• ETRI Journal
• /
• v.32 no.3
• /
• pp.390-394
• /
• 2010

In this paper, we propose a new super-resolution technique based on interpolation of the high-frequency subband images obtained by discrete wavelet transform (DWT) and the input image. The proposed technique uses DWT to decompose an image into different subband images. Then the high-frequency subband images and the input low-resolution image have been interpolated, followed by combining all these images to generate a new super-resolved image by using inverse DWT. The proposed technique has been tested on Lena, Elaine, Pepper, and Baboon. The quantitative peak signal-to-noise ratio (PSNR) and visual results show the superiority of the proposed technique over the conventional and state-of-art image resolution enhancement techniques. For Lena's image, the PSNR is 7.93 dB higher than the bicubic interpolation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.6
• /
• pp.497-502
• /
• 2013

In this paper, we aim to evaluate the performance of the digital signal processing (DSP) algorithm used in [8] in order to improve the resolution of spectrometers with fixed number of low-cost, non-ideal filters. In such spectrometers, the resolution is limited by the number of filters. We aim to demonstrate via new experiments that the resolution improvement by six times over the conventional limit is possible by using the DSP algorithm as claimed by [8].

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.06a
• /
• pp.98-101
• /
• 2019

At present, deep convolutional neural networks have made a very important contribution in single-image super-resolution. Through the learning of the neural networks, the features of input images are transformed and combined to establish a nonlinear mapping of low-resolution images to high-resolution images. Some previous methods are difficult to train and take up a lot of memory. In this paper, we proposed a simple and compact deep recursive residual network learning the features for single image super resolution. Global residual learning and local residual learning are used to reduce the problems of training deep neural networks. And the recursive structure controls the number of parameters to save memory. Experimental results show that the proposed method improved image qualities that occur in previous methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.63-66
• /
• 2018

Recently, a convolutional neural network (CNN) models at single image super-resolution have been very successful. Residual learning improves training stability and network performance in CNN. In this paper, we compare four convolutional neural network models for super-resolution (SR) to learn nonlinear mapping from low-resolution (LR) input image to high-resolution (HR) target image. Four models include general CNN model, global residual learning CNN model, local residual learning CNN model, and the CNN model with global and local residual learning. Experiment results show that the results are greatly affected by how skip connections are connected at the basic CNN network, and network trained with only global residual learning generates highest performance among four models at objective and subjective evaluations.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.272-275
• /
• 2020

Light Field (LF) cameras capture both spatial and directional information of light rays. Current LF cameras have a problem of a low spatial resolution. There have been lots of existing works carried out to improve the resolution of LF images. In this paper, those existing works will be divided into two categories: hardware-based approaches and software-based approaches, and we will look into and compare several experiment results in order for LF spatial resolution enhancement. Finally, the direction for the future spatial resolution enhancement will be suggested.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.6
• /
• pp.67-74
• /
• 2006

In this paper, a tiled display system is developed to get a high-resolution image in visualization of large-scale structural analysis data with low-resolution display devices and low-cost cluster computer system. Concerning the hardware system, some of the crucial points are investigated, and a new beam-projector positioner is designed and manufactured to resolve the keystone phenomena which result in distorted image. In the development of tiled display software, Qt and OpenGL are utilized for GUI and rendering, respectively. To obtain the entire tiled image, LAM-MPI is utilized to synchronize the several sub-images produced from each cluster computer node.

• Childhood Kidney Diseases
• /
• v.20 no.1
• /
• pp.18-22
• /
• 2016

Purpose: The previous reports regarding VUR resolution were not precise due to early frequent surgical intervention. We evaluated the spontaneous resolution (SR) rate and the incidence of new renal scars in primary VUR, focusing on severe reflux. Methods: Medical records of 334 patients with primary VUR who were on medical prophylaxis without surgery for 1 to 9 years, were retrospectively reviewed. Medical prophylaxis was initiated with low-dose antibiotic prophylaxis or probiotics. Radioisotope cystourethrography was performed every 1 to 3 years until SR of reflux. New renal scar was evaluated with follow-up $^{99m}Tc$ DMSA renal scan. Results: The SR rates decreased as VUR grades were getting higher (P=0.00). The overall and annual SR were 58.4% and 14.9%/yr in grade IV reflux and 37.5% and 9.3%/yr in grade V reflux. The median times of SR were 38 months in grade IV reflux and 66 months in grade V reflux. The probable SR rates in grade IV and V reflux were 7.8% and 8.9% in the 1st year, 46.0% and 30.8% in the 3rd year and 74.4% and 64.4% in the 5th year. The incidences of new renal scars between low to moderate reflux and severe reflux showed no significant difference (P=0.32). Conclusion: The SR rates of severe primary VUR were higher than previously reported and most new renal scars were focal and mild.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.113-117
• /
• 2020

Images were taken under various weather such as rain, haze, snow often show low visibility, which can dramatically decrease accuracy of some tasks in computer vision: object detection, segmentation. Besides, previous work to enhance image usually downsample the image to receive consistency features but have not yet good upsample algorithm to recover original size. So, in this research, we jointly implement removal streak in heavy rain image and super resolution using a deep network. We put forth a 2-stage network: a multi-model network followed by a refinement network. The first stage using rain formula in the single image and two operation layers (addition, multiplication) removes rain streak and noise to get clean image in low resolution. The second stage uses refinement network to recover damaged background information as well as upsample, and receive high resolution image. Our method improves visual quality image, gains accuracy in human action recognition task in datasets. Extensive experiments show that our network outperforms the state of the art (SoTA) methods.

• Korean Journal of Remote Sensing
• /
• v.40 no.2
• /
• pp.141-150
• /
• 2024

Ground subsidence in urban areas is mainly caused by anthropogenic factors such as excessive groundwater extraction and underground infrastructure development in the subsurface composed of soft materials. Global Navigation Satellite System data with high temporal resolution have been widely used to measure surface displacements accurately. However, these point-based terrestrial measurements with the low spatial resolution are somewhat limited in observing two-dimensional continuous surface displacements over large areas. The synthetic aperture radar interferometry (InSAR) technique can construct relatively high spatial resolution surface displacement information with accuracy ranging from millimeters to centimeters. Although constellation operations of SAR satellites have improved the revisit cycle, the temporal resolution of space-based observations is still low compared to in-situ observations. In this study, we evaluate the extraction of a time-series of surface displacement in Incheon Metropolitan City, South Korea, using the small baseline subset technique implemented using the commercial software, Gamma. For this purpose, 24 COSMO-SkyMed X-band SAR observations were collected from July 12, 2011, to August 27, 2012. The time-series surface displacement results were improved by reducing random phase noise, correcting residual phase due to satellite orbit errors, and mitigating nonlinear atmospheric phase artifacts. The perpendicular baseline of the collected COSMO-SkyMed SAR images was set to approximately 2-300 m. The surface displacement related to the ground subsidence was detected approximately 1 cm annually around a few Incheon Subway Line 2 route stations. The sufficient coherence indicates that the satellite orbit has been precisely managed for the interferometric processing.

• Journal of Broadcast Engineering
• /
• v.24 no.5
• /
• pp.703-712
• /
• 2019

Single image Super-Resolution (SISR) aims to generate a visually pleasing high-resolution image from its degraded low-resolution measurement. In recent years, deep learning - based super - resolution methods have been actively researched and have shown more reliable and high performance. A typical method is WaveletSRNet, which restores high-resolution images through wavelet coefficient learning based on feature maps of images. However, there are two disadvantages in WaveletSRNet. One is a big processing time due to the complexity of the algorithm. The other is not to utilize feature maps efficiently when extracting input image's features. To improve this problems, we propose an efficient single image super resolution method, named RDB-WaveletSRNet. The proposed method uses the residual dense block to effectively extract low-resolution feature maps to improve single image super-resolution performance. We also adjust appropriated growth rates to solve complex computational problems. In addition, wavelet packet decomposition is used to obtain the wavelet coefficients according to the possibility of large scale ratio. In the experimental result on various images, we have proven that the proposed method has faster processing time and better image quality than the conventional methods. Experimental results have shown that the proposed method has better image quality by increasing 0.1813dB of PSNR and 1.17 times faster than the conventional method.