• Polymer(Korea)
• /
• v.38 no.6
• /
• pp.752-759
• /
• 2014

Crosslinked polyimides (PIs) were synthesized by reacting 4,4'-(hexafluoroisopropylidene)-diphthalic anhydride (6FDA) and 2,2'-bis(trifluoromethyl)benzidine (TFDB) with various ratios of the cross-linkable, end-capping agent cis-1,2,3,6-tetrahydrophthalic anhydride (CDBA) via ring-opening metathesis polymerization. Residual stress behaviors were investigated in-situ during thermal imidization of the crosslinked PI precursors using a thin film stress analyzer (TFSA) by wafer bending method. The thermal properties were investigated via differential scanning calorimetry (DSC), thermomechanical analysis (TMA), and thermogravimetric analysis (TGA). The optical properties were measured by ultraviolet-visible spectrophotometer (UV-vis) and spectrophotometry. All properties were interpreted with respect to their morphology of crosslinked networks. With increasing the amounts of the end-capping agent, the residual stress decreased from 27.9 to -1.3 MPa, exhibited ultra-low stress and high thermal properties. The minimized residual stress and enhanced thermal properties of the crosslinked PI makes them potential candidates for versatile high-density multi-layer structure applications.

• Journal of Internet Computing and Services
• /
• v.21 no.5
• /
• pp.21-29
• /
• 2020

Accurate estimation of human pose relies on backbone method in which its role is to extract feature map. Up to dated, the method of backbone feature extraction is conducted by the plain convolutional neural networks named by CNN and the residual neural networks named by Resnet, both of which have various architectures and performances. The CNN family network such as VGG which is well-known as a multiple stacked hidden layers architecture of deep learning methods, is base and simple while Resnet which is a bottleneck layers architecture yields fewer parameters and outperform. They have achieved inspired results as a backbone network in human pose estimation. However, they were used then followed by different pose estimation networks named by pose parsing module. Therefore, in this paper, we present a comparison between the plain CNN family network (VGG) and bottleneck network (Resnet) as a backbone method in the same pose parsing module. We investigate their performances such as number of parameters, loss score, precision and recall. We experiment them in the bottom-up method of human pose estimation system by adapted the pose parsing module of openpose. Our experimental results show that the backbone method using VGG network outperforms the Resent network with fewer parameter, lower loss score and higher accuracy of precision and recall.

• Korean Journal of Remote Sensing
• /
• v.36 no.5_2
• /
• pp.961-973
• /
• 2020

In this manuscript, a new pansharpening model based on Convolutional Neural Network (CNN) was developed. Dilated convolution, which is one of the representative convolution technologies in CNN, was applied to the model by making it deep and complex to improve the performance of the deep learning architecture. Based on the dilated convolution, the residual network is used to enhance the efficiency of training process. In addition, we consider the spatial correlation coefficient in the loss function with traditional L1 norm. We experimented with Dilated Residual Networks (DRNet), which is applied to the structure using only a panchromatic (PAN) image and using both a PAN and multispectral (MS) image. In the experiments using KOMPSAT-3A, DRNet using both a PAN and MS image tended to overfit the spectral characteristics, and DRNet using only a PAN image showed a spatial resolution improvement over existing CNN-based models.

• Journal of Korea Water Resources Association
• /
• v.55 no.spc1
• /
• pp.1197-1210
• /
• 2022

Recently, users' complaints on drinking water quality are increasing according to emerging interest in the drinking water service issues such as pipe aging and various water quality accidents. In the case of drinking water quality complaints, not only the water pollution but also the inconvenience on the chlorine residual for disinfection are included, thus various efforts, such as rechlorination treatment, are being attempted in order to keep the chlorine concentration supplied evenly. In this research, for a more accurate water quality simulation of water distribution network, the water quality reaction coefficients were estimated, and an optimization method of chlorination/ rechlorination scheduling was proposed consideirng satisfaction of water quality standards and chlorine residual equalization. The proposed method was applied to a large-scale real water network, and various chlorination schemes were comparatively analyzed through the grid search algorithm and optimized based on the suitability and uniformity of supplied chlorine residual concentration.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.9
• /
• pp.55-63
• /
• 2014

Up to date, numerous studies on wireless sensor networks have been performed to overcome the Energy-Constraint of the sensor system. Existing schemes for estimating the residual energy have considered only voltage of sensor system. However battery performance in the real is affected by temperature and load. In this paper we introduce more accurate scheme, for the use in wireless sensor node, based on the interpolation of lookup tables which allow for temperature and load characteristics, as well as battery voltage.

• ETRI Journal
• /
• v.39 no.3
• /
• pp.398-405
• /
• 2017

In this paper, an artificial stereo extension method that creates stereophonic sound from a mono sound source is proposed. The proposed method first trains deep neural networks (DNNs) that model the nonlinear relationship between the dominant and residual signals of the stereo channel. In the training stage, the band-wise log spectral magnitude and unwrapped phase of both the dominant and residual signals are utilized to model the nonlinearities of each sub-band through deep architecture. From that point, stereo extension is conducted by estimating the residual signal that corresponds to the input mono channel signal with the trained DNN model in a sub-band domain. The performance of the proposed method was evaluated using a log spectral distortion (LSD) measure and multiple stimuli with a hidden reference and anchor (MUSHRA) test. The results showed that the proposed method provided a lower LSD and higher MUSHRA score than conventional methods that use hidden Markov models and DNN with full-band processing.

• Journal of Communications and Networks
• /
• v.18 no.2
• /
• pp.210-217
• /
• 2016

In this paper, we propose a low complexity multiple-input multiple-output (MIMO) detection algorithm with adaptive interference mitigation in downlink multiuser MIMO (DL MU-MIMO) systems with quantization error of the channel state information (CSI) feedback. In DL MU-MIMO systems using the imperfect precoding matrix caused by quantization error of the CSI feedback, the station receives the desired signal as well as the residual interference signal. Therefore, a complexMIMO detection algorithm with interference mitigation is required for mitigating the residual interference. To reduce the computational complexity, we propose a MIMO detection algorithm with adaptive interference mitigation. The proposed algorithm adaptively mitigates the residual interference by using the maximum likelihood detection (MLD) error criterion (MEC). We derive a theoretical MEC by using the MLD error condition and a practical MEC by approximating the theoretical MEC. In conclusion, the proposed algorithm adaptively performs interference mitigation when satisfying the practical MEC. Simulation results show that the proposed algorithm reduces the computational complexity and has the same performance, compared to the generalized sphere decoder, which always performs interference mitigation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.10
• /
• pp.4957-4976
• /
• 2016

This paper investigates the error performance of the amplify-and-forward (AF) relaying systems in the context of full-duplex (FD) communication. In addition to the inherent self-interference (SI) due to simultaneous transmission and reception, coexistent FD terminals may cause crosstalk. In this paper, we utilize the information exchange via the crosstalk channel to construct a particular distributed space-time code (DSTC). The residual SI is also considered. Closed-form pairwise error probability (PEP) is first derived. Then we obtain the upper bound of PEP in high transmit power region to provide more insights of diversity and coding gain. The proposed DSTC scheme can attain full cooperative diversity if the variance of SI is not a function of the transmit power. The coding gain can be improved by lengthening the frame and proper power control. Feasibility and efficiency of the proposed DSTC are verified in numerical simulations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.3
• /
• pp.757-770
• /
• 2022

Palmprint recognition has drawn increasingly attentions in the past decade due to its uniqueness and reliability. Traditional palmprint recognition methods usually use high-resolution images as the identification basis so that they can achieve relatively high precision. However, high-resolution images mean more computation cost in the recognition process, which usually cannot be guaranteed in mobile computing. Therefore, this paper proposes an improved low-resolution palmprint image recognition method based on residual networks. The main contributions include: 1) We introduce a channel attention mechanism to refactor the extracted feature maps, which can pay more attention to the informative feature maps and suppress the useless ones. 2) The ResStage group structure proposed by us divides the original residual block into three stages, and we stabilize the signal characteristics before each stage by means of BN normalization operation to enhance the feature channel. Comparison experiments are conducted on a public dataset provided by the Hong Kong Polytechnic University. Experimental results show that the proposed method achieve a rank-1 accuracy of 98.17% when tested on low-resolution images with the size of 12dpi, which outperforms all the compared methods obviously.

• The Journal of Korea Robotics Society
• /
• v.13 no.1
• /
• pp.1-7
• /
• 2018

The image-to-image translation is one of the deep learning applications using image data. In this paper, we aim at improving the performance of object transfiguration which transforms a specific object in an image into another specific object. For object transfiguration, it is required to transform only the target object and maintain background images. In the existing results, however, it is observed that other parts in the image are also transformed. In this paper, we have focused on the structure of artificial neural networks that are frequently used in the existing methods and have improved the performance by adding constraints to the exiting structure. We also propose the advanced structure that combines the existing structures to maintain their advantages and complement their drawbacks. The effectiveness of the proposed methods are shown in experimental results.