• 융합신호처리학회 학술대회논문집
• /
• 한국신호처리시스템학회 2000년도 하계종합학술대회논문집
• /
• pp.285-288
• /
• 2000

본 논문에서는 카메라 플랫홈의 흔들림 등으로 인한 외부 영향으로 출렁이는 비디오를 전자적으로 안정화시키는 방법을 제안한다. LOG operator〔1〕을 이용하여 특징점을 잡고 그 특징점을 중심으로 일정크기의 subblock에 대해서만 correlation을 구한다. Least Square를 이용하여 모션 파라메타를 측정하고 모션 보상을 행하는데 현재의 영상을 기준 좌표계로 변환하고 명암값 보간을 하게된다. 이 때 기존 연구에서 많이 사용한 bilinear 보간법의 단점인 대비가 첨예한 곳에서의 averaging 효과를 없애기 위해 본 연구에서는 대비가 첨예한 곳에서는 nearest neighbor 보간을 사용하고 그렇지 않은 곳에서는 bilinear 보간을 사용하는 hybrid한 방법을 채택하였다. 그 결과 사람이 카메라를 손에 들고 움직일 때 생기는 출렁이는 비디오에 대해 대부분의 카메라 움직임을 안정화시킬 수 있었다.

• 대한디지털의료영상학회논문지
• /
• 제3권1호
• /
• pp.110-118
• /
• 1997

Recently, to make a diagnosis of the patient different X-Ray examinations are used. To name a few, Computed Tomography(CT). Magnetic Resonance Image(MRI) Single Photon Emission Computed Tomography(SPET) and Positron Emission Tomography(PET). But diagnosticians face difficulties sometimes when they make a diagnosis with images from those examinations. One of the problem is whether the Lesions of the patient is captured in the image correctly. Another one is whether the images are taken with same angle. in this paper, a study 9 on the method to obtain the hybrid image from the different images to different examinations. The procedure done in this paper is described as future study. Although small errors in position between images would occurred, this method more useful as it does not make patients in convenient. To reconstruct a image, some images are scanned by scanner and stored to personal computer for further image processing with Aldus photostyler program. The method to generate a sharpened image are also described.

• Fibers and Polymers
• /
• 제7권1호
• /
• pp.57-61
• /
• 2006

A study has been made on the quantification and evaluation of fabric pilling using two-dimensional and three-dimensional hybrid imaging methods. Two-dimensional imaging method was good for some samples while three-dimensional measurement method for others, according to the properties of their base fabric. Various image processing techniques as well as three-dimensional data processing algorithms were applied for the extraction of pills from measured data and a series of shape parameters have been defined for the objective evaluation of fabric pilling. An evaluation criterion that is compatible with the conventional evaluation method has been proposed by applying the new evaluation method to the current photographic standards.

• Journal of Information Processing Systems
• /
• 제14권5호
• /
• pp.1176-1194
• /
• 2018

The object classification using the images' contents is a big challenge in computer vision. The superpixels' information can be used to detect and classify objects in an image based on locations. In this paper, we proposed a methodology to detect and classify the image's pixels' locations using enhanced bag of words (BOW). It calculates the initial positions of each segment of an image using superpixels and then ranks it according to the region score. Further, this information is used to extract local and global features using a hybrid approach of Scale Invariant Feature Transform (SIFT) and GIST, respectively. To enhance the classification accuracy, the feature fusion technique is applied to combine local and global features vectors through weight parameter. The support vector machine classifier is a supervised algorithm is used for classification in order to analyze the proposed methodology. The Pascal Visual Object Classes Challenge 2007 (VOC2007) dataset is used in the experiment to test the results. The proposed approach gave the results in high-quality class for independent objects' locations with a mean average best overlap (MABO) of 0.833 at 1,500 locations resulting in a better detection rate. The results are compared with previous approaches and it is proved that it gave the better classification results for the non-rigid classes.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2005년도 Proceedings of ISRS 2005
• /
• pp.571-574
• /
• 2005

In this paper, we propose an efficient image segmentation technique for large-sized monochrome images using a hybrid approach which combines threshold and region-based techniques. First, an image is partitioned into fixed-size blocks and for each block the representative intensity is determined by averaging pixel intensities within the block. Next, the neighborhood blocks that have similar characteristics with respect to a specific threshold are merged in order to form candidate regions. Finally, those candidate regions are refined to get final target object regions by merging regions considering the spatial locality and certain criteria. We have performed experiments on images selected from various domains and showed that our technique was able to extract target object regions appropriately from most images.

• Journal of Power Electronics
• /
• 제17권1호
• /
• pp.200-211
• /
• 2017

Conventional boost-full-bridge and boost-hybrid-bridge two-stage inverters are widely applied in order to adapt to the wide dc input voltage range of photovoltaic arrays. However, the efficiency of the conventional topology is not fully optimized because additional switching losses are generated in the voltage conversion so that the input voltage rises and then falls. Moreover, the electrolytic capacitors in a dc-link lead to a larger volume combined with increases in both weight and cost. This paper proposes a higher efficiency inverter with time-sharing synchronous modulation. The energy transmission paths, wheeling branches and switching losses for the high-frequency switches are optimized so that the overall efficiency is greatly improved. In this paper, a contrastive analysis of the component losses for the conventional and proposed inverter topologies is carried out in MATLAB. Finally, the high-efficiency under different switching frequencies and different input voltages is verified by a 3 kW prototype.

• 한국정보통신학회논문지
• /
• 제16권6호
• /
• pp.1285-1290
• /
• 2012

3D 산업이 발달하고 3D와 관련된 콘텐츠 및 장비가 점차 보편화 되면서 사람들은 3D 다음 세대의 미디어 및 서비스에 대한 관심을 보이고 있다. 많은 사람들은 그 강력한 후보자로 디지털 홀로그램을 생각하고 있다. 홀로그램 기술은 일반적인 비디오 시스템들과 마찬가지로 획득(혹은 생성), 처리, 그리고 재생으로 나누어질 수 있다. 본 논문에서는 하이브리드 영상획득 장치를 이용한 디지털 홀로그래픽 디스플레이 시스템에 대해 논의하고 새로운 시스템의 구조를 제안한다. 제안한 방식을 통해서 다양한 해상도를 갖는 디스플레이, 수신단의 연산 능력, 그리고 네트워크의 대역폭에 따라 적응적으로 홀로그래픽 비디오를 서비스할 수 있다는 것을 보였다.

• 디지털산업정보학회논문지
• /
• 제5권2호
• /
• pp.51-76
• /
• 2009

In this paper, a new hybrid opto-digital multiple information hiding and real-time extracting system was suggested and implemented using a volume holographic optical correlator. The multiple information hiding system in which the multiple information can be hided in an arbitrary cover image was digitally implemented by using the combination of pseudo random sequence and Hadamard matrix. In addition, a real-time optical extraction system in which the hided multiple information in a cover image can be extracted in real-time was implemented by using a volume holographic optical correlator. In the VanderLugt-type holographic optical correlator used in this experiment, the multiful matched spatial filters corresponding each of the hided informations were recorded in a photorefractive crystal by using the moving window-based angular multiplexing scheme.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권9호
• /
• pp.2882-2903
• /
• 2022

Image super-resolution (SR) processing is of great value in the fields of digital image processing, intelligent security, film and television production and so on. This paper proposed a densely connected deep learning network based on cascade architecture, which can be used to solve the problem of super-resolution in the field of image quality enhancement. We proposed a more efficient residual scaling dense block (RSDB) and the multi-channel cascade architecture to realize more efficient feature reuse. Also we proposed a hybrid loss function based on L₁ error and L_∞ error to achieve better L_∞ error performance. The experimental results show that the overall performance of the network is effectively improved on cascade architecture and residual scaling. Compared with the residual dense net (RDN), the PSNR / SSIM of the new method is improved by 2.24% / 1.44% respectively, and the L_∞ performance is improved by 3.64%. It shows that the cascade connection and residual scaling method can effectively realize feature reuse, improving the residual convergence speed and learning efficiency of our network. The L_∞ performance is improved by 11.09% with only a minimal loses of 1.14% / 0.60% on PSNR / SSIM performance after adopting the new loss function. That is to say, the L_∞ performance can be improved greatly on the new loss function with a minor loss of PSNR / SSIM performance, which is of great value in L_∞ error sensitive tasks.

• 전기전자학회논문지
• /
• 제28권3호
• /
• pp.352-359
• /
• 2024

본 논문에서는 이미지의 지역적 및 전역적 특징을 결합하여 이미지의 미학적 품질을 자동으로 평가할 수 있는 CNN-ViT 하이브리드 모델을 제안한다. 이 접근 방식에서는 CNN을 사용하여 색상 및 객체 배치와 같은 지역적 특징을 추출하고, ViT를 통해 전역적 특징을 반영하여 이미지의 미학적 가치를 분석한다. Color composition은 입력 이미지에서 주요 색상을 추출해 생성한 컬러 팔레트를 CNN에 통과시켜 얻은 값이며, Rule of Third는 이미지 속 오브젝트가 삼등분할점에 얼마나 근접한지를 정량적으로 평가한 점수로 사용된다. 이러한 값들은 모델에 이미지의 주요 평가 요소인 색채와 공간 균형에 대한 정보를 제공한다. 모델은 이를 바탕으로 이미지의 점수와 색상, 공간의 균형 간에 연관성을 분석하며, 인간의 평가 분포와 유사한 점수를 추측하도록 설계되었다. 실험 결과, AADB 이미지 데이터베이스에서 스피어만순위상관계수(SRCC)에서는 0.716을 기록하여 순위 예측에서 더 일관된 결과를 제공 했으며, 피어슨상관계수(LCC)에서도 0.72을 기록하여 기존 연구 모델보다 2~4% 정도 향상된 결과를 보였다.