• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.466-472
• /
• 2021

Speaker Change Detection (SCD) refers to finding the moment when the main speaker changes from one person to the next in a speech conversation. In speaker change detection, difficulties arise due to overlapping speakers, inaccuracy in the information labeling, and data imbalance. To solve these problems, TIMIT corpus widely used in speech recognition have been concatenated artificially to obtain a sufficient amount of training data, and the detection of changing speaker has performed after identifying overlapping speakers. In this paper, we propose an speaker change detection system that considers the speaker overlapping. We evaluated and verified the performance using various approaches. As a result, a detection system similar to the X-Vector structure was proposed to remove the speaker overlapping region, while the Bi-LSTM method was selected to model the speaker change system. The experimental results show a relative performance improvement of 4.6 % and 13.8 % respectively, compared to the baseline system. Additionally, we determined that a robust speaker change detection system can be built by conducting related studies based on the experimental results, taking into consideration text and speaker information.

• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1445-1456
• /
• 2018

Environment perception and three-dimensional (3D) reconstruction tasks are used to provide unmanned ground vehicle (UGV) with driving awareness interfaces. The speed of obstacle segmentation and surrounding terrain reconstruction crucially influences decision making in UGVs. To increase the processing speed of environment information analysis, we develop a CPU-GPU hybrid system of automatic environment perception and 3D terrain reconstruction based on the integration of multiple sensors. The system consists of three functional modules, namely, multi-sensor data collection and pre-processing, environment perception, and 3D reconstruction. To integrate individual datasets collected from different sensors, the pre-processing function registers the sensed LiDAR (light detection and ranging) point clouds, video sequences, and motion information into a global terrain model after filtering redundant and noise data according to the redundancy removal principle. In the environment perception module, the registered discrete points are clustered into ground surface and individual objects by using a ground segmentation method and a connected component labeling algorithm. The estimated ground surface and non-ground objects indicate the terrain to be traversed and obstacles in the environment, thus creating driving awareness. The 3D reconstruction module calibrates the projection matrix between the mounted LiDAR and cameras to map the local point clouds onto the captured video images. Texture meshes and color particle models are used to reconstruct the ground surface and objects of the 3D terrain model, respectively. To accelerate the proposed system, we apply the GPU parallel computation method to implement the applied computer graphics and image processing algorithms in parallel.

• Korean Journal of Community Nutrition
• /
• v.24 no.1
• /
• pp.1-17
• /
• 2019

Objectives: This study was performed to compare the dietary life of single- and non single-person households in the Korea National Health and Nutrition Examination Survey (KNHANES). Methods: A nationally representative sample of 20,421 19-64-year-olds who had 24-hour recall data was taken from the 2014-2016 Korea National Health and Nutrition Examination Survey (KNHANES). Single- and non single-persons were compared for nutrient intake, dietary behaviors, food consumption patterns, nutrition education and confirm nutrition label. Results: The dietary intakes of dietary fiber and iron were lower in single-person households than in non single-person households. The lower the level of education and income, the lower the nutrient intake of single-person households. In the case of those aged 19 to 29, the breakfast skipping rate was higher in single-person households than in non single-person households. The higher the education level, the higher the breakfast skipping rate and the eating out frequency in the single-person households. In the food intake survey, the frequency of healthy food intake in single-person households was much lower than that of non single-person households. The confirmation rate of nutrition labeling was lower in single-person households than in non single-person households. Conclusions: This study shows that single-person households have poorer health-nutritional behaviors than multi-person households. Therefore, a nutrition education program based on the data of this study needs to be developed for health promotion of single-person households.

• International Journal of Oral Biology
• /
• v.45 no.4
• /
• pp.190-196
• /
• 2020

Several factors, including genetic and environmental insults, impede protein folding and secretion in the endoplasmic reticulum (ER). Accumulation of unfolded or mis-folded protein in the ER manifests as ER stress. To cope with this morbid condition of the ER, recent data has suggested that the intracellular event of an unfolded protein response plays a critical role in managing the secretory load and maintaining proteostasis in the ER. Tauroursodeoxycholic acid (TUDCA) is a chemical chaperone and hydrophilic bile acid that is known to inhibit apoptosis by attenuating ER stress. Numerous studies have revealed that TUDCA affects hepatic diseases, obesity, and inflammatory illnesses. Recently, molecular regulation of ER stress in tooth development, especially during the secretory stage, has been studied. Therefore, in this study, we examined the developmental role of ER stress regulation in tooth morphogenesis using in vitro organ cultivation methods with a chemical chaperone treatment, TUDCA. Altered cellular events including proliferation, apoptosis, and dentinogenesis were examined using immunostaining and terminal deoxynucleotidyl transferase dUTP nick end labeling assay. In addition, altered localization patterns of the formation of hard tissue matrices related to molecules, including amelogenin and nestin, were examined to assess their morphological changes. Based on our findings, modulating the role of the chemical chaperone TUDCA in tooth morphogenesis, especially through the modulation of cellular proliferation and apoptosis, could be applied as a supporting data for tooth regeneration for future studies.

• The Korea Journal of Herbology
• /
• v.37 no.5
• /
• pp.37-51
• /
• 2022

Objectives : The purpose of this study was to apply informations related to herbal medicines to IDMP (Identification of Medicinal Products), an ISO standards related to medicinal products substances, for systematic collection of data through the integration of informations on distribution, manufacturing, and management of herbal medicines. Methods : By analyzing ISO 11238 and ISO/TS 19844, elements that can be used in the information model of herbal medicine were derived from the identification of medicinal products information model on substances. The labeling specified in the safety and quality control regulations for herbal medicines was mapped to the IDMP information model, and ginseng was applied as an example. Results : Herbal medicine corresponded to substance in IDMP. Among the five types of substances specified by IDMP, herbal medicines were expressed as structurally diverse. Scientific name was used as an invariant property of herbal medicine, and the substance level included information about source material and modification, and specifically included information about the scientific name, medicinal part, fraction, and processing. In addition, the specified substance level had information on the constituents, characteristic attributes, manufacturing, and grade of the herbal medicine. Conclusions : It is necessary to establish a code system for identifying herbal medicines. In order to apply the IDMP standards, research on the development of standard terms is required to express the characteristics of herbal medicines. In addition, information for identification of herbal medicines is also required, and information from production to consumption should be systematically accumulated and managed for actual application.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.4
• /
• pp.97-113
• /
• 2023

Swift and appropriate responses in emergency situations like objects falling on the road can bring convenience to road users and effectively reduces secondary traffic accidents. In Korea, current intelligent transportation system (ITS)-based detection systems for emergency road situations mainly rely on loop detectors and CCTV cameras, which only capture road data within detection range of the equipment. Therefore, a new detection method is needed to identify emergency situations in spatially shaded areas that existing ITS detection systems cannot reach. In this study, we propose a ResNet-based algorithm that detects and classifies emergency situations from vehicle camera footage. We collected front-view driving videos recorded on Korean highways, labeling each video by defining the type of emergency, and training the proposed algorithm with the data.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.10
• /
• pp.1-9
• /
• 2018

The purpose of this exploratory study is to identify the mobile phone applications (apps) used by foreign tourists visiting South Korea through a pilot study using focus groups and individual interviews. Concentrating on tourist mobile app use in a smart tourism environment and categorized through a taxonomy of mobile applications lays the framework and determines the factors boosting tourism smartphone app trends by foreign tourists visiting South Korea. Researchers collected data through ethnographic methods and analyzed it through qualitative research to uncover major themes within the smart tourism app use phenomenon. The researchers coded, counted, analyzed, and then divided the findings gleaned from a pilot study and interviews into a taxonomy of seven logical smartphone app categories. The labeling and coding of all the data accounting for similarities and differences can be recognized and are logically discussed in the implications of the apps used by tourists to assist tourist destinations. More specifically these findings will assist smart tourism destinations by better understanding foreign tourist smartphone app use behavior. Tourists visiting South Korea interviewed in this study exhibited significant mastery of Internet of Things (IoT) technologies, craved free WiFi access, and utilized smartphone apps for all facets of their travel. Findings show major concentrations of app use in bookings of accommodations, tourist attractions, online shopping, navigation, wayfinding, augmented reality, information searching, language translation, gaming, and online dating while traveling in South Korea.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.607-610
• /
• 2021

When learning a model in supervised learning, input data and the label of the data are required. However, labeling is high cost task and if automated, there is no guarantee that the label will always be correct. In the case of supervised learning in such a noisy labels environment, the accuracy of the model increases at the initial stage of learning, but decrease significantly after a certain period of time. There are various methods to solve the noisy label problem. But in most cases, the probability predicted by the model is used as the pseudo label. So, we proposed a method to predict the true label more quickly by refining the probabilities predicted by the model. Result of experiments on the same environment and dataset, it was confirmed that the performance improved and converged faster. Through this, it can be applied to methods that use the probability distribution predicted by the model among existing studies. And it is possible to reduce the time required for learning because it can converge faster in the same environment.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.37-53
• /
• 2023

The AI based intelligent command and control system can automatically analyzes the properties of intricate battlefield information and tactical data. In addition, commanders can receive situation analysis results and battlefield awareness through the system to support decision-making. It is necessary to build a battlefield situation analysis dataset similar to the actual battlefield situation for learning AI in order to provide decision-making support to commanders. In this paper, we explain the next step of the dataset construction method of the existing previous research, 'A Virtual Battlefield Situation Dataset Generation for Battlefield Analysis based on Artificial Intelligence'. We proposed a method to build the dataset required for the final battlefield situation analysis results to support the commander's decision-making and recognize the future battlefield. We developed 'Dataset Generator SW', a software tool to build a learning dataset for battlefield situation analysis, and used the SW tool to perform data labeling. The constructed dataset was input into the Siamese Network model. Then, the output results were inferred to verify the dataset construction method using a post-processing ranking algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.10
• /
• pp.2643-2657
• /
• 2023

Scene graphs are structured representations that can clearly convey objects and the relationships between them, but are often heavily biased due to the highly skewed, long-tailed relational labeling in the dataset. Indeed, the visual world itself and its descriptions are biased. Therefore, Unbiased Scene Graph Generation (USGG) prefers to train models to eliminate long-tail effects as much as possible, rather than altering the dataset directly. To this end, we propose Geometric and Semantic Improvement (GSI) for USGG to mitigate this issue. First, to fully exploit the feature information in the images, geometric dimension and semantic dimension enhancement modules are designed. The geometric module is designed from the perspective that the position information between neighboring object pairs will affect each other, which can improve the recall rate of the overall relationship in the dataset. The semantic module further processes the embedded word vector, which can enhance the acquisition of semantic information. Then, to improve the recall rate of the tail data, the Class Balanced Seesaw Loss (CBSLoss) is designed for the tail data. The recall rate of the prediction is improved by penalizing the body or tail relations that are judged incorrectly in the dataset. The experimental findings demonstrate that the GSI method performs better than mainstream models in terms of the mean Recall@K (mR@K) metric in three tasks. The long-tailed imbalance in the Visual Genome 150 (VG150) dataset is addressed better using the GSI method than by most of the existing methods.