• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.479-481
• /
• 2023

Generative Adversarial Networks (GANs) have facilitated substantial improvement in single-image super-resolution (SR) by enabling the generation of photo-realistic images. However, the high memory requirements of GAN-based SRs (mainly generators) lead to reduced performance and increased energy consumption, making it difficult to implement them onto resource-constricted devices. In this study, we propose an efficient and compressed architecture for the SR-GAN (generator) model using the model compression technique Knowledge Distillation. Our approach involves the transmission of knowledge from a heavy network to a lightweight one, which reduces the storage requirement of the model by 58% with also an increase in their performance. Experimental results on various benchmarks indicate that our proposed compressed model enhances performance with an increase in PSNR, SSIM, and image quality respectively for x4 super-resolution tasks.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.45 no.12
• /
• pp.90-96
• /
• 2008

Recently, various recording technologies are studied for optical data storage. After standardization of BD (Blu-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 proposed a neural network-based nonlinear equalizer (NNEQ) for Super-RENS discs. To mitigate the nonlinear ISI (Inter-Symbol Interference), we applied NARX (Nonlinear AutoRegressive eXogenous) which is a kind of neural networks. Its validity is tested with the RF signal samples obtained from a Super-RENS disc. The performance of the proposed equalizer is superior to the one without equalization and that of the Limit-EQ in terms of BER (Bit Error Rate).

• Journal of Broadcast Engineering
• /
• v.24 no.1
• /
• pp.132-141
• /
• 2019

The necessity of transmitting video data over a narrow-bandwidth exists steadily despite that video service over broadband is common. In this paper, we propose a scalable video coding framework for low-resolution video transmission over a very narrow-bandwidth network by super-resolution of decoded frames of a base layer using a convolutional neural network based super resolution technique to improve the coding efficiency by using it as a prediction for the enhancement layer. In contrast to the conventional scalable high efficiency video coding (SHVC) standard, in which upscaling is performed with a fixed filter, we propose a scalable video coding framework that replaces the existing fixed up-scaling filter by using the trained convolutional neural network for super-resolution. For this, we proposed a neural network structure with skip connection and residual learning technique and trained it according to the application scenario of the video coding framework. For the application scenario where a video whose resolution is $352{\times}288$ and frame rate is 8fps is encoded at 110kbps, the quality of the proposed scalable video coding framework is higher than that of the SHVC framework.

• The Bulletin of The Korean Astronomical Society
• /
• v.37 no.1
• /
• pp.61.2-61.2
• /
• 2012

We observed selected super cluster regions in M82 with 5'5 arcsec field-of-view using near-IR high resolution echelle spectrometer, IRCS, at the SUBARU 8.2 m telescope. The slit width of 0.15 arcsec makes the high resolution (R ${\approx}$ 20,000) spectra in the H and K bands. In this poster, we present sample spectra of [FeII] lines and ro-vibration lines of $H_2$ which trace ionic shocks in the intercloud regions and molecular shocks. The line widths of $Br{\gamma}$ line are also measured to derive the velocity dispersion within the super star clusters.

• Imaging Science in Dentistry
• /
• v.50 no.4
• /
• pp.331-337
• /
• 2020

Purpose: As cone-beam computed tomography (CBCT) has become the most widely used 3-dimensional (3D) imaging modality in the dental field, storage space and costs for large-capacity data have become an important issue. Therefore, if 3D data can be stored at a clinically acceptable compression rate, the burden in terms of storage space and cost can be reduced and data can be managed more efficiently. In this study, a deep learning network for super-resolution was tested to restore compressed virtual CBCT images. Materials and Methods: Virtual CBCT image data were created with a publicly available online dataset (CQ500) of multidetector computed tomography images using CBCT reconstruction software (TIGRE). A very deep super-resolution (VDSR) network was trained to restore high-resolution virtual CBCT images from the low-resolution virtual CBCT images. Results: The images reconstructed by VDSR showed better image quality than bicubic interpolation in restored images at various scale ratios. The highest scale ratio with clinically acceptable reconstruction accuracy using VDSR was 2.1. Conclusion: VDSR showed promising restoration accuracy in this study. In the future, it will be necessary to experiment with new deep learning algorithms and large-scale data for clinical application of this technology.

• 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.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.3
• /
• pp.234-245
• /
• 2023

The Arena Fragmentation Test(AFT) is designed to analyze warhead performance by measuring fragmentation data. In order to evaluate the results of the AFT, a set of AFT images are captured by high-speed cameras. To detect objects in the AFT image set, ResNet-50 based Faster R-CNN is used as a detection model. However, because of the low resolution of the AFT image set, a detection model has shown low performance. To enhance the performance of the detection model, Super-resolution(SR) methods are used to increase the AFT image set resolution. To this end, The Bicubic method and three SR models: ZSSR, EDSR, and SwinIR are used. The use of SR images results in an increase in the performance of the detection model. While the increase in the number of pixels representing a fragment flame in the AFT images improves the Recall performance of the detection model, the number of pixels representing noise also increases, leading to a slight decreases in Precision performance. Consequently, the F1 score is increased by up to 9 %, demonstrating the effectiveness of SR in enhancing the performance of the detection model.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.2
• /
• pp.209-216
• /
• 2018

High-resolution satellite imagery is used in diverse fields such as meteorological observation, topography observation, remote sensing (RS), military facility monitoring and protection of cultural heritage. In satellite imagery, low-resolution imagery can take place depending on the conditions of hardware (e.g., optical system, satellite operation altitude, image sensor, etc.) even though the images were obtained from the same satellite imaging system. Once a satellite is launched, the adjustment of the imaging system cannot be done to improve the resolution of the degraded images. Therefore, there should be a way to improve resolution, using the satellite imagery. In this study, a super resolution (SR) algorithm was adopted to improve resolution, using such low-resolution satellite imagery. The SR algorithm is an algorithm which enhances image resolution by matching multiple low-resolution images. In satellite imagery, however, it is difficult to get several images on the same region. To take care of this problem, this study performed the SR algorithm by calibrating geometric changes on images after applying automatic extraction of feature points and projection transform. As a result, a clear edge was found just like the SR results in which feature points were manually obtained.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.1
• /
• pp.17-25
• /
• 2007

In most electronic imaging applications, image with high resolution(HR) are desired. HR means that pixel density within an image is high, and therefore HR image can offer more details that may be critical in various applications. Digital images that are captured by CCD and CMOS cameras usually have a very low resolution, which significantly limits the performance of image recognition systems. Image super-resolution techniques can be applied to overcome the limits of these imaging systems. Super-resolution techniques have been proposed to increase the resolution by combining information from multiple images. To techniques were consisted of the registration algorithm for estimation and shift, the nearest neighbor interpolation using weight of acquired frames and presented frames. In this paper, it is proposed the image interpolation techniques using the wavelet base function. This is applied to embody a correct edge image and natural image when expend part of the still image by applying the wavelet base function coefficient to the conventional Super-Resolution interpolation method. And the proposal algorithm in this paper is confirmed to improve the image applying the nearest neighbor interpolation algorithm, bilinear interpolation algorithm.,bicubic interpolation algorithm through the computer simulation.

• Journal of Broadcast Engineering
• /
• v.24 no.4
• /
• pp.633-642
• /
• 2019

Recently, the convolution neural network (CNN) model at a single image super-resolution (SISR) have been very successful. The residual learning method can improve training stability and network performance in CNN. In this paper, we propose a SISR using recursive residual network architecture by introducing dense skip connections for learning nonlinear mapping from low-resolution input image to high-resolution target image. The proposed SISR method adopts a method of the recursive residual learning to mitigate the difficulty of the deep network training and remove unnecessary modules for easier to optimize in CNN layers because of the concise and compact recursive network via dense skip connection method. The proposed method not only alleviates the vanishing-gradient problem of a very deep network, but also get the outstanding performance with low complexity of neural network, which allows the neural network to perform training, thereby exhibiting improved performance of SISR method.