• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.3
• /
• pp.439-449
• /
• 2024

Recent rapid advancements in voice generation technology have enabled the natural synthesis of voices using text alone. However, this progress has led to an increase in malicious activities, such as voice phishing (voishing), where generated voices are exploited for criminal purposes. Numerous models have been developed to detect the presence of synthesized voices, typically by extracting features from the voice and using these features to determine the likelihood of voice generation.This paper proposes a new model for extracting voice features to address misuse cases arising from generated voices. It utilizes a deep learning-based audio codec model and the pre-trained natural language processing model BERT to extract novel voice features. To assess the suitability of the proposed voice feature extraction model for voice detection, four generated voice detection models were created using the extracted features, and performance evaluations were conducted. For performance comparison, three voice detection models based on Deepfeature proposed in previous studies were evaluated against other models in terms of accuracy and EER. The model proposed in this paper achieved an accuracy of 88.08%and a low EER of 11.79%, outperforming the existing models. These results confirm that the voice feature extraction method introduced in this paper can be an effective tool for distinguishing between generated and real voices.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.4
• /
• pp.1721-1737
• /
• 2020

Human apparent age estimation has become a popular research topic and attracted great attention in recent years due to its wide applications, such as personal security and law enforcement. To achieve the goal of age estimation, a large number of methods have been pro-posed, where the models derived through the cumulative attribute coding achieve promised performance by preserving the neighbor-similarity of ages. However, these methods afore-mentioned ignore the geometric structure of extracted facial features. Indeed, the geometric structure of data greatly affects the accuracy of prediction. To this end, we propose an age estimation algorithm through joint feature selection and manifold learning paradigms, so-called Feature-selected and Geometry-preserved Least Square Regression (FGLSR). Based on this, our proposed method, compared with the others, not only preserves the geometry structures within facial representations, but also selects the discriminative features. Moreover, a deep learning extension based FGLSR is proposed later, namely Feature selected and Geometry preserved Neural Network (FGNN). Finally, related experiments are conducted on Morph2 and FG-Net datasets for FGLSR and on Morph2 datasets for FGNN. Experimental results testify our method achieve the best performances.

• The Bulletin of The Korean Astronomical Society
• /
• v.42 no.1
• /
• pp.50.1-50.1
• /
• 2017

We present our study of the correlation between star formation rate(SFR) and merging activities of nearby galaxies(d<150Mpc). Our study uses 265 UV-selected galaxies which are not classified as AGN. The UV selection is made using the GALEX Atlas of Galaxies (Gil de Paz+07) and the updated UV catalog of nearby galaxies (Bai+15). We use deep R band optical images reaching to $1{\sigma}$ surface brightness detection limit ${\sim}27mag/arcsec^2$ to classify merger features by visual inspection. We also estimated unobscured SFR($SFR_{NUV}$) and obscured SFR($SFR_{W4}$) using Near-UV continuum and 22 micron Mid-IR luminosity respectively as a indicator of star forming activity. The fraction of galaxies with merger features in each SFR bin is obtained to see if how the fraction of galaxies with merging features($F_m$) changes as a function of SFR. As a result, for 203 late type galaxies(LTGs), we found that merger fraction increases from ~8% up to 50% with $SFR_{W4}$, while for 229 LTGs $SFR_{NUV}$ shows relatively consistent fraction(~18%) of merger fraction. For early type galaxies(ETGs), we could also find no significant correlation between $F_m$ and SFR(both $SFR_{NUV}$ and $SFR_{W4}$). This result suggests that a main driver of star forming activity of UV bright galaxies, especially for obscured late types, is mergers.

• Journal of the Korean Society of Costume
• /
• v.62 no.6
• /
• pp.200-217
• /
• 2012

The purpose of this research is to examine the specificity of grids to define the characteristics of clothes styles in the Joseon Dynasty period. The significance of examining of the specificity of grids is to find out arbitrary types of the features of grids involved in structuring the Jeogori in the Joseon Dynasty period one by one. The Visual Linguistic Theory was introduced as a methodological tool to exquisitely analyze the characteristics of grids in deep structures of Jeogori in the Joseon Dynasty period. This theory strives to examine sample distribution, the distribution of samples by quality and the distribution of the types of ploidy features. Through the examination, the results are as follows. The grid systems of the Jeogori consisted of diverse proportion systems reaching 86 cases, that is, sequence systems composed of multi-functional, multi-combined bodies. Most ornamental grids had feature angles distributed in a range of $2-20^{\circ}$ that showed a common preference for low sloped diagonal lines or small curvature. Although the preference for certain feature angles were prominent, the feature angles that were used were generally distributed evenly among diverse feature angles to show the characteristics of separation. Therefore, Jeogori makers in the Joseon Dynasty period can be considered as having experimented with many proportion systems to show their aesthetics. In conclusion, based on the results of the examination of feature distributions and related methods to allocate ploidy features, O-type accounted for 66% and thus it was identified that the Jeogori was characterized by O-type. Therefore, it was identified that the characteristic of the Jeogori in the Joseon Dynasty period consisted of O-type fractal structures which are formative structures unique to our nation.

• Clinical and Experimental Pediatrics
• /
• v.59 no.sup1
• /
• pp.14-18
• /
• 2016

Pediatric epilepsy can be caused by various conditions, including specific syndromes. 1p36 deletion syndrome is reported in 1 in 5,000-10,000 newborns, and its characteristic clinical features include developmental delay, mental retardation, hypotonia, congenital heart defects, seizure, and facial dysmorphism. However, detection of the terminal deletion in chromosome 1p by conventional G-banded karyotyping is difficult. Here we present a case of epilepsy with profound developmental delay and characteristic phenotypes. A 7-year-and 6-month-old boy experienced afebrile generalized seizure at the age of 5 years and 3 months. He had recurrent febrile seizures since 12 months of age and showed severe global developmental delay, remarkable hypotonia, short stature, and dysmorphic features such as microcephaly; small, low-set ears; dark, straight eyebrows; deep-set eyes; flat nasal bridge; midface hypoplasia; and a small, pointed chin. Previous diagnostic work-up, including conventional chromosomal analysis, revealed no definite causes. However, array-comparative genomic hybridization analysis revealed 1p36 deletion syndrome with a 9.15-Mb copy loss of the 1p36.33-1p36.22 region, and fluorescence in situ hybridization analysis (FISH) confirmed this diagnosis. This case highlights the need to consider detailed chromosomal study for patients with delayed development and epilepsy. Furthermore, 1p36 deletion syndrome should be considered for patients presenting seizure and moderate-to-severe developmental delay, particularly if the patient exhibits dysmorphic features, short stature, and hypotonia.

• Publications of The Korean Astronomical Society
• /
• v.30 no.2
• /
• pp.463-464
• /
• 2015

It has been suggested that only the most luminous AGNs ($L{\gtrsim}10^{45}erg/s$) are triggered by galaxy mergers, while less luminous AGNs ($L{\sim}10^{43}erg/s$) are driven by other internal processes. The lack of merging features in low luminosity AGN host galaxies has been a primary argument against the idea of merger triggering of low luminosity AGNs. But a merger, especially a rather minor one, might still have played an important role in low luminosity AGNs, as minor merging features at low luminosities are more difficult to identify than major merging features. Using SNUCAM on the 1.5 m telescope at Maidanak observatory, we obtained deep optical images of NGC 7743, a barred spiral galaxy classified as a Seyfert 2 AGN with a low bolometric luminosity of $5{\times}10^{42}erg/s$. Surprisingly, we discovered a merging feature around the galaxy, which indicates past merging activity in the galaxy. This example indicates that the merging fraction of low luminosity AGNs may be much higher than previously thought, hinting at the importance of galaxy mergers even in low luminosity AGNs.

• Journal of Digital Convergence
• /
• v.19 no.5
• /
• pp.307-313
• /
• 2021

In this paper, a lightweight network with fewer parameters compared to the existing object detection method is proposed. In the case of the currently used detection model, the network complexity has been greatly increased to improve accuracy. Therefore, the proposed network uses EfficientNet as a feature extraction network, and the subsequent layers are formed in a pyramid structure to utilize low-level detailed features and high-level semantic features. An attention process was applied between pyramid structures to suppress unnecessary noise for prediction. All computational processes of the network are replaced by depth-wise and point-wise convolutions to minimize the amount of computation. The proposed network was trained and evaluated using the PASCAL VOC dataset. The features fused through the experiment showed robust properties for various objects through a refinement process. Compared with the CNN-based detection model, detection accuracy is improved with a small amount of computation. It is considered necessary to adjust the anchor ratio according to the size of the object as a future study.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.1
• /
• pp.24-29
• /
• 2022

Semantic role labeling task used in various natural language processing fields, such as information extraction and question answering systems, is the task of identifying the arugments for a given sentence and predicate. Predicate used as semantic role labeling input are extracted using lexical analysis results such as POS-tagging, but the problem is that predicate can't extract all linguistic patterns because predicate in korean language has various patterns, depending on the meaning of sentence. In this paper, we propose a korean predicate recognition method using neural network model with pre-trained embedding models and lexical features. The experiments compare the performance on the hyper parameters of models and with or without the use of embedding models and lexical features. As a result, we confirm that the performance of the proposed neural network model was 92.63%.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.2
• /
• pp.201-210
• /
• 2023

Deep learning-based face verificattion model show high performance and are used in many fields, but there is a possibility the user's face image may be leaked in the process of inputting the face image to the model. Althoughde-identification technology exists as a method for minimizing the exposure of face features, there is a problemin that verification performance decreases when the existing technology is applied. In this paper, after combining the face features of other person, a de-identified face image is created through StyleGAN. In addition, we propose a method of optimizingthe combining ratio of features according to the face verification model using HopSkipJumpAttack. We visualize the images generated by the proposed method to check the de-identification performance, and evaluate the ability to maintain the performance of the face verification model through experiments. That is, face verification can be performed using the de-identified image generated through the proposed method, and leakage of face personal information can be prevented.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.37 no.4
• /
• pp.225-232
• /
• 2024

This study focuses on the determination of the fracture mode in composite laminates using deep learning. With the increase in the use of laminated composites in numerous engineering applications, the insurance of their integrity and performance is of paramount importance. However, owing to the complex nature of these materials, the identification of fracture modes is often a tedious and time-consuming task that requires critical domain knowledge. Therefore, to alleviate these issues, this study aims to utilize modern artificial intelligence technology to automate the fractographic analysis of laminated composites. To accomplish this goal, scanning electron microscopy (SEM) images of fractured tensile test specimens are obtained from laminated composites to showcase various fracture modes. These SEM images are then categorized based on numerous fracture modes, including fiber breakage, fiber pull-out, mix-mode fracture, matrix brittle fracture, and matrix ductile fracture. Next, the collective data for all classes are divided into train, test, and validation datasets. Two state-of-the-art, deep learning-based pre-trained models, namely, DenseNet and GoogleNet, are trained to learn the discriminative features for each fracture mode. The DenseNet models shows training and testing accuracies of 94.01% and 75.49%, respectively, whereas those of the GoogleNet model are 84.55% and 54.48%, respectively. The trained deep learning models are then validated on unseen validation datasets. This validation demonstrates that the DenseNet model, owing to its deeper architecture, can extract high-quality features, resulting in 84.44% validation accuracy. This value is 36.84% higher than that of the GoogleNet model. Hence, these results affirm that the DenseNet model is effective in performing fractographic analyses of laminated composites by predicting fracture modes with high precision.