• The Journal of the Acoustical Society of Korea
• /
• v.25 no.6
• /
• pp.277-281
• /
• 2006

This paper presents a method for the automatic recognition of pitch accents over syllables. The method that we propose is based on the time-delay recursive neural network (TDRNN). which is a neural network classifier with two different representation of dynamic context: the delayed input nodes allow the representation of an explicit trajectory F0(t) along time. while the recursive nodes provide long-term context information that reflects the characteristics of pitch accentuation in spoken English. We apply the TDRNN to pitch accent recognition in two forms: in the normal TDRNN. all of the prosodic features (pitch. energy, duration) are used as an entire set in a single TDRNN. while in the distributed TDRNN. the network consists of several TDRNNs each taking a single prosodic feature as the input. The final output of the distributed TDRNN is weighted sum of the output of individual TDRNN. We used the Boston Radio News Corpus (BRNC) for the experiments on the speaker-independent pitch accent recognition. π 1e experimental results show that the distributed TDRNN exhibits an average recognition accuracy of 83.64% over both pitch events and non-events.

• The Journal of the Acoustical Society of Korea
• /
• v.29 no.6
• /
• pp.381-387
• /
• 2010

Both timbral features and spatial features are important in the assessment of multichannel audio coding systems. The prediction model, extending the ITU-R Rec. BS. 1387-1 to multichannel audio coding systems, with the use of spatial features such as ITDDist (Interaural Time Difference Distortion), ILDDist (Interaural Level Difference Distortion), and IACCDist (InterAural Cross-correlation Coefficient Distortion) was proposed by Choi et al. In that model, ITDDistswere only computed for low frequency bands (below 1500Hz), and ILDDists were computed only for high frequency bands (over 2500Hz) according to classical duplex theory. However, in the high frequency range, information in temporal envelope is also important in spatial perception, especially in sound localization. A new model to compute the ITD distortions of temporal envelopes in high frequency components is introduced in this paper to investigate the role of such ITD on spatial perception quantitatively. The computed ITD distortions of temporal envelopes in high frequency components were highly correlated with perceived sound quality of multichannel audio sounds.

• IMMUNE NETWORK
• /
• v.22 no.5
• /
• pp.42.1-42.20
• /
• 2022

Vaccination with tumor peptide epitopes associated with MHC class I molecules is an attractive approach directed at inducing tumor-specific CTLs. However, challenges remain in improving the therapeutic efficacy of peptide epitope vaccines, including the low immunogenicity of peptide epitopes and insufficient stimulation of innate immune components in vivo. To overcome this, we aimed to develop and test an innovative strategy that elicits potent CTL responses against tumor epitopes. The essential feature of this strategy is vaccination using tumor epitope-loaded nanoparticles (NPs) in combination with polyinosinic-polycytidylic acid (poly-IC) and anti-PD1 mAb. Carboxylated NPs were prepared using poly(lactic-co-glycolic acid) and poly(ethylene/maleic anhydride), covalently conjugated with anti-H-2K^b mAbs, and then attached to H-2K^b molecules isolated from the tumor mass (H-2^b). Native peptides associated with the H-2K^b molecules of H-2K^b-attached NPs were exchanged with tumor peptide epitopes. Tumor peptide epitope-loaded NPs efficiently induced tumor-specific CTLs when used to immunize tumor-bearing mice as well as normal mice. This activity of the NPs significantly was increased when co-administered with poly-IC. Accordingly, the NPs exerted significant anti-tumor effects in mice implanted with EG7-OVA thymoma or B16-F10 melanoma, and the anti-tumor activity of the NPs was significantly increased when applied in combination with poly-IC. The most potent anti-tumor activity was observed when the NPs were co-administered with both poly-IC and anti-PD1 mAb. Immunization with tumor epitope-loaded NPs in combination with poly-IC and anti-PD1 mAb in tumor-bearing mice can be a powerful means to induce tumor-specific CTLs with therapeutic anti-tumor activity.

• Journal of the Korean earth science society
• /
• v.45 no.2
• /
• pp.111-120
• /
• 2024

Cloud coverage is a key factor in determining whether to proceed with observations. In the past, human judgment played an important role in weather evaluation for observations. However, the development of remote and robotic observation has diminished the role of human judgment. Moreover, it is not easy to evaluate weather conditions automatically because of the diverse cloud shapes and their rapid movement. In this paper, we present the development of a cloud monitoring program by applying a machine learning-based Python module "cloudynight" on all-sky camera images obtained at Miryang Arirang Astronomical Observatory (MAAO). The machine learning model was built by training 39,996 subregions divided from 1,212 images with altitude/azimuth angles and extracting 16 feature spaces. For our training model, the F1-score from the validation samples was 0.97, indicating good performance in identifying clouds in the all-sky image. As a result, this program calculates "Cloudiness" as the ratio of the number of total subregions to the number of subregions predicted to be covered by clouds. In the robotic observation, we set a policy that allows the telescope system to halt the observation when the "Cloudiness" exceeds 0.6 during the last 30 minutes. Following this policy, we found that there were no improper halts in the telescope system due to incorrect program decisions. We expect that robotic observation with the 0.7 m telescope at MAAO can be successfully operated using the cloud monitoring program.

• Journal of Internet Computing and Services
• /
• v.25 no.2
• /
• pp.57-67
• /
• 2024

This paper explores the potential of electromyography (EMG) as a means of gesture recognition for user input in gesture-based interaction. EMG utilizes small electrodes within muscles to detect and interpret user movements, presenting a viable input method. To classify user gestures based on EMG data, machine learning techniques are employed, necessitating the preprocessing of raw EMG data to extract relevant features. EMG characteristics can be expressed through formulas such as Integrated EMG (IEMG), Mean Absolute Value (MAV), Simple Square Integral (SSI), Variance (VAR), and Root Mean Square (RMS). Additionally, determining the suitable time for gesture classification is crucial, considering the perceptual, cognitive, and response times required for user input. To address this, segment sizes ranging from a minimum of 100ms to a maximum of 1,000ms are varied, and feature extraction is performed to identify the optimal segment size for gesture classification. Notably, data learning employs overlapped segmentation to reduce the interval between data points, thereby increasing the quantity of training data. Using this approach, the paper employs four machine learning models (KNN, SVC, RF, XGBoost) to train and evaluate the system, achieving accuracy rates exceeding 96% for all models in real-time gesture input scenarios with a maximum segment size of 200ms.

• Journal of IKEEE
• /
• v.28 no.2
• /
• pp.221-227
• /
• 2024

This paper presents a smartphone-attachable vascular compliance monitoring module. The proposed sensor module measures photoplethysmogram (PPG) and reconstructs an accelerated PPG waveform. The feature points are extracted from the accelerated PPG waves, and vascular compliance is estimated using these extracted features. The module is powered via the smartphone's USB terminal and transmits the acquired waveforms along with vascular compliance values through Bluetooth. The transmitted waveforms and vascular compliance value are displayed through the smartphone application. This work proposes an assessment method for consistency of PPG instrumentation, and it was implemented in a processor of sensor module. The proposed sensor module can be easily attached to smartphone that does not support PPG instrumentation, providing simple measurment and numerical analysis of vascular compliance. To verify the performance of the implemented sensor module, we acquired vascular compliance and pulse pressure data from 29 subjects. Pulse pressure, which serves as a representative indicator of vascular compliance, was obtained using a commercial blood pressure monitor. The analysis results showed that the Pearson coefficient between vascular compliance and pulse pressure was 0.778, confirming a relatively high correlation between two metrics.

• Journal of Cadastre & Land InformatiX
• /
• v.54 no.1
• /
• pp.75-87
• /
• 2024

There There is a difference in terms of administrative power in that the address of things are not an address under Public Act. In terms of location expression, it is possible to express the location more flexibly and in more detail than the road name address, so it should be improved so that it can be assigned and managed in an appropriate location, so that the location of the entire territory can be expressed together with the road name address. As a result of the comparison between the road name address and the address of things based on the analysis results of related laws such as the existing Road Name Address Act, the Building Act, and the Regulations on the Preparation and Management of Basic Address Information, it was confirmed that there are fundamental limitations of the address of things system. Accordingly, this study attempted to suggest ways to improve the address of thing system by broadly dividing it into the legal aspect and the addressable object aspect. From the legal point of view, firstly, it is necessary to improve the upper and lower level laws by unification together with a clear definition of the term subject of addressable object; secondly, according to the Building Act, facilities that are not used for residence among buildings must be given an address of thing; and thirdly, it is necessary to make it easy to use and link with heterogeneous public data by classifying the registration items of the basic address information map by type of geographical feature to be assigned an address. From the point of view of addressability, firstly, it must be given to all facilities in the relevant category so that it can be recognised that all specific facilities have object addresses, and secondly, it is necessary to be able to address the address of things to places that are used by many, even if there are no facilities.

• Smart Media Journal
• /
• v.12 no.11
• /
• pp.48-56
• /
• 2023

The technology of Three-dimensional human posture estimation is used in sports, motion recognition, and special effects of video media. Among various methods for this, multi-view 3D human pose estimation is essential for precise estimation even in complex real-world environments. But Existing models for multi-view 3D human posture estimation have the disadvantage of high order of time complexity as they use 3D feature maps. This paper proposes a method to extend an existing monocular viewpoint multi-frame model based on Transformer with lower time complexity to 3D human posture estimation for multi-viewpoints. To expand to multi-viewpoints our proposed method first generates an 8-dimensional joint coordinate that connects 2-dimensional joint coordinates for 17 joints at 4-vieiwpoints acquired using the 2-dimensional human posture detector, CPN(Cascaded Pyramid Network). This paper then converts them into 17×32 data with patch embedding, and enters the data into a transformer model, finally. Consequently, the MLP(Multi-Layer Perceptron) block that outputs the 3D-human posture simultaneously updates the 3D human posture estimation for 4-viewpoints at every iteration. Compared to Zheng[5]'s method the number of model parameters of the proposed method was 48.9%, MPJPE(Mean Per Joint Position Error) was reduced by 20.6 mm (43.8%) and the average learning time per epoch was more than 20 times faster.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.65-76
• /
• 2024

With the recent activation of the offshore wind power industry, there has been a development of power plants with a scale exceeding 400MW, comparable to traditional thermal power plants. Renewable energy, characterized by intermittency depending on the energy source, is a prominent feature of modern renewable power generation facilities, which are structured based on controllable inverter technology. As the integration of renewable energy sources into the grid expands, the grid codes for power system connection are progressively becoming more defined, leading to active discussions and evaluations in this area. In this paper, we propose a method for selecting optimal reactive power compensation capacity when multiple offshore wind farms are integrated and connected through a shared interconnection facility to comply with grid codes. Based on the requirements of the grid code, we analyze the reactive power compensation and excessive stability of the 400MW wind power generation site under development in the southwest sea of Jeonbuk. This analysis involves constructing a generation site database using PSS/E (Power System Simulation for Engineering), incorporating turbine layouts and cable data. The study calculates reactive power due to charging current in internal and external network cables and determines the reactive power compensation capacity at the interconnection point. Additionally, static and dynamic stability assessments are conducted by integrating with the power system database.

• The Mathematical Education
• /
• v.63 no.2
• /
• pp.165-186
• /
• 2024

Despite the growing attention on artificial intelligence-based automated scoring technology as a support method for the introduction of descriptive items in school environments and large-scale assessments, there is a noticeable lack of foundational research in mathematics compared to other subjects. This study developed an automated scoring model for two descriptive items in first-year middle school mathematics using the Random Forest algorithm, evaluated its performance, and explored ways to enhance this performance. The accuracy of the final models for the two items was found to be between 0.95 to 1.00 and 0.73 to 0.89, respectively, which is relatively high compared to automated scoring models in other subjects. We discovered that the strategic selection of the number of evaluation categories, taking into account the amount of data, is crucial for the effective development and performance of automated scoring models. Additionally, text preprocessing by mathematics education experts proved effective in improving both the performance and interpretability of the automated scoring model. Selecting a vectorization method that matches the characteristics of the items and data was identified as one way to enhance model performance. Furthermore, we confirmed that oversampling is a useful method to supplement performance in situations where practical limitations hinder balanced data collection. To enhance educational utility, further research is needed on how to utilize feature importance derived from the Random Forest-based automated scoring model to generate useful information for teaching and learning, such as feedback. This study is significant as foundational research in the field of mathematics descriptive automatic scoring, and there is a need for various subsequent studies through close collaboration between AI experts and math education experts.