• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.355-359
• /
• 2009

In this paper, we describe self-training super-resolution. Our approach is based on example based algorithms. Example based algorithms need training images, and selection of those changes the result of the algorithm. Consequently it is important to choose training images. We propose self-training based super-resolution algorithm which use an input image itself as a training image. It seems like other example based super-resolution methods, but we consider training phase as the step to collect primitive information of the input image. And some artifacts along the edge are visible in applying example based algorithms. We reduce those artifacts giving weights in consideration of the edge direction. We demonstrate the performance of our approach is reasonable several synthetic images and real images.

• International Journal of Computer Science & Network Security
• /
• v.21 no.12spc
• /
• pp.463-468
• /
• 2021

Single image super-resolution (SISR) is an image processing technique, and its main target is to reconstruct the high-quality or high-resolution (HR) image from the low-quality or low-resolution (LR) image. Currently, deep learning-based convolutional neural network (CNN) image super-resolution approaches achieved remarkable improvement over the previous approaches. Furthermore, earlier approaches used hand designed filter to upscale the LR image into HR image. The design architecture of such approaches is easy, but it introduces the extra unwanted pixels in the reconstructed image. To resolve these issues, we propose novel deep learning-based approach known as Lightweight deep CNN-based approach for Single Image Super-Resolution (LDCSIR). In this paper, we propose a new architecture which is inspired by ResNet with Inception blocks, which significantly drop the computational cost of the model and increase the processing time for reconstructing the HR image. Compared with the other state of the art methods, LDCSIR achieves better performance in terms of quantitively (PSNR/SSIM) and qualitatively.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.46 no.5
• /
• pp.39-48
• /
• 2009

Many super-resolution reconstruction algorithms have been proposed since it was the first proposed in 1984. The spatial domain approach of the super-resolution reconstruction methods is accomplished by mapping the low resolution image pixels into the high resolution image pixels. Generally, a super-resolution reconstruction algorithm by using the spatial domain approach has the noise problem because the low resolution images have different noise component, different PSF, and distortion, etc. In this paper, we proposed the new super-resolution reconstruction method that uses the L1 norm to minimize noise source and also uses the Huber norm to preserve edges of image. The proposed algorithm obtained the higher image quality of the result high resolution image comparing with other algorithms by experiment.

• Journal of Korea Multimedia Society
• /
• v.22 no.11
• /
• pp.1242-1250
• /
• 2019

Super resolution technique aims to convert a low-resolution image with coarse details to a corresponding high-resolution image with refined details. In the past decades, the performance is greatly improved due to progress of deep learning models. However, universal solution for various objects is a still challenging issue. We observe that learning super resolution with a general dataset has poor performance on faces. In this paper, we propose a super resolution fusion scheme that works well for both general- and face datasets to achieve more universal solution. In addition, object-specific feature extractor is employed for better reconstruction performance. In our experiments, we compare our fusion image and super-resolved images from one- of the state-of-the-art deep learning models trained with DIV2K and FFHQ datasets. Quantitative and qualitative evaluates show that our fusion scheme successfully works well for both datasets. We expect our fusion scheme to be effective on other objects with poor performance and this will lead to universal solutions.

• Korean Journal of Remote Sensing
• /
• v.39 no.4
• /
• pp.395-407
• /
• 2023

Super-resolution (SR) has great significance in image processing because it enables downstream vision tasks with high spatial resolution. Recently, SR studies have adopted deep learning networks and achieved remarkable SR performance compared to conventional example-based methods. Deep-learning-based SR models generally require low-resolution (LR) images and the corresponding high-resolution (HR) images as training dataset. Due to the difficulties in obtaining real-world LR-HR datasets, most SR models have used only HR images and generated LR images with predefined degradation such as bicubic downsampling. However, SR models trained on simple image degradation do not reflect the properties of the images and often result in deteriorated SR qualities when applied to real-world images. In this study, we propose an image degradation model for HR satellite images based on the modulation transfer function (MTF) of an imaging sensor. Because the proposed method determines the image degradation based on the sensor properties, it is more suitable for training SR models on remote sensing images. Experimental results on HR satellite image datasets demonstrated the effectiveness of applying MTF-based filters to construct a more realistic LR-HR training dataset.

• Spatial Information Research
• /
• v.19 no.1
• /
• pp.29-40
• /
• 2011

GOCI the world first Ocean Color Imager in Geostationary Orbit, which could obtain total 8 images of the same region a day, however, its spatial resolution(500m) is not enough to use for the accurate land application, Super Resolution(SR), reconstructing the high resolution(HR) image from multiple low resolution(LR) images introduced by computer vision field. could be applied to the time-series remotely sensed images such as GOCI data, and the higher resolution image could be reconstructed from multiple images by the SR, and also the cloud masked area of images could be recovered. As the precedent study for developing the efficient SR method for GOCI images, on this research, it reproduced the simulated data under the acquisition process of the remote sensed data, and then the simulated images arc applied to the proposed algorithm. From the proposed algorithm result of the simulated data, it turned out that low resolution(LR) images could be registered in sub-pixel accuracy, and the reconstructed HR image including RMSE, PSNR, SSIM Index value compared with original HR image were 0.5763, 52.9183 db, 0.9486, could be obtained.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.4
• /
• pp.156-163
• /
• 2021

Attention mechanisms have been used in deep learning-based computer vision systems, including single image super-resolution (SISR) networks. However, existing SISR networks with attention mechanism focused on real image super-resolution, so it is hard to know whether they are available for CG or VR images. In this paper, we attempt to apply a recent attention module, called balanced attention mechanism (BAM) module, to 12 state-of-the-art SISR networks, and then check whether the BAM module can achieve performance improvement in CG or VR image super-resolution. In our experiments, it has been confirmed that the performance improvement in CG or VR image super-resolution is limited and depends on data characteristics, size, and network type.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.4
• /
• pp.23-31
• /
• 2012

This paper proposes a motion estimation-based super resolution algorithm to restore input low-resolution images of large movement into a super-resolution image. It is difficult to find the sub-pixel motion estimation in images of large movement compared to typical experimental images. Also, it has disadvantage which have high computational complexity to find reference images and candidate images using general motion estimation method. In order to solve these problems for the traditional two-dimensional motion estimation using the proposed registration threshold that satisfy the conditions based on the reference image is determined. Candidate image with minimum weight among the best candidates for super resolution images, the restoration process to proceed with to find a new image registration algorithm is proposed. According to experimental results, the average PSNR of the proposed algorithm is 31.89dB and this is better than PSNR of traditional super-resolution algorithm and it also shows improvement of computational complexity.

• 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.

• Journal of the Korea Convergence Society
• /
• v.11 no.3
• /
• pp.19-26
• /
• 2020

In this paper, we propose a method to analyze the loss region of the dictionary-based super resolution result learned for image quality improvement and to map the learning data according to the analyzed loss region. In the conventional learned dictionary-based method, a result different from the feature configuration of the input image may be generated according to the learning image, and an unintended artifact may occur. The proposed method estimate loss information of low resolution images by analyzing the reconstructed contents to reduce inconsistent feature composition and unintended artifacts in the example-based super resolution process. By mapping the training data according to the final interpolation feature map, which improves the noise and pixel imbalance of the estimated loss information using a Gaussian-based kernel, it generates super resolution with improved noise, artifacts, and staircase compared to the existing super resolution. For the evaluation, the results of the existing super resolution generation algorithms and the proposed method are compared with the high-definition image, which is 4% better in the PSNR (Peak Signal to Noise Ratio) and 3% in the SSIM (Structural SIMilarity Index).