• Journal of Korea Multimedia Society
• /
• v.25 no.10
• /
• pp.1375-1385
• /
• 2022

In this paper, we propose a semi-supervised domain adaptation solution to deal with practical face recognition (FR) scenarios where a single face image for each target identity (to be recognized) is only available in the training phase. Main goal of the proposed method is to reduce the discrepancy between the target and the source domain face images, which ultimately improves FR performances. The proposed method is based on the Domain Adatation network (DAN) using an MMD loss function to reduce the discrepancy between domains. In order to train more effectively, we develop a novel loss function learning strategy in which MMD loss and cross-entropy loss functions are adopted by using different weights according to the progress of each epoch during the learning. The proposed weight adoptation focuses on the training of the source domain in the initial learning phase to learn facial feature information such as eyes, nose, and mouth. After the initial learning is completed, the resulting feature information is used to training a deep network using the target domain images. To evaluate the effectiveness of the proposed method, FR performances were evaluated with pretrained model trained only with CASIA-webface (source images) and fine-tuned model trained only with FERET's gallery (target images) under the same FR scenarios. The experimental results showed that the proposed semi-supervised domain adaptation can be improved by 24.78% compared to the pre-trained model and 28.42% compared to the fine-tuned model. In addition, the proposed method outperformed other state-of-the-arts domain adaptation approaches by 9.41%.

• Journal of Information Processing Systems
• /
• v.15 no.6
• /
• pp.1472-1488
• /
• 2019

Critical elements in the China's Lunar Exploration reside in that the lunar rover travels over the surrounding undetermined environment and it conducts scientific exploration under the ground control via teleoperation system. Such an interplanetary transportation mission teleoperation system belongs to the ground application system in deep space mission, which performs terrain reconstruction, visual positioning, path planning, and rover motion control by receiving telemetry data. It plays a vital role in the whole lunar exploration operation and its so-called trustworthy evidence must be assessed before and during its implementation. Taking ISO standards and China's national military standards as trustworthy evidence source, the net assessment model and net assessment method of teleoperation system are established in this paper. The multi-dimensional net assessment model covering the life cycle of software is defined by extracting the trustworthy evidences from trustworthy evidence source. The qualitative decisions are converted to quantitative weights through the net assessment method (NAM) combined with fuzzy analytic hierarchy process (FAHP) and entropy weight method (EWM) to determine the weight of the evidence elements in the net assessment model. The paper employs the teleoperation system for interplanetary transportation as a case study. The experimental result drawn shows the validity and rationality of net assessment model and method. In the final part of this paper, the untrustworthy elements of the teleoperation system are discovered and an improvement scheme is established upon the "net result". The work completed in this paper has been applied in the development of the teleoperation system of China's Chang'e-3 (CE-3) "Jade Rabbit-1" and Chang'e-4 (CE-4) "Jade Rabbit-2" rover successfully. Besides, it will be implemented in China's Chang'e-5 (CE-5) mission in 2019. What's more, it will be promoted in the Mars exploration mission in 2020. Therefore it is valuable to the development process improvement of aerospace information system.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.2
• /
• pp.36-43
• /
• 2009

In this paper, we propose a Cooperative Routing using ARQ-based Transmission (CRARQ) in Mobile Ad-hoc Wireless Sensor Networks (MAWSN). The goals of the CRARQ are to improve both the efficiency of transmission and the reliability of transmission together. The main features and contributions of the proposed CRARQ for supporting these goals are as follows. First, CRARQ uses a reactive concept to construct a route from a source to a destination. Second, CRARQ chooses the most stable relay at each hop along the route by mobility-based strategy. Third, cooperative transmission using the ARQ technique which can improve reliability is used at each hop to send data packets from source to destination. Finally, we present a theoretical analysis model for the proposed CRARQ. The performance evaluation of our protocol is implemented via simulation using Optimized Network Engineering Tool (OPNET) and analysis. The results of both simulation and analysis show that CRARQ achieves the improved packet delivery ratio (PDR) and outage probability.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.5
• /
• pp.677-690
• /
• 1995

In this paper, we propose the high efficiency data compression capable of error control using block-sorting, move to front(MTF) and arithmetic code with variable length in to fixed out. First, the substring with is parsed into length N is shifted one by one symbol. The cyclic shifted rows are sorted in lexicographical order. Second, the MTF technique is applied to get the reference of locality in the sorted substring. Then the preprocessed sequence is coded using VF(variable to fixed) arithmetic code which can be limited the error propagation in one codeword. The key point is how to split the fixed length codeword in proportion to symbol probabilities in VF arithmetic code. We develop the new VF arithmetic coding that split completely the codeword set for arbitrary source alphabet. In addition to, an extended representation for symbol probability is designed by using recursive Gray conversion. The performance of proposed method is compared with other well-known source coding methods with respect to entropy, compression ratio and coding times.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.4
• /
• pp.1134-1147
• /
• 1995

POD (proper orthogonal decomposition) is applied to turbulent leading-edge separation bubble to extract coherent structures. A two-dimensional leading-edge separation bubble is simulated by discrete-vortex method, where a time-dependent source forcing is incorporated. Based on the wealth of numerical data, POD is applied in a range of the forcing amplitude ( $A_{o}$ = 0, 0.5, 1.0 and 1.5) and forcing frequency (0 .leq. $f_{F}$H/ $U_{\infty}$.leq. 0.3). It is demonstrated that the structures of POD have noticeable changes with local forcings. In an effort to investigate the mechanism of decreasing reattachment length, dynamic behaviors of the expansion coefficients and contributions of the eigenfunctions of POD are scrutinized. As the forcing amplitude increases, the large-scale vortex structures are formed near the forcing amplitude increases, the large-scale vortex structures are formed near the separation point and the flow structures become more organized and more regular, accompanying with the reduction of reattachment length. By further inverstigation of POD global entropy, it is seen that the reattachment length is closely linked to the degree of organization of the flow structures.es.s.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.11
• /
• pp.78-87
• /
• 1994

In this paper, we have addressed a new approach to encode the DCT coefficients efficiently by exploiting the structural differences among the probability distribution functions of them. We have proved analytically that if signals are separated according to their source models, the entropy of the separated signals decreases below that of the unseparated signals. And simulation results show that the average codelength decrease when each (runlength, level) symbol is encoded adaptively according to the index associated with the symbol.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.18 no.12
• /
• pp.3287-3296
• /
• 1994

For the numerical computation of three-dimensional compressible flow field within a blade row in a centrifugal compressor, a quasi 3-dimensional solver which combines a reversible B-B flow and an irreversible H-S flow using finite element methods was developed. In a reversible B-B flow, the governing coordinates are modified in order to be applied to any type of turbomachinery, and two kinds of stream functions are introduced in order to make the Kutta condition exactly satisfied. In an irreversible H-S flow, the changes of entropy in the irreversible governing equations are determined not by empirical source but by the theoretical treatment of dissipation forces. The dissipation forces are obtained from the distribution of shear stresses in the flow passage which are given from the wall shear stresses using the exponential functions. A more accurate quasi-3-dimensional solver is established where the effect of body forces is involved in the non-axisymmetric H-S flow. Some numerical results obtained from authors' previous studies for axial flow machines assure that the present method is able to predict well as long as the flow is subsonic and not under strong viscous effect.

• Journal of Korea Multimedia Society
• /
• v.20 no.12
• /
• pp.1992-1999
• /
• 2017

This study proposes a decorative typeface which is produced through the concept of trigonometric functions in an open-source programming language known as Processing. First, the theoretical background of Processing and trigonometric functions as well as previous research in this area are analyzed. Second, basic modules of 'V', 'I', 'O', and 'M' were created for use as the final alphabet typeface with the concept of a trigonometric function. Third, a decorative parabolic curve that encircles the base module was created. Finally, the modules created on Processing were edited in Adobe Illustrator to create a typeface set with characters from A to Z. Various artworks using Programming can produce an infinite number of different versions by modifying only some of the variables and codes, and this method can include multimedia features such as text, images, videos, interactive art and various forms of content and media. Therefore, with regard to expression, the possibilities are endless. In this study, I attempt to expand the field of visual culture using programming and computational methodologies. In contrast to the digital typeface production method, which relies on existing graphic tools, this study is meaningful because it expands the range of use of decorative typefaces.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.1
• /
• pp.111-119
• /
• 1994

In this paper we propose a reversible image coding methods with progressive build-up function by bit-plane of multi-lavel image. Firstly, the differencial conversion is applied to reduce the entropy of source images. Then the bit-plane sequences of converted images are coded by means of the newly designed VFRL(Variable to Fixed Run-Llength) code and the RDHC(Run-length Dynamic Huffman Coding). We aim to the simple construction to reduce the complexity. The computer simulated results show that the proposed methods are very effective to the multi-level digital images. For "GIRL" and "COUPLE" of 1EEE monochromatic standard images, the compressibility are superior to the results obtained by the well-known universal codes.own universal codes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.4
• /
• pp.1442-1462
• /
• 2014

Packet classification is a key technology of the Internet for routers to classify the arriving packets into different flows according to the predefined rulesets. Previous packet classification algorithms have mainly focused on search speed and memory usage, while overlooking update performance. In this paper, we propose PreCuts, which can drastically improve the update speed. According to the characteristics of IP field, we implement three heuristics to build a 3-layer decision tree. In the first layer, we group the rules with the same highest byte of source and destination IP addresses. For the second layer, we cluster the rules which share the same IP prefix length. Finally, we use the heuristic of information entropy-based bit partition to choose some specific bits of IP prefix to split the ruleset into subsets. The heuristics of PreCuts will not introduce rule duplication and incremental update will not reduce the time and space performance. Using ClassBench, it is shown that compared with BRPS and EffiCuts, the proposed algorithm not only improves the time and space performance, but also greatly increases the update speed.