• Journal of Broadcast Engineering
• /
• v.26 no.4
• /
• pp.429-440
• /
• 2021

With the recent development of deep convolutional neural network learning, deep learning techniques applied to single image super-resolution are showing good results. One of the existing deep learning-based super-resolution techniques is RDN(Residual Dense Network), in which the initial feature information is transmitted to the last layer using residual dense blocks, and subsequent layers are restored using input information of previous layers. However, if all hierarchical features are connected and learned and a large number of residual dense blocks are stacked, despite good performance, a large number of parameters and huge computational load are needed, so it takes a lot of time to learn a network and a slow processing speed, and it is not applicable to a mobile system. In this paper, we use the residual dense structure, which is a continuous memory structure that reuses previous information, and the residual dense channel attention block using the channel attention method that determines the importance according to the feature map of the image. We propose a method that can increase the depth to obtain a large receptive field and maintain a concise model at the same time. As a result of the experiment, the proposed network obtained PSNR as low as 0.205dB on average at 4× magnification compared to RDN, but about 1.8 times faster processing speed, about 10 times less number of parameters and about 1.74 times less computation.

• Journal of the Society of Naval Architects of Korea
• /
• v.59 no.2
• /
• pp.72-79
• /
• 2022

Deep-learning-based Super-Resolution (SR) methods were evaluated to reconstruct pressure fields with a high resolution from low-resolution images taken from a coarse grid simulation. In addition to a canonical SRCNN(super-resolution convolutional neural network) model, two modified models from SRCNN, adding an activation function (ReLU or Sigmoid function) to the output layer, were considered in the present study. High resolution images obtained by three models were more vivid and reliable qualitatively, compared with a conventional super-resolution method of bicubic interpolation. A quantitative comparison of statistical similarity showed that SRCNN model with Sigmoid function achieved best performance with less dependency on original resolution of input images.

• 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 Broadcast Engineering
• /
• v.25 no.6
• /
• pp.965-972
• /
• 2020

Depth maps have distance information of objects. They play an important role in organizing 3D information. Color and depth images are often simultaneously obtained. However, depth images have lower resolution than color images due to limitation in hardware technology. Therefore, it is useful to upsample depth maps to have the same resolution as color images. In this paper, we propose a novel method to upsample depth map by shifting the pixel position instead of compensating pixel value. This approach moves the position of the pixel around the edge to the center of the edge, and this process is carried out in several steps to restore blurred depth map. The experimental results show that the proposed method improves both quantitative and visual quality compared to the existing methods.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.22 no.1
• /
• pp.21-27
• /
• 2004

It will be a real good news fer the people who were lost their hometown by the construction of a large dam to be restored to the farmer state. Focused on Cheung-pyung around where most part were submerged by the Chungju large Dam founded in eurly 1980s, It used remote sensing image restoration Technique in this study in order to restore topographical features before the flood with stereo effects. We gathered comparatively good satellite photos and remotely sensed digital images, then its made a new fusion image from these various satellite images and the topographical map which had been made before the water filled by the DAM. This task was putting together two kinds of different timed images. And then, we generated DEM including the outskirts of that area as matching current contour lines with the map. That could be a perfect 3D image of test areas around before when it had been water filled by making perspective images from all directions included north, south, east and west, fer showing there in 3 dimensions. Also, for close range visiting made of flying simulation can bring to experience their real space at that time. As a result of this experimental task, it made of new fusion images and 3-D perspective images and simulation live images by remotely sensed photos and images, old paper maps about vanished submerged Dam areas and gained of possibility 3-D terrain image restoration about submerged area by large Dam construction.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.10
• /
• pp.150-161
• /
• 2013

This paper presents a new system which produces high resolution 3D contents by capturing multiview images of an object using multiple cameras, and estimating geometric and texture information of the object from the captured images. Even though a variety of multiview image-based 3D reconstruction systems have been proposed, it was difficult to generate high resolution 3D contents because multiview image-based 3D reconstruction requires a large amount of memory and computation. In order to reduce computational complexity and memory size for 3D reconstruction, the proposed system predetermines the regions in input images where an object can exist to extract object boundaries fast. And for fast computation of a visual hull, the system represents silhouettes and 3D-2D projection/back-projection relations by chain codes and 1D homographies, respectively. The geometric data of the reconstructed object is compactly represented by a 3D segment-based data format which is called DoCube, and the 3D object is finally reconstructed after 3D mesh generation and texture mapping are performed. Experimental results show that the proposed system produces 3D object contents of $800{\times}800{\times}800$ resolution with a rate of 2.2 seconds per frame.

• Journal of Advanced Navigation Technology
• /
• v.6 no.2
• /
• pp.97-103
• /
• 2002

A recent technology trends in synthetic aperture radar(SAR) requires the ultra high resolution performance in detecting and precisely identifying the targets. In this paper, as a technique for enhancing the radar range resolution, the wide band chirp connection algorithm is presented by stitching the several chirp modules with unit bandwidth based on the linear frequency modulated chirp signal waveform. The principles of the digital chirp signal generation and its architecture for implementation is described with the wide band chirp signal generator, modulator, and demodulator. The performance analysis for the presented algorithm is given with the simulation results.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.257-269
• /
• 2023

There is currently a burgeoning demand for image synthesis from photos and videos using deep learning models. Existing video synthesis models solely extract motion information from the provided video to generate animation effects on photos. However, these synthesis models encounter challenges in achieving accurate lip synchronization with the audio and maintaining the image quality of the synthesized output. To tackle these issues, this paper introduces a novel framework based on an image animation approach. Within this framework, upon receiving a photo, a video, and audio input, it produces an output that not only retains the unique characteristics of the individuals in the photo but also synchronizes their movements with the provided video, achieving lip synchronization with the audio. Furthermore, a super-resolution model is employed to enhance the quality and resolution of the synthesized output.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.4
• /
• pp.71-76
• /
• 2020

In this paper, we propose a super-resolution method that reconstructs compressed low-resolution images into high-resolution images. We propose a CNN model with a small number of parameters, and even if quantization is applied to the proposed model, super-resolution can be implemented without deteriorating the image quality. To further improve the quality of the compressed low-resolution image, a new degradation model was proposed instead of the existing bicubic degradation model. The proposed degradation model is used only in the training process and can be applied by changing only the parameter values to the original CNN model. In the super-resolution image applying the proposed degradation model, visual artifacts caused by image compression were effectively removed. As a result, our proposed method generates higher PSNR values at compressed images and shows better visual quality, compared to conventional CNN-based SR methods.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.1
• /
• pp.54-63
• /
• 2011

This paper proposes a fast learning-based interpolation algorithm to up-scale an input low-resolution image into a high-resolution image. In conventional learning-based super-resolution, a certain relationship between low-resolution and high-resolution images is learned from various training images and a specific high frequency synthesis information is derived. And then, an arbitrary low resolution image can be super-resolved using the high frequency synthesis information. However, such super-resolution algorithms require heavy memory space to store huge synthesis information as well as significant computation due to two-dimensional matching process. In order to mitigate this problem, this paper presents one-dimensional patch-based learning and synthesis. So, we can noticeably reduce memory cost and computational complexity. Simulation results show that the proposed algorithm provides higher PSNR and SSIM of about 0.7dB and 0.01 on average, respectively than conventional bicubic interpolation algorithm.