• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.1100-1118
• /
• 2021

In a wide range of ML applications, the training data contains privacy-sensitive information that should be kept secure. Training the ML systems by privacy-sensitive data makes the ML model inherent to the data. As the structure of the model has been fine-tuned by training data, the model can be abused for accessing the data by the estimation in a reverse process called model inversion attack (MIA). Although, MIA has been applied to shallow neural network models of recognizers in literature and its threat in privacy violation has been approved, in the case of a deep learning (DL) model, its efficiency was under question. It was due to the complexity of a DL model structure, big number of DL model parameters, the huge size of training data, big number of registered users to a DL model and thereof big number of class labels. This research work first analyses the possibility of MIA on a deep learning model of a recognition system, namely a face recognizer. Second, despite the conventional MIA under the white box scenario of having partial access to the users' non-sensitive information in addition to the model structure, the MIA is implemented on a deep face recognition system by just having the model structure and parameters but not any user information. In this aspect, it is under a semi-white box scenario or in other words a gray-box scenario. The experimental results in targeting five registered users of a CNN-based face recognition system approve the possibility of regeneration of users' face images even for a deep model by MIA under a gray box scenario. Although, for some images the evaluation recognition score is low and the generated images are not easily recognizable, but for some other images the score is high and facial features of the targeted identities are observable. The objective and subjective evaluations demonstrate that privacy cyber-attack by MIA on a deep recognition system not only is feasible but also is a serious threat with increasing alert state in the future as there is considerable potential for integration more advanced ML techniques to MIA.

• Korean Institute of Interior Design Journal
• /
• v.20 no.3
• /
• pp.98-109
• /
• 2011

To interpret a change of discourse can be a method to understand architectural space in progress. With this idea, features of modern age which motivated sense of the contemporary were considered in this study and subsequently characteristics of contemporary space differentiated from the modern were researched. First of all, features of subject which provided a base of modern thoughts were contemplated. The word 'modern' is used in wide and various terms but basically its core conception consists of reason and universal rationality. The subject of the modem age has vision-centric features just like an ideal representation principle of perspective. Given the fact, it was confirmed that a position to become a right subject, that is, a position controlled by reason existed and that it was to guarantee subject a truth. However, the contemporary subject keeps changing with a purpose of escaping from modern characteristics. It presents a tendency to escape from rationalism of the modern age and Platonism of the ancient Greece which established a basis of western ideology. The subject-centered ideas came to focus on the structure and relationship firmed fundamentally in deep inside of subject. The contemporary subject which escaped from the stiffen ideas bears a meaning through events taking place on immanence surface and serialization. Also, the contemporary architectural space is considered to go abreast with the change and trend. In conclusion, this study proved that features of event-oriented architectural space based on the changing contemporary subject appear as process-based space, user-participated space and individual-cognition space and the like.

• Korean Journal of Health Education and Promotion
• /
• v.35 no.5
• /
• pp.17-24
• /
• 2018

Objectives: After many national physical activity guidelines have established, recent changes are deep and rapid. So the major features and implication to reverse worsening Korean physical activity indicator is desprate but related knowledge and informations are few. So review of recent features and implications of new physical activity guidelines have made. Methods: National physical activity guidelines of advanced countries were searched through snowballing methods. Major features were described according to the nation. Implication were drew through discussion for Korean realitiy. Results: New Australian physical activity and sedentary behaviour guideline explicitly included sedentary behaviour. The age in the guideline expanded to early years. Canada also presented 24-hour movement guidelines to early years. The second generation of the physical activity guidelines reflects the extensive amount of new knowledge. New aspects include discussions of additional health benefits related to brain health, additional cancer sites, and fall-related injuries; immediate and longer term benefits for how people feel, function, and sleep; further benefits among older adults and people with additional chronic conditions; risks of sedentary behavior and their relationship with physical activity; elimination of the requirement for physical activity benefits to occur in bouts of at least 10 minutes; and tested strategies that can be used to get the population more active. Conclusions: The most important message from the new guidelines is that the greatest health benefits accrue by moving from no, to even small amounts of, physical activity. Multiple studies demonstrate that the steepest reduction in disease risk occurs at the lowest levels of physical activity. People need to understand that even small amounts of physical activity are beneficial and that reductions in the risk of disease and disability occur by simply getting moving. So various evidence based proven strstegies are needed in Korea including workforce training.

• Journal of Broadcast Engineering
• /
• v.24 no.5
• /
• pp.791-801
• /
• 2019

The serial arc is one of factors causing electrical fires. Over past decades, various researches have been carried out to detect arc occurrences. Even though frequency analysis, wavelet and statistical features have been used, arc detection performance is degraded due to diverse arc waveforms. Therefore, there is a need to develop a method that could increase the feature dimension, thereby improving the detection performance. In this paper, we use variational mode decomposition (VMD) to obtain multiple decomposed signals and then extract statistical features from them. The features from VMD outperform those from no-VMD in terms of detection performance. Further, artificial neural network is employed as an arc classifier. Experiments validated that the use of VMD improves the classification accuracy by up to 4 percent, based on 14,000 training data.

• International Journal of Computer Science & Network Security
• /
• v.21 no.8
• /
• pp.288-296
• /
• 2021

Automated face recognition in a runtime environment is gaining more and more important in the fields of surveillance and urban security. This is a difficult task keeping in mind the constantly volatile image landscape with varying features and attributes. For a system to be beneficial in industrial settings, it is pertinent that its efficiency isn't compromised when running on roads, intersections, and busy streets. However, recognition in such uncontrolled circumstances is a major problem in real-life applications. In this paper, the main problem of face recognition in which full face is not visible (Occlusion). This is a common occurrence as any person can change his features by wearing a scarf, sunglass or by merely growing a mustache or beard. Such types of discrepancies in facial appearance are frequently stumbled upon in an uncontrolled circumstance and possibly will be a reason to the security systems which are based upon face recognition. These types of variations are very common in a real-life environment. It has been analyzed that it has been studied less in literature but now researchers have a major focus on this type of variation. Existing state-of-the-art techniques suffer from several limitations. Most significant amongst them are low level of usability and poor response time in case of any calamity. In this paper, an improved face recognition system is developed to solve the problem of occlusion known as FRS-OCC. To build the FRS-OCC system, the color and texture features are used and then an incremental learning algorithm (Learn++) to select more informative features. Afterward, the trained stack-based autoencoder (SAE) deep learning algorithm is used to recognize a human face. Overall, the FRS-OCC system is used to introduce such algorithms which enhance the response time to guarantee a benchmark quality of service in any situation. To test and evaluate the performance of the proposed FRS-OCC system, the AR face dataset is utilized. On average, the FRS-OCC system is outperformed and achieved SE of 98.82%, SP of 98.49%, AC of 98.76% and AUC of 0.9995 compared to other state-of-the-art methods. The obtained results indicate that the FRS-OCC system can be used in any surveillance application.

• Smart Media Journal
• /
• v.10 no.1
• /
• pp.63-69
• /
• 2021

When a person sees a sentence and understands the sentence, the person understands the sentence by reminiscent of the main word in the sentence as an image. Text-to-image is what allows computers to do this associative process. The previous deep learning-based text-to-image model extracts text features using Convolutional Neural Network (CNN)-Long Short Term Memory (LSTM) and bi-directional LSTM, and generates an image by inputting it to the GAN. The previous text-to-image model uses basic embedding in text feature extraction, and it takes a long time to train because images are generated using several modules. Therefore, in this research, we propose a method of extracting features by using the attention mechanism, which has improved performance in the natural language processing field, for sentence embedding, and generating an image by inputting the extracted features into the GAN. As a result of the experiment, the inception score was higher than that of the model used in the previous study, and when judged with the naked eye, an image that expresses the features well in the input sentence was created. In addition, even when a long sentence is input, an image that expresses the sentence well was created.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.117-127
• /
• 2022

Data anomalies seriously threaten the reliability of the bridge structural health monitoring system and may trigger system misjudgment. To overcome the above problem, an efficient and accurate data anomaly detection method is desiderated. Traditional anomaly detection methods extract various abnormal features as the key indicators to identify data anomalies. Then set thresholds artificially for various features to identify specific anomalies, which is the artificial experience method. However, limited by the poor generalization ability among sensors, this method often leads to high labor costs. Another approach to anomaly detection is a data-driven approach based on machine learning methods. Among these, the bidirectional long-short memory neural network (BiLSTM), as an effective classification method, excels at finding complex relationships in multivariate time series data. However, training unprocessed original signals often leads to low computation efficiency and poor convergence, for lacking appropriate feature selection. Therefore, this article combines the advantages of the two methods by proposing a deep learning method with manual experience statistical features fed into it. Experimental comparative studies illustrate that the BiLSTM model with appropriate feature input has an accuracy rate of over 87-94%. Meanwhile, this paper provides basic principles of data cleaning and discusses the typical features of various anomalies. Furthermore, the optimization strategies of the feature space selection based on artificial experience are also highlighted.

• Analyses & Alternatives
• /
• v.6 no.1
• /
• pp.259-282
• /
• 2022

This article examines features and implications of 'new northeast phenomenon' in China's new normal period. Different from previous studies with economy as a single factor, this paper regards the northeast phenomenon as not an single economic phenomenon but a compound socio economic phenomenon that economic, demographic and financial issues are linked together. This study finds that since 2014 decline of economic growth, deepening of population decline and brain drain, accelerated aging, the increase in fiscal deficit and a surge in social security spending, these phenomena occur simultaneously and influence each other, forming a vicious cycle in northeast China, and also finds that the difficulty of 'new northeast phenomenon' lies in this compound features. If so, what are the implications of 'new northeast phenomenon's' these features for China as a whole? This study proposes two points. First, based on the recent changes in some relevant situations in China, northeast region can be regarded as a microcosm of China, northeast phenomenon is likely to spread to other regions of China in the near future, it will become a common phenomenon all over China. Second, the emergence of the 'new northeast phenomenon' in the new normal period requires deep reflection and rethinking about the fundamental effect of the regional development strategies, such as 'The Development of the Western Region', 'The Rise of Central China', 'The revitalization of the Northeast', implemented since the reform and development. The 'new northeast phenomenon' has become one of the most urgent problems to be solved by the Chinese government, if the solution is successful, it can be a very useful direction for reconstructing regional development strategies in contemporary China.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.3
• /
• pp.701-720
• /
• 2023

Aerial objects are more challenging to segment than normal objects, which are usually smaller and have less textural detail. In the process of segmentation, target objects are easily omitted and misdetected, which is problematic. To alleviate these issues, we propose local aggregation feature pyramid networks (LAFPNs) and pyramid integrated context modules (PICMs) for aerial object segmentation. First, using an LAFPN, while strengthening the deep features, the extent to which low-level features interfere with high-level features is reduced, and numerous dense and small aerial targets are prevented from being mistakenly detected as a whole. Second, the PICM uses global information to guide local features, which enhances the network's comprehensive understanding of an entire image and reduces the missed detection of small aerial objects due to insufficient texture information. We evaluate our network with the MS COCO dataset using three categories: airplanes, birds, and kites. Compared with Mask R-CNN, our network achieves performance improvements of 1.7%, 4.9%, and 7.7% in terms of the AP metrics for the three categories. Without pretraining or any postprocessing, the segmentation performance of our network for aerial objects is superior to that of several recent methods based on classic algorithms.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.12
• /
• pp.1786-1793
• /
• 2022

Cyber-attacks such as smishing and hacking mail exploiting COVID-19, political and social issues, have recently been continuous. Machine learning and deep learning technology research are conducted to prevent any damage due to cyber-attacks inducing malicious links to breach personal data. It has been concluded as a lack of basis to judge the attacks to be malicious in previous studies since the features of data set were excessively simple. In this paper, nine main features of three types, "URL Days", "URL Word", and "URL Abnormal", were proposed in addition to lexical features of URL which have been reflected in previous research. F1-Score and accuracy index were measured through four different types of machine learning algorithms. An improvement of 0.9% in a result and the highest value, 98.5%, were examined in F1-Score and accuracy through comparatively analyzing an existing research. These outcomes proved the main features contribute to elevating the values in both accuracy and performance.