• International journal of advanced smart convergence
• /
• 제12권3호
• /
• pp.61-67
• /
• 2023

Gait analysis plays a key role in the research field of exploring and understanding human movement. By quantitatively analyzing the complexity of human movement and the various factors that influence it, it is possible to identify individual gait characteristics and abnormalities. This is especially true for people with walking difficulties or special circumstances, such as amputee, for example. This is because it can help us understand their gait characteristics and provide individualized rehabilitation plans. In this paper, we compare and analyze the differences in ankle joint motion and angles between normal and amputee. In particular, a filtering process was applied to the ankle joint angle data to obtain high accuracy results. The results of this study can contribute to a more accurate understanding and improvement of the gait patterns of normal and amputee.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권7호
• /
• pp.3217-3238
• /
• 2018

In this paper, a depth image up-sampling method is put forward by using pixel classifying and jointed bilateral filtering. By analyzing the edge maps originated from the high-resolution color image and low-resolution depth map respectively, pixels in up-sampled depth maps can be classified into four categories: edge points, edge-neighbor points, texture points and smooth points. First, joint bilateral up-sampling (JBU) method is used to generate an initial up-sampling depth image. Then, for each pixel category, different refinement methods are employed to modify the initial up-sampling depth image. Experimental results show that the proposed algorithm can reduce the blurring artifact with lower bad pixel rate (BPR).

• Journal of Information Processing Systems
• /
• 제16권4호
• /
• pp.820-831
• /
• 2020

Extractive document summarization aims to select a few sentences while preserving its main information on a given document, but the current extractive methods do not consider the sentence-information repeat problem especially for news document summarization. In view of the importance and redundancy of news text information, in this paper, we propose a neural extractive summarization approach with joint sentence semantic clipping and selection, which can effectively solve the problem of news text summary sentence repetition. Specifically, a hierarchical selective encoding network is constructed for both sentence-level and document-level document representations, and data containing important information is extracted on news text; a sentence extractor strategy is then adopted for joint scoring and redundant information clipping. This way, our model strikes a balance between important information extraction and redundant information filtering. Experimental results on both CNN/Daily Mail dataset and Court Public Opinion News dataset we built are presented to show the effectiveness of our proposed approach in terms of ROUGE metrics, especially for redundant information filtering.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2011년도 춘계학술대회 논문집
• /
• pp.281-282
• /
• 2011

• IEIE Transactions on Smart Processing and Computing
• /
• 제6권2호
• /
• pp.93-101
• /
• 2017

This paper proposes an algorithm for adaptive loop filter (ALF) complexity reduction in the decoding process. In the original ALF algorithm, filtering for I frames is performed in the frame unit, and thus, all of the pixels in a frame are filtered if the current frame is an I frame. The proposed algorithm is designed on top of the local gradient calculation. On both the encoder side and the decoder side, homogeneous areas are checked and skipped in the filtering process, and the filter coefficient calculation is only performed in the inhomogeneous areas. The proposed algorithm is implemented in Joint Exploration Model (JEM) version 3.0 future video coding reference software. The proposed algorithm is applied for frame-level filtering and intra configuration. Compared with the JEM 3.0 anchor, the proposed algorithm has 0.31%, 0.76% and 0.73% bit rate loss for luma (Y) and chroma (U and V), respectively, with about an 8% decrease in decoding time.

• 지능정보연구
• /
• 제5권1호
• /
• pp.103-123
• /
• 1999

Input filtering as a preprocessing method is so much crucial to get good performance in time series forecasting. There are a few preprocessing methods (i.e. ARMA outputs as time domain filters, and Fourier transform or wavelet transform as time-frequency domain filters) for handling time series. Specially, the time-frequency domain filters describe the fractal structure of financial markets better than the time domain filters due to theoretically additional frequency information. Therefore, we, first of all, try to describe and analyze specially some issues on the effectiveness of different filtering methods from viewpoint of the performance of a neural network based forecasting. And then we discuss about neural network model architecture issues, for example, what type of neural network learning architecture is selected for our time series forecasting, and what input size should be applied to a model. In this study an input selection problem is limited to a size selection of the lagged input variables. To solve this problem, we simulate on analyzing and comparing a few neural networks having different model architecture and also use an embedding dimension measure as chaotic time series analysis or nonlinear dynamic analysis to reduce the dimensionality (i.e. the size of time delayed input variables) of the models. Throughout our study, experiments for integration methods of joint time-frequency analysis and neural network techniques are applied to a case study of daily Korean won / U. S dollar exchange returns and finally we suggest an integration framework for future research from our experimental results.

• The Journal of the Acoustical Society of Korea
• /
• 제27권1E호
• /
• pp.30-34
• /
• 2008

An improvement to the existing blind signal separation (BSS) method has been made in this paper. The proposed method models the inherent signal dependency observed in acoustic object to separate the real-world convolutive sound mixtures. The frequency domain approach requires solving the well known permutation problem, and the problem had been successfully solved by a vector representation of the sources whose multidimensional joint densities have a certain amount of dependency expressed by non-spherical distributions. Especially for speech signals, we observe strong dependencies across neighboring frequency bins and the decrease of those dependencies as the bins become far apart. The non-spherical joint density model proposed in this paper reflects this property of real-world speech signals. Experimental results show the improved performances over the spherical joint density representations.

• ETRI Journal
• /
• 제37권1호
• /
• pp.43-53
• /
• 2015

In this paper, a selective coefficient updating (SCU) approach at each branch of the per-tone equalization (PTEQ) structure has been applied for insufficient cyclic prefix (CP) length. Because of the high number of adaptive filters and their complex adaption process in the PTEQ structure, SCU has been proposed. Using this method leads to a reduction in the computational complexity, while the performance remains almost unchanged. Moreover, the use of set-membership filtering with variable step size is proposed for a sufficient CP case to increase convergence speed and decrease the average number of calculations. Simulation results show that despite the aforementioned algorithms having similar performance in comparison with conventional algorithms, they are able to reduce the number of calculations necessary. In addition, compensation of both the channel effect and the transmitter/receiver in-phase/quadrature-phase imbalances are achievable by these algorithms.

• 한국게임학회 논문지
• /
• 제17권5호
• /
• pp.133-142
• /
• 2017

최근 스크린 클라이밍용 콘텐츠로 클라이밍 학습 프로그램과 스크린 클라이밍 게임이 등장하였으며, 특히 스크린 클라이밍 게임에 대한 연구가 활발히 진행되고 있다. 본 논문에서는 스크린 클라이밍 콘텐츠 구현의 핵심 기술인 자세 인식 성능의 개선을 위하여 등반자의 신체영역을 기반으로 하는 스켈레톤 보정 방법을 제안한다. 스켈레톤 보정 과정은 비정상적인 스켈레톤 정보를 걸러내는 스켈레톤 프레임 안정화와 신체 영역을 관절부위별로 나누어 각 관절부위의 중점을 보정위치로 하는 신체영역 기반 스켈레톤 수정 과정으로 이루어진다. 이렇게 보정한 스켈레톤 정보는 클라이밍 콘텐츠에서 등반자의 자세가 이상적인 자세와 얼마나 유사한지 판단하는 데 사용될 수 있다.

• 한국멀티미디어학회논문지
• /
• 제5권2호
• /
• pp.147-157
• /
• 2002

영상 시퀀스에 포함된 잡음은 화질을 열화시키고 중요한 영상의 미세 정보를 훼손시킨다. 또한 잡음은 영상의 복잡성을 증가시키므로 영상의 압축효율을 저하시킨다. 본 논문에서는 특성이 서로 다른 두개의 적응형 잡음제거 필터를 결합적으로 연결하여 잡음을 제거하는 공간결합필터와 시간결합필터를 제시하고, 이 결합필터들을 연결한 IIR필터 구조를 갖는 3차원 잡음제거 필터를 제시한다. 제시된 잡음제거 필터는 영상의 에지와 미세정보를 유지하면서도 평탄한 영상영역에 대해서는 강한 잡음제거를 수행하며 이웃한 프레임 사이에서도 잡음에 의한 시간적인 깜박임을 억제한다. 실험결과 제안된 잡음제거 필터는 기존 방식들에 비하여 객관적 화질뿐만 아니라 주관적인 화질을 크게 향상시킴을 알 수 있었다.