• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.1
• /
• pp.26-31
• /
• 2010

We proposed a new directional edge based interpolation, DEBI, by combining two weighted directional information to reduce blurred edges and annoying artifacts. Four isotropic gradient masks are employed in defining edge directions and they are proven to hold a first order derivative relation with respect to a rotating coordinate. Two minimum gradients among four absolute directional results are shown to be sufficient to describe slant edges efficiently. Compared with widely used bilinear and bicubic interpolation methods, the proposed algorithm results in a noticeable improvement along edge area.

• Progress in Medical Physics
• /
• v.11 no.1
• /
• pp.39-47
• /
• 2000

In Stereotactic Radiosurgery (SRS), there are three imaging methods of target localization, such as digital subtraction Angiography (DSA), computed tomography (CT), magnetic resonance imaging (MRI). Especially, DSA and MR images have a distortion effect generated by each modality. In this research, image properties of DSA were studied. A first essential condition in SRS is an accurate information of target locations, since high dose used to treat a patient may give a complication on critical organ and normal tissue. Hut previous localization program did not consider distortion effect which was caused by image intensifier (II) of DSA. A neurosurgeon could not have an accurate information of target locations to operate a patient. In this research, through distortion correction, we tried to calculate accurate target locations. We made a grid phantom to correct distortion, and a target phantom to evaluate localization algorithm. The grid phantom was set on the front of II, and DSA images were obtained. Distortion correction methods consist of two parts: 1. Bilinear transform for geometrical correction and bilinear interpolation for gray level correction. 2. Automatic detection method for calculating locations of grid crosses, fiducial markers, and target balls. Distortion was corrected by applying bilinear transform and bilinear interpolation to anterior-posterior and left-right image, and locations of target and fiducial markers were calculated by the program developed in this study. Localization errors were estimated by comparing target locations calculated in DSA images with absolute locations of target phantom. In the result, the error in average with and without distortion correction is $\pm$0.34 mm and $\pm$0.41 mm respectively. In conclusion, it could be verified that our localization algorithm has an improved accuracy and acceptability to patient treatment.

• Journal of Korea Multimedia Society
• /
• v.20 no.12
• /
• pp.1865-1873
• /
• 2017

The feature-based visual SLAM requires 3D positions for the extracted feature points to perform 3D-2D motion estimation. LiDAR can provide reliable and accurate 3D position information with low computational burden, while stereo camera has the problem of the impossibility of stereo matching in simple texture image region, the inaccuracy in depth value due to error contained in intrinsic and extrinsic camera parameter, and the limited number of depth value restricted by permissible stereo disparity. However, the sparsity of LiDAR data may increase the inaccuracy of motion estimation and can even lead to the result of motion estimation failure. Therefore, in this paper, we propose three interpolation methods which can be applied to interpolate sparse LiDAR data. Simulation results obtained by applying these three methods to a visual odometry algorithm demonstrates that the selective bilinear interpolation shows better performance in the view point of computation speed and accuracy.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.2
• /
• pp.59-67
• /
• 2012

In this paper, an efficient interpolation scheme using gradient information between adjacent pixel values was proposed to estimate the value of interpolated pixel to have the locality which exists in real image and the characteristic of simple convex surface and simple concave surface which exist partially in the real image. PSNR(Peak Signal to Noise Ratio) was used to evaluate the performance of the proposed scheme. The PSNR values of the magnified images using the proposed scheme are greater than those of the magnified images using the previous interpolation schemes.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.21 no.6
• /
• pp.153-160
• /
• 2011

Recently, the use of security surveillance system including CCTV is increasing due to the increase of terrors and crimes. At the same time, interest of face recognition at a distance using surveillance cameras has been increasing. Accordingly, we analyzed the performance of face recognition according to distance using PCA-based face recognition and interpolation. In this paper, we used Nearest, Bilinear, Bicubic, Lanczos3 interpolations to interpolate face image. As a result, we confirmed that existing interpolation have an few effect on performance of PCA-based face recognition and performance of PCA-based face recognition is improved by including face image according to distance in traning data.

• Journal of Multimedia Information System
• /
• v.8 no.2
• /
• pp.101-110
• /
• 2021

In computer vision, single-image super resolution has been an area of research for a significant period. Traditional techniques involve interpolation-based methods such as Nearest-neighbor, Bilinear, and Bicubic for image restoration. Although implementations of convolutional neural networks have provided outstanding results in recent years, efficiency and single model multi-scalability have been its challenges. Furthermore, previous works haven't placed enough emphasis on real-number scalability. Interpolation-based techniques, however, have no limit in terms of scalability as they are able to upscale images to any desired size. In this paper, we propose a convolutional neural network possessing the advantages of the interpolation-based techniques, which is also efficient, deeming it suitable in practical implementations. It consists of convolutional layers applied on the low-resolution space, post-up-sampling along the end hidden layers, and additional layers on high-resolution space. Up-sampling is applied on a multiple channeled feature map via bicubic interpolation using a single model. Experiments on architectural structure, layer reduction, and real-number scale training are executed with results proving efficient amongst multi-scale learning (including scale multi-path-learning) based models.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2001.05a
• /
• pp.1050-1055
• /
• 2001

A software for prediction of apartment noise level ihas been improved. The program is based on the ray tracing technique which has been widely used in the environmental noise analysis and prediction. Especially for prediction of environmental noise in apartment complex, this program is advanced in the graphics routine by bilinear interpolation. In this paper, we analyze the railway noise distribution in apartment environment and develop a 3D graphics routine for illustrating the noise level.

• Journal of the Korean Mathematical Society
• /
• v.53 no.6
• /
• pp.1411-1429
• /
• 2016

We consider the nite volume element method for elliptic problems using isoparametric bilinear elements on quadrilateral meshes. A gradient recovery method is presented by using the patch interpolation technique. Based on some superclose estimates, we prove that the recovered gradient $R({\nabla}u_h)$ possesses the superconvergence: ${\parallel}{\nabla}u-R({\nabla}u_h){\parallel}=O(h^2){\parallel}u{\parallel}_3$. Finally, some numerical examples are provided to illustrate our theoretical analysis.

• Journal of Biomedical Engineering Research
• /
• v.30 no.1
• /
• pp.34-40
• /
• 2009

Transforms including translation and rotation are required for registering two or more images. In medical applications, different registration methods have been applied depending on the structures: for rigid bodies such as bone structures, affine transformation was widely used. In most previous research, a single transform was used for registering the whole images, which resulted in low registration accuracy especially when the degree of deformation was high between two images. In this paper, a novel registration method is introduced which is based image sub-division and bilinear interpolation of transformations. The proposed method enhanced the registration accuracy by 40% comparing with Trimmed ICP for registering color and MRI images.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1998.06a
• /
• pp.127-130
• /
• 1998

3차원의 영상을 2차원 화면에 투영시키기 위해서는 영상에 대한 카메라의 촬영 각도와 촬영 거리를 일치시켜주어야 한다. 이를 위해서 다른 각도와 거리에서 촬영한 영상을 같은 각도와 거리에서 촬영한 영상으로 만들어 주는 변환을 만들었다. 그러나 이러한 변환을 하였을 때 디지털 영상은 경계부분이나 직선 등이 끊어지는 현상이 발생하게 된다. 본 논문에서는 변환된 영역에서의 정수 점에 원래 영상의 점을 바로 대응시키지 않고 주변 점의 다항식으로 나타낼 수 있다. 이 다항식의 계수를 정하는 방법으로 변환된 영역에서의 정수 점에 대응하는 원래 영상의 점을 구하여 2차 선형 보간법(Bilinear interpolation)을 사용하였다. 변환된 영역에서의 정수 점에 대응하는 원래 영상에서의 좌표를 얻기 위해 주변 4개의 점 내에서는 부분적으로 선형적임을 가정하였다. 위의 방법으로 주변 점들의 에너지를 각 정수 점들로 분산시킴으로써 경계 점이나 직선들이 잘 보존된 상태로 변환할 수 있었다.