method. The proposed method improved feature information from low-resolution images using a super resolution method and also applied face recognition using a feature extraction and an classifier. In experiments, the proposed method improves the face recognition rate when compared to conventional methods.

• 한국산업정보학회:학술대회논문집
• /
• 한국산업정보학회 1998년도 공동추계학술대회 경제위기 극복을 위한 정보기술의 효율적 활용
• /
• pp.479-488
• /
• 1998

This study investigates the effect of conflict resolution methods on the IS outsourcing performance. A survey of 207 IS oursourcing relationship of U.S. firms indicates that(1) constructive/proactive methods, such as joint problem solving, is positively related to the success of IS outsourcing relationships, and that (2) destructive/passive methods, such as persuasion, smoothing/avoiding, domination, and use of harsh words are negatively related to the success of IS outsourcing relationships. The findings suggest that the manner in which conflict is resolved can be productive or destructive for the IS outsourcing performance.

• 대한전자공학회논문지SP
• /
• 제46권6호
• /
• pp.68-77
• /
• 2009

본 논문에서는 저해상도 유추 보간법(interpolation algorithm derived from low resolution: ILR)을 이용한 공간적 디인터레이싱 기법을 제안한다. 일반적인 에지 기반 디인터레이싱 방법들은 화소 단위의 상관도를 이용하기 때문에 잡음과 밝기의 변화에 민감하다는 단점이 있다. 또 보간의 성능을 좌우하는 정확한 에지 방향을 판단함에 있어서 만족스럽지 못한 성능을 보인다. 이러한 한계를 극복하기 위해 본 논문에서는 먼저 보간 하고자 하는 픽셀 주위의 저해상도 블록의 특성을 먼저 탐구하고 이를 고해상도 블록에 적용함으로써 missing pixel 을 보간 하는 방법을 제안한다. 실험결과 제안하는 방법은 기존의 화소 단위의 에지 기반 디인터레이싱 방법보다 PSNR로 대표되는 객관적 성능과 주관적 화질 측면에서 우수한 결과를 나타내는 것을 볼 수 있었다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2008년도 International Symposium on Remote Sensing
• /
• pp.297-300
• /
• 2008

High-resolution mesoscale meteorological modeling requires more accurate and higher resolution digital elevation model (DEM) data. Shuttle Radar Topographic Mission (SRTM) has created 90 m DEM for entire globe and that is freely available for meteorological modeling and environmental applications. In this research, the effects of the topographic interpolation methods on high-resolution wind field simulation in the coastal regions were quantitatively analyzed using Weather Research and Forecasting (WRF) model with SRTM data. Sensitivity experiments with three different interpolation schemes (four-point bilinear, sixteen-point overlapping parabolic and nearest neighbor interpolation methods) were preformed using SRTM. In WRF modeling with sixteen-point overlapping parabolic interpolation, the coastal line and some small islands show more clearly than other cases. The maximum height of inland is around 140 meters higher, while the minimum of sea height is about 80 meter lower. As it concerns the results of each scheme it seems that the sixteen-point overlapping parabolic scheme indicates the well agreement with observed surface wind data. Consequently, topographic changes due to interpolation methods can lead to the significant influence on mesoscale wind field simulation of WRF modeling.

• 한국멀티미디어학회논문지
• /
• 제20권11호
• /
• pp.1750-1758
• /
• 2017

In super-resolution, various methods with Convolutional Neural Network(CNN) have recently been proposed. CNN based methods provide much higher image quality than conventional methods. Especially, VDSR outperforms other CNN based methods in terms of image quality. However, it requires a high computational complexity which prevents real-time processing. In this paper, the method to apply a deconvolution layer to VDSR is proposed to reduce computational complexity. Compared to original VDSR, the proposed method achieves the 4.46 times speed-up and its degradation in image quality is less than -0.1 dB which is negligible.

• 대한원격탐사학회지
• /
• 제26권3호
• /
• pp.305-316
• /
• 2010

The MODerate Resolution Imaging Spectroradiometer (MODIS) sensor has 36 bands at 250m, 500m, 1km spatial resolution. However, 500m or 1km MODIS data exhibits a few limitations when low resolution data is applied at small areas that possess complex land cover types. In this study, we produce seven 250m spectral bands by fusing two MODIS 250m bands into five 500m bands. In order to recommend the best fusion method by which one acquires MODIS data, we compare seven fusion methods including the Brovey transform, principle components algorithm (PCA) fusion method, the Gram-Schmidt fusion method, the least mean and variance matching method, the least square fusion method, the discrete wavelet fusion method, and the wavelet-PCA fusion method. Results of the above fusion methods are compared using various evaluation indicators such as correlation, relative difference of mean, relative variation, deviation index, peak signal-to-noise ratio index and universal image quality index, as well as visual interpretation method. Among various fusion methods, the local mean and variance matching method provides the best fusion result for the visual interpretation and the evaluation indicators. The fusion algorithm of 250m MODIS data may be used to effectively improve the accuracy of various MODIS land products.

• Spatial Information Research
• /
• 제13권4호
• /
• pp.345-354
• /
• 2005

초해상도 영상복원은 동일 지역을 촬영한 여러 장의 저해상도 영상을 이용하여 고해상도의 영상으로 재구성하는 영상처리 알고리즘 기법이다. 이 기법은 위성영상, 비디오 감시, 영상 강조 및 복원, 영상 모자이킹, 의료 영상과 같이 여러 장의 프레임 영상을 획득할 수 있는 분야에서 유용하게 사용될 수 있다. 본 연구에서는 지상을 촬영한 비디오 영상 열에 공간영역 초해상도 기법을 적용하였다. 실험에 사용된 영상은 높은 중복도로 촬영된 연속적인 비디오 영상에서 부표본화되었으며, 저해상도 영상과 고해상도 영상간의 관측 모델을 구성하고 초해상도 영상복원 기법중의 하나인 MAP 알고리즘을 적용하였다. MAP 기법을 이용하여 여러 장의 저해상도 영상에서 고해상도 영상으로 복원하였으며, 그 결과를 기존의 영상보간 방법들과 비교하였다.

• 한국정보통신학회논문지
• /
• 제22권1호
• /
• pp.147-154
• /
• 2018

본 논문에서는 다중 해상도 모델 기반 워게임 체계에서 빈번히 실행되어 모의 성능에 영향을 미치는 적군의 객체 정보를 해상도별로 획득하는 방안을 제안한다. 정보 요청법은 다중 해상도 모델이 특정 해상도의 정보를 선별하여 외부 객체에게 제공한다. 정보 공지법은 다중 해상도 모델이 모든 정보를 공지하고, 외부 객체가 특정 해상도의 정보를 선별하여 획득한다. 정보 공유법은 외부 객체가 공유 공간에 취합된 다중 해상도 모델의 모든 정보를 조회하여 특정 해상도의 정보를 획득한다. 제안 방법들에 대한 모의실험 결과, 정보 공유법이 정보 요청법과 정보 공지법에 비해 정보 획득 시간이 빨라 효과적인 방법임을 확인하였다. 또한 본 제안은 워게임에서 적군의 정보 획득 시간을 단축할 수 있어 워게임 운영 효율 증대 효과가 있다.

• 스마트미디어저널
• /
• 제8권4호
• /
• pp.64-71
• /
• 2019

The performance of an OMR (Optical Music Recognition) system is usually determined by the characterizing features of the input music score images. Low resolution is one of the main factors leading to degraded image quality. In this paper, we handle the low-resolution problem using the super-resolution technique. We propose the use of a deep neural network with instance normalization to improve the quality of music score images. We apply instance normalization which has proven to be beneficial in single image enhancement. It works better than batch normalization, which shows the effectiveness of shifting the mean and variance of deep features at the instance level. The proposed method provides an end-to-end mapping technique between the high and low-resolution images respectively. New images are then created, in which the resolution is four times higher than the resolution of the original images. Our model has been evaluated with the dataset "DeepScores" and shows that it outperforms other existing methods.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.648-650
• /
• 2003

Efficient multi-resolution image fusion aims to take advantage of the high spectral resolution of Landsat TM images and high spatial resolution of SPOT panchromatic images simultaneously. This paper presents a multi-resolution data fusion scheme, based on multirate image representation. Motivated by analytical results obtained from high-resolution multispectral image data analysis: the energy packing the spectral features are distributed in the lower frequency bands, and the spatial features, edges, are distributed in the higher frequency bands. This allows to spatially enhancing the multispectral images, by adding the high-resolution spatial features to them, by a multirate filtering procedure. The proposed method is compared with some conventional methods. Results show it preserves more spectral features with less spatial distortion.