• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.6A
• /
• pp.464-472
• /
• 2012

In this paper, we propose a new super-resolution algorithm that uses successive frames by applying the block matching motion estimation algorithm. Usually, single frame super-resolution algorithms are based on probability or discrete wavelet transform (DWT) approach to extract high-frequency components of the input image, but only limited information is available for these algorithms. To solve this problem, various multiple-frame based super-resolution algorithms are proposed. The accuracy of registration between frames is a very important factor for the good performance of an algorithm. We therefore propose an algorithm using 6-Tap FIR filter to increase the accuracy of the image registration with sub-pixel unit. Proposed algorithm shows better performance than other conventional interpolation based algorithms such as nearest neighborhood, bi-linear and bi-cubic methods and results in about the same image quality as DWT based super-resolution algorithm.

• Korean Journal of Remote Sensing
• /
• v.36 no.3
• /
• pp.449-460
• /
• 2020

Super-resolution is a technique used to reconstruct an image with low-resolution into that of high-resolution. Recently, deep-learning based super resolution has become the mainstream, and applications of these methods are widely used in the remote sensing field. In this paper, we propose a super-resolution method based on the deep back-projection network model to improve the satellite image resolution by the factor of four. In the process, we customized the loss function with the edge loss to result in a more detailed feature of the boundary of each object and to improve the stability of the model training using generative adversarial network based on Wasserstein distance loss. Also, we have applied the detail preserving image down-scaling method to enhance the naturalness of the training output. Finally, by including the modified-residual learning with a panchromatic feature in the final step of the training process. Our proposed method is able to reconstruct fine features and high frequency information. Comparing the results of our method with that of the others, we propose that the super-resolution method improves the sharpness and the clarity of WorldView-3 and KOMPSAT-2 images.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.12
• /
• pp.2946-2952
• /
• 2014

Among the Example based Super Resolution(SR) techniques, Neighbor embedding(NE) has been inspired by manifold learning method, particularly locally linear embedding. However, the poor generalization of NE decreases the performance of such algorithm. The sizes of local training sets are always too small to improve the performance of NE. We propose the Learning Sparse-Neighbor Image Representation baesd on SVR having an excellent generalization ability to solve this problem. Given a low resolution image, we first use bicubic interpolation to synthesize its high resolution version. We extract the patches from this synthesized image and determine whether each patch corresponds to regions with high or low spatial frequencies. After the weight of each patch is obtained by our method, we used to learn separate SVR models. Finally, we update the pixel values using the previously learned SVRs. Through experimental results, we quantitatively and qualitatively confirm the improved results of the proposed algorithm when comparing with conventional interpolation methods and NE.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.3
• /
• pp.240-248
• /
• 2013

In this paper, we propose a super-resolution algorithm using an adaptive weighted interpolation(AWI) and discrete wavelet transform(DWT). In general, super-resolution algorithms for single-image, probability based operations have been used for searching high-frequency components. Consequently, the complexity of the algorithm is increased and it causes the increase of processing time. In the proposed algorithm, we first find high-frequency sub-bands by using DWT. Then we apply an AWI to the obtained high-frequency sub-bands to make them have the same size as the input image. Now, the interpolated high-frequency sub-bands and input image are properly combined and perform the inverse DWT. For the experiments, we use the down-sampled version of the original image($512{\times}512$) as a test image($256{\times}256$). Through experiment, we confirm the improved efficiency of the proposed algorithm comparing with interpolation algorithms and also save the processing time comparing with the probability based algorithms even with the similar performance.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.129-132
• /
• 2022

Super-resolution is the process of converting a low-quality image into a high-quality image. This study was conducted using ESPCN. In a super-resolution deep neural network, different quality images can be output even when receiving the same input data according to the activation function that determines the weight when passing through each node. Therefore, the purpose of this study is to find the most suitable activation function for super-resolution by applying the activation functions ReLU, ELU, and Swish and compare the quality of the output image for the same input images. The CelebaA Dataset was used as the dataset. Images were cut into a square during the pre-processing process then the image quality was lowered. The degraded image was used as the input image and the original image was used for evaluation. As a result, ELU and swish took a long time to train compared to ReLU, which is mainly used for machine learning but showed better performance.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.45 no.3
• /
• pp.53-60
• /
• 2008

Recently, various recording technologies are studied for optical data storage. After standardization of BD (Blue-ray Disc) and HD-DVD (High-Definition Digital Versatile Disc), the industry is looking for a suitable technology for next generation optical data storage. Super-RENS (Super-resolution near field structure) technique, which is capable of compatibility with other systems, is one of next optical data storage. In this paper, we analyze the nonlinearity of Super-RENS read-out signal through the bicoherence test, which uses HOS (Higher-Order Statistics) and apply neural networks for nonlinear modeling. The model structure considered in this paper is the NARX (Nonlinear AutoRegressive eXogenous) model. The experiment results indicate that the read-out signals have nonlinear characteristics. In addition, it verified the possibility that neural networks can be utilized for nonlinear modeling of Super-RENS systems.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.378-390
• /
• 2022

With the recent development of deep composite multiplication neural network learning, deep learning techniques applied to single-image super-resolution have shown good results, and the strong expression ability of deep networks has enabled complex nonlinear mapping between low-resolution and high-resolution images. However, there are limitations in applying it to real-time or low-power devices with increasing parameters and computational amounts due to excessive use of composite multiplication neural networks. This paper uses blocks that extract hierarchical characteristics little by little using information distillation and suggests the Recursive Distillation Super Resolution Network (RDSRN), a lightweight network that improves performance by making more accurate high frequency components through high frequency residual purification blocks. It was confirmed that the proposed network restores images of similar quality compared to RDN, restores images 3.5 times faster with about 32 times fewer parameters and about 10 times less computation, and produces 0.16 dB better performance with about 2.2 times less parameters and 1.8 times faster processing time than the existing lightweight network CARN.

• Journal of the Korean Society for Precision Engineering
• /
• v.19 no.9
• /
• pp.172-178
• /
• 2002

One of the ever-increasing demands on the performances of heterodyne interferometers is to improve the measurement resolution, of which current state -of-the-art reaches the region of sub-nanometers. So far, the demand has been met by increasing the clock speed that drives the electronics involved fur the phase measurement of the Doppler shift, but its further advance is being hampered by the technological limit of modem electronics. To cope with the problem, in this investigation, we propose a new scheme of phase -measuring electronics that reduces the measurement resolution without further increase in clock speed. Our scheme adopts a super-heterodyne technique that lowers the original beat frequency to a level of 1 MHz by mixing it with a stable reference signal generated from a special phase- locked-loop. The technique enables us to measure the phase of Doppler shift with a resolution of 1.58 nanometer at a sampling rate of 1 MHz. To avoid the undesirable decrease in the maximum measurable speed caused by the lowered beat frequency, a special form of frequency up-down counting technique is combined with the super-heterodyning. This allows performing required phase unwrapping simply by using programmable digital gates without 2n ambiguities up to the maximum velocity guaranteed by the original beat frequency.

• Journal of Korea Multimedia Society
• /
• v.20 no.11
• /
• 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.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.4
• /
• pp.543-548
• /
• 2021

The higher the resolution of the image, the higher the satisfaction of the viewers of the image, and the super-resolution imaging has a considerable increase in research value among the fields of computer vision and image processing. In this study, the main features of low-resolution image LR are extracted mainly using deep learning super-resolution models. It learns and reconstructs the extracted features, and focuses on reconstruction-based algorithms that generate high-resolution image HR. In this paper, we investigate SRCNN and VDSR in a super-resolution algorithm model based on reconstruction. The structure and algorithm process of the SRCNN and VDSR model are briefly introduced, and the multi-channel and special form are also examined in the improved performance evaluation function, and understand the performance of each algorithm through experiments. In the experiment, an experiment was performed to compare the results of the SRCNN and VDSR models with the peak signal-to-noise ratio and image structure similarity, so that the results can be easily judged.