• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제21권11호
• /
• pp.713-720
• /
• 2015

컴퓨팅 기술과 네트워크의 발달로 인해, 정보 기기가 소형화되고 이동성이 강조되고 있다. 이에 따라 모바일 환경에서 작동 가능한 음성 인식 시스템에 대한 수요가 최근 급격히 증대되고 있다. 본 논문은 음소 기반 한국어 음성 인식 시스템의 일부로서, 한국어 모음 'ㅐ'에 대한 새로운 인식 방식을 제안한다. 제안하는 방식은 주파수 영역에서의 분석을 배제하고, 시간 영역에서의 시퀀스 패턴에 기반하여 인식을 수행함으로써, 계산 비용을 현저히 절감할 수 있다. 'ㅐ'의 전형적인 시퀀스 패턴들을 탐지하기 위한 세 가지 알고리즘이 제시되며, 이를 결합하여 최종 판별을 수행한다. 실험 결과를 통해, 제안하는 방식이 89.1%의 정확도로 모음 'ㅐ'를 인식할 수 있음을 확인하였다.

• Journal of Information Processing Systems
• /
• 제13권5호
• /
• pp.1213-1228
• /
• 2017

The pitch tracking of music has been researched for several decades. Several possible improvements are available for creating a good t-distribution, using the instantaneous robust algorithm for pitch tracking framework to perfectly detect pitch. This article shows how to detect the pitch of music utilizing an improved detection method which applies a statistical method; this approach uses a pitch track, or a sequence of frequency bin numbers. This sequence is used to create an index that offers useful features for comparing similar songs. The pitch frequency spectrum is extracted using a modified instantaneous robust algorithm for pitch tracking (IRAPT) as a base combined with the statistical method. The pitch detection algorithm was implemented, and the percentage of performance matching in Thai classical music was assessed in order to test the accuracy of the algorithm. We used the longest common subsequence to compare the similarities in pitch sequence alignments in the music. The experimental results of this research show that the accuracy of retrieval of Thai classical music using the t-distribution of instantaneous robust algorithm for pitch tracking (t-IRAPT) is 99.01%, and is in the top five ranking, with the shortest query sample being five seconds long.

• Steel and Composite Structures
• /
• 제29권6호
• /
• pp.749-758
• /
• 2018

This article derived a hybrid coupling technique using the higher-order displacement polynomial and three soft computing techniques (teaching learning-based optimization, particle swarm optimization, and artificial bee colony) to predict the optimal stacking sequence of the layered structure and the corresponding frequency values. The higher-order displacement kinematics is adopted for the mathematical model derivation considering the necessary stress and stain continuity and the elimination of shear correction factor. A nine noded isoparametric Lagrangian element (eighty-one degrees of freedom at each node) is engaged for the discretisation and the desired model equation derived via the classical Hamilton's principle. Subsequently, three soft computing techniques are employed to predict the maximum natural frequency values corresponding to their optimum layer sequences via a suitable home-made computer code. The finite element convergence rate including the optimal solution stability is established through the iterative solutions. Further, the predicted optimal stacking sequence including the accuracy of the frequency values are verified with adequate comparison studies. Lastly, the derived hybrid models are explored further to by solving different numerical examples for the combined structural parameters (length to width ratio, length to thickness ratio and orthotropicity on frequency and layer-sequence) and the implicit behavior discuss in details.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권7호
• /
• pp.1794-1806
• /
• 2023

This study presents a method for capturing photographs of users as input and converting them into 2D character animation sprites using a generative adversarial network-based artificial intelligence network. Traditionally, 2D character animations have been created by manually creating an entire sequence of sprite images, which incurs high development costs. To address this issue, this study proposes a technique that combines motion videos and sample 2D images. In the 2D sprite generation process that uses the proposed technique, a sequence of images is extracted from real-life images captured by the user, and these are combined with character images from within the game. Our research aims to leverage cutting-edge deep learning-based image manipulation techniques, such as the GAN-based motion transfer network (impersonator) and background noise removal (U² -Net), to generate a sequence of animation sprites from a single image. The proposed technique enables the creation of diverse animations and motions just one image. By utilizing these advancements, we focus on enhancing productivity in the game and animation industry through improved efficiency and streamlined production processes. By employing state-of-the-art techniques, our research enables the generation of 2D sprite images with various motions, offering significant potential for boosting productivity and creativity in the industry.

• Asian-Australasian Journal of Animal Sciences
• /
• 제30권2호
• /
• pp.254-261
• /
• 2017

Objective: The effects of aging and freezing/thawing sequence on color, physicochemical, and enzymatic characteristics of two beef muscles (Mm. gluteus medius, GM and biceps femoris, BF) were evaluated. Methods: Beef muscles at 3 d postmortem were assigned to four different combinations of aging and freezing/thawing sequence as follows; aging at $2^{\circ}C$ for 3 wk (A3, never-frozen control), freezing at $-28^{\circ}C$ for 2 wk then thawing (F2, frozen/thawed-only), aging at $2^{\circ}C$ for 3 wk, freezing at $-28^{\circ}C$ for 2 wk then thawing (A3F2), and freezing at $-28^{\circ}C$ for 2 wk, thawing then further aging at $2^{\circ}C$ for 3 wk (F2A3). Results: No significant interactions between different aging/freezing/thawing treatments and muscle type on all measurements were found. Postmortem aging, regardless of aging/freezing/thawing sequence, had no impact on color stability of frozen/thawed beef muscles (p<0.05). F2A3 resulted in higher purge loss than F2 and A3F2 treatments (p<0.05). A3F2 and F2A3 treatments resulted in lower shear force of beef muscles compared to F2 (p<0.05). Although there was no significant difference in glutathione peroxidase (GSH-Px) activity, F2A3 had the highest ${\beta}-N-acetyl$ glucominidase (BNAG) activity in purge, but the lowest BNAG activity in muscle (p<0.05). GM muscle exhibited higher total color changes and purge loss, and lower GSH-Px activity than BF muscle. Conclusion: The results from this present study indicate that different combinations of aging/freezing/thawing sequence would result in considerable impacts on meat quality attributes, particularly thaw/purge loss and tenderness. Developing a novel freezing strategy combined with postmortem aging will be beneficial for the food/meat industry to maximize its positive impacts on tenderness, while minimizing thaw/purge loss of frozen/thawed meat.

• 대한자기공명의과학회:학술대회논문집
• /
• 대한자기공명의과학회 2001년도 제6차 학술대회 초록집
• /
• pp.104-104
• /
• 2001

Purpose： In this study, we investigated the possible motor pathways of hemiplegic stroke patients usin combined TMS and BOLD fMRI approach and evaluated the correlation between TMS a fMRI methods. Method： Four subjects, who demonstrated left hemiplegia after stroke, are included. TMS was performed using a Dantec Mag2 stimulator (Dantec Company, USA) in single puls mode with figure eight-shaped coil. Following TMS localization, The BOLD T2＊-weight images were acquired with echo planar imaging sequence (TR = 1.2 sec, TE = 60 msec, and flip angle = 90). Motor activation was studied by means of a repetitive fing flexion-extension task. The stimulation protocol comprised 10 cycles of alternating activati and rest (10 images per cycle). Total 60 cycles were performed and each cycle take abou 1.5 sec. The resulting images were then analyzed with STIMULATE (CMRR, U, o Minnesota) to generate functional maps using a student t-test (p < 0.0005) and cluste analysis.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2006년도 가을 학술발표논문집 Vol.33 No.2 (B)
• /
• pp.509-514
• /
• 2006

Human action recognition is an active research area in computer vision. In this paper, we present a robust method for human action recognition by using combined information of human body shape and motion information with multiple views image sequence. The principal component analysis is used to extract the shape feature of human body and multiple block motion of the human body is used to extract the motion features of human. This combined information with multiple view sequences enhances the recognition of human action. We represent each action using a set of hidden Markov model and we model each action by multiple views. This characterizes the human action recognition from arbitrary view information. Several daily actions of elderly persons are modeled and tested by using this approach and they are correctly classified, which indicate the robustness of our method.

• Nuclear Engineering and Technology
• /
• 제29권1호
• /
• pp.1-6
• /
• 1997

This paper presents a new dynamic approach for assessing feasibility associated with the implementation of accident management strategies by the operators. This approach includes the combined use of both the concept of reliability physics and a dynamic event tree generation scheme. The reliability physics is based on the concept of a comparison between two competing variables, i.e., the requirement and the achievement parameter, while the dynamic event tree generation scheme on the continuous generation of the possible event sequences at every branch point up to the desired solution. This approach is applied to a cavity flooding strategy in a reference plant, which is to supply water into the reactor cavity using emergency fire systems in the station blackout sequence. The MAAP code and Latin Hypercube sampling technique are used to determine the uncertainty of the requirement parameter. It has been demonstrated that this combined methodology may contribute to assessing the success likelihood of the operator actions required during accidents and therefore to developing the accident management procedures.

• 한국멀티미디어학회논문지
• /
• 제25권6호
• /
• pp.807-818
• /
• 2022

In order to further improve the accuracy and time efficiency of behavior recognition in intelligent monitoring scenarios, a human behavior recognition algorithm based on YOLO combined with LSTM and CNN is proposed. Using the real-time nature of YOLO target detection, firstly, the specific behavior in the surveillance video is detected in real time, and the depth feature extraction is performed after obtaining the target size, location and other information; Then, remove noise data from irrelevant areas in the image; Finally, combined with LSTM modeling and processing time series, the final behavior discrimination is made for the behavior action sequence in the surveillance video. Experiments in the MSR and KTH datasets show that the average recognition rate of each behavior reaches 98.42% and 96.6%, and the average recognition speed reaches 210ms and 220ms. The method in this paper has a good effect on the intelligence behavior recognition.

• Journal of Electrical Engineering and Technology
• /
• 제12권6호
• /
• pp.2106-2117
• /
• 2017

An effective statistical feature extraction approach of data sampling of fault in the combined transmission system is presented in this paper. The proposed algorithm leads to high accuracy at minimum cost to predict fault location and fault type classification. This algorithm requires impedance measurement data from one end of the transmission line. Modal decomposition is used to extract positive sequence impedance. Then, the fault signal is decomposed by using discrete wavelet transform. Statistical sampling is used to extract appropriate fault features as benchmark of decomposed signal to train classifier. Support Vector Machine (SVM) is used to illustrate the performance of statistical sampling performance. The overall time of sampling is not exceeding 1 1/4 cycles, taking into account the interval time. The proposed method takes two steps of sampling. The first step takes 3/4 cycle of during-fault and the second step takes 1/4 cycle of post fault impedance. The interval time between the two steps is assumed to be 1/4 cycle. Extensive studies using MATLAB software show accurate fault location estimation and fault type classification of the proposed method. The classifier result is presented and compared with well-established travelling wave methods and the performance of the algorithms are analyzed and discussed.