• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.5
• /
• pp.2366-2395
• /
• 2018

Multi-view super-resolution (MVSR) refers to the process of reconstructing a high-resolution (HR) image from a set of low-resolution (LR) images captured from different viewpoints typically by different cameras. These multi-view images are usually obtained by a camera array. In our previous work [1], we super-resolved multi-view LR images via image fusion (IF) and blind deblurring (BD). In this paper, we present a new MVSR method that jointly realizes IF and BD based on an integrated energy function optimization. First, we reformulate the MVSR problem into a multi-channel blind deblurring (MCBD) problem which is easier to be solved than the former. Then the depth map of the desired HR image is calculated. Finally, we solve the MCBD problem, in which the optimization problems with respect to the desired HR image and with respect to the unknown blur are efficiently addressed by the alternating direction method of multipliers (ADMM). Experiments on the Multi-view Image Database of the University of Tsukuba and images captured by our own camera array system demonstrate the effectiveness of the proposed method.

• 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 IKEEE
• /
• v.21 no.1
• /
• pp.73-76
• /
• 2017

SiC devices have drawn much attentions for its wide band gap material properties. Especially 4H-SiC Schottky barrier diode is widely used for its rapid switching speed and low forward voltage drop. However, the low reliability of Schottky barrier diode has many problems that Super Barrier Rectifier(SBR) was researched for alternative. makes 4H-SiC trench-type accumulation super barrier rectifier(TASBR) is analyzed and proposed in this paper. We could verified that forward voltage drop was improved 21.06% without severe degradation of reverse breakdown voltage and leakage current based on the results from 2-D numerical simulations. With this novel rectifier structure, we can expect application with less power loss.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.45 no.3
• /
• pp.34-39
• /
• 2008

In this paper, we propose a bit-interleaved coded modulation (BICM) a lied to improved super-orthogonal space-time block code(SOSTBC). The proposed system achieves a greater diversity gain than that of super-orthogonal space-time trellis code (SOSTTC) with similar decoding complexity. Since, using the improved SOSTBC, the bit diversity carl be full diversity of SOSTBC. In contrast, BICM applied to Jafarkhani's SOSTBC is difficult to achieve a greater diversity gain than that of SOSTTC, because every bit diversity of the system is 1.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.9
• /
• pp.2882-2903
• /
• 2022

Image super-resolution (SR) processing is of great value in the fields of digital image processing, intelligent security, film and television production and so on. This paper proposed a densely connected deep learning network based on cascade architecture, which can be used to solve the problem of super-resolution in the field of image quality enhancement. We proposed a more efficient residual scaling dense block (RSDB) and the multi-channel cascade architecture to realize more efficient feature reuse. Also we proposed a hybrid loss function based on L₁ error and L_∞ error to achieve better L_∞ error performance. The experimental results show that the overall performance of the network is effectively improved on cascade architecture and residual scaling. Compared with the residual dense net (RDN), the PSNR / SSIM of the new method is improved by 2.24% / 1.44% respectively, and the L_∞ performance is improved by 3.64%. It shows that the cascade connection and residual scaling method can effectively realize feature reuse, improving the residual convergence speed and learning efficiency of our network. The L_∞ performance is improved by 11.09% with only a minimal loses of 1.14% / 0.60% on PSNR / SSIM performance after adopting the new loss function. That is to say, the L_∞ performance can be improved greatly on the new loss function with a minor loss of PSNR / SSIM performance, which is of great value in L_∞ error sensitive tasks.

• Journal of Multimedia Information System
• /
• v.8 no.4
• /
• pp.203-210
• /
• 2021

Deep leaning convolutional neural networks (CNN) have successfully been applied to image super-resolution (SR). Despite their great performances, SR techniques tend to focus on a certain upscale factor when training a particular model. Algorithms for single model multi-scale networks can easily be constructed if images are upscaled prior to input, but sub-pixel convolution upsampling works differently for each scale factor. Recent SR methods employ multi-scale and multi-path learning as a solution. However, this causes unshared parameters and unbalanced parameter distribution across various scale factors. We present a multi-scale single-path upsample module as a solution by exploiting the advantages of sub-pixel convolution and interpolation algorithms. The proposed model employs sub-pixel convolution for the highest scale factor among the learning upscale factors, and then utilize 1-dimension interpolation, compressing the learned features on the channel axis to match the desired output image size. Experiments are performed for the single-path upsample module, and compared to the multi-path upsample module. Based on the experimental results, the proposed algorithm reduces the upsample module's parameters by 24% and presents slightly to better performance compared to the previous algorithm.

• Proceedings of the KSME Conference
• /
• 2003.04a
• /
• pp.531-536
• /
• 2003

Equal channel angular pressing (ECAP) has been employed to produce materials with ultra-fine grains that have high strength and high corrosion resistance properties. In order to obtain super plastic deformation during ECAP, multipass angular pressing is frequently employed. In this paper, three-dimensional finite element analyses have been performed to investigate the deformation behavior of pure-Zr specimen and the effects of process parameters for equal channel multi-angular pressing (ECMAP) process. The results have been compared with some experimental results

• Journal of Communications and Networks
• /
• v.16 no.2
• /
• pp.238-244
• /
• 2014

A key technical challenge in TV white spaces is the efficient spectrum usage without interfering with primary users. This paper considers available spectrum discovery scheme using in-band sensing signal to support super Wi-Fi services effectively. The proposed scheme in this paper adopts non-contiguous orthogonal frequency-division multiplexing (NC-OFDM) to utilize the fragmented channel in TV white space due to microphones while this channel cannot be used in IEEE 802.11af. The proposed solution is a novel available spectrum discovery scheme by exploiting the advantages of a sensing signaling. The proposed method achieves considerable improvement in throughput and delay time. The proposed method can use more subcarriers for transmission by applying NC-OFDM in contrast with the conventional IEEE 802.11af standard. Moreover, the increased number of wireless microphones (WMs) hardly affects the throughput of the proposed method because our proposal only excludes some subcarriers used by WMs. Additionally, the proposed method can cut discovery time down to under 10 ms because it can find available channels in real time by exchanging sensing signal without interference to the WM.

• Journal of the Institute of Convergence Signal Processing
• /
• v.7 no.2
• /
• pp.81-86
• /
• 2006

In this paper, a new process detection circuit is proposed. The proposed process detection circuit compares a long channel MOS transistor (L > 0.4um) to a short channel MOS transistor which uses lowest feature size of the process. The circuit generates the differential current proportional to the deviation of carrier mobilities according to the process variation. This method keep the two transistor's drain voltage same by implementing the feedback using a high gain OPAMP. This paper also shows the new design of the simple high gam self-biased rail-to-rail OPAMP using a proposed self-biased super MOS composite circuit. The gain of designed OPAMP is measured over 100dB with $0.2{\sim}1.6V$ wide range CMR in single stage. Finally, the proposed process detection circuit is applied to a differential VCO and the VCO showed that the proposed process detection circuit compensates the process corners successfully and ensures the wide rage operation.

• Journal of Electrical Engineering and information Science
• /
• v.3 no.2
• /
• pp.261-266
• /
• 1998

This paper describes the fabrication and characterization of dual-polysilicon gated surface channel 0.1$\mu\textrm{m}$ CMOSFETs using BF2 and arsenic as channel dopants. We have used and LDD structure and 40${\AA}$ gate oxide as an insulator. To suppress short channel effects down to 0.1$\mu\textrm{m}$ channel length, shallow source/drain extensions implemented by low energy implantation and SSR(Super Steep Retrograde) channel structure were used. The threshold voltages of fabricated CMOSFETs are 0.6V. The maximum transconductance of nMOSFET is 315${\mu}$S/$\mu\textrm{m}$, and that of pMOSFET is 156 ${\mu}$S/$\mu\textrm{m}$. The drain saturation current of 418 ${\mu}$A/$\mu\textrm{m}$, 187${\mu}$A/$\mu\textrm{m}$ are obtained. Subthreshold swing is 85mV/dec and 88mV/dec, respectively. DIBL(Drain Induced Barrier Lowering) is below 100mV. In the device with 2000${\AA}$ thick gate polysilicon, depletion in polysilicon near the gate oxide results in an increase of equivalent gate oxide thickness and degradation of device characteristics. The gate delay time is measured to be 336psec at operation voltage of 2V.