• The Journal of the Acoustical Society of Korea
• /
• v.39 no.6
• /
• pp.615-621
• /
• 2020

In this paper, we propose a multi-site based earthquake event classification method using graph convolution networks. In the traditional earthquake event classification methods using deep learning, they used single-site observation to estimate seismic event class. However, to achieve robust and accurate earthquake event classification on the seismic observation network, the method using the information from the multi-site observations is needed, instead of using only single-site data. Firstly, our proposed model employs convolution neural networks to extract informative embedding features from the single-site observation. Secondly, graph convolution networks are used to integrate the features from several stations. To evaluate our model, we explore the model structure and the number of stations for ablation study. Finally, our multi-site based model outperforms up to 10 % accuracy and event recall rate compared to single-site based model.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.171-171
• /
• 2011

상수도의 합리적인 운용과 관리를 위해서는 급수량 예측이 매우 중요하다. 기존 급수량 예측은 신경망과 칼만 필터법을 사용한 연구들이 대부분이었다. 이러한 연구결과들은 높은 상관결과를 갖고 있지만 이는 자기상관계수에 대한 높은 의존도에 따른 결과로 볼 수 있다. 즉, 예측의 결과가 전날 수요량을 거의 그대로 따라오는 경향을 띄어, 급수량 예측 그래프가 기존 그래프를 오른쪽으로 이동시킨 것과 같이 나타난다. 본 연구에서는 이러한 문제점들을 해결하기 위해서 물수요량을 예측하는데 있어서 효과적인 예측인자를 도출하는 것이 우선되어야 할 것으로 판단되었다. 이에, 물수요량 특성을 효과적으로 나타내어 줄 수 있는 예측인자로서 강수량, 최저온도, 최고온도, 평균온도 등을 1차적으로 선정하였다. 이들 예측인자들과 서울시 물수요량과의 상관성을 평가하여 최적의 예측인자 Set과 지체시간 등을 산정하였다. 이렇게 선정된 예측인자와 Bayesian 통계기법 기반의 회귀분석 모형을 구축하여 물수요량을 예측하였다. 본 연구에서 적용하고자 하는 계층적 Bayesian 모형은 유사한 특성을 가지는 자료계열들 사이에서 서로 보완이 될 수 있는 정보들을 추출함으로써 모형이 갖는 불확실성을 상당히 줄일 수 있는 방법이다. 이러한 모형적 특징은 생산량 예측에 대한 불확실성 저감 측면에서 장점이 있을 것으로 판단된다. 본 연구에서는 광암, 암사, 구의, 뚝도, 영등포, 강북 정수장을 대상으로 모형의 적합성을 평가하였다. 이러한 연구결과는 향후 정수장 운영계획 및 동일한 시스템을 갖는 상수도 급수량 예측 시 유용하게 사용할 수 있을 것이다.

• The KIPS Transactions:PartB
• /
• v.8B no.6
• /
• pp.651-656
• /
• 2001

In this paper, we propose Improved Self-Organized Graph(Improved SOG：$SOG^*$)thinning method, which maintains the excellent thinning results of Self-organized graph(SOG) built from Self-Organizing features map and improves the performance of modified SOG using a new incremental learning method of Kohonen features map. In the experiments, this method shows the thinning results equal to those of SOG and the time complexity O((logM)3) superior to it. Therefore, the proposed method is useful for the feature extraction from digits and characters in the preprocessing step.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.10
• /
• pp.63-70
• /
• 2014

Process of recognizing objects in binary images consists of image segmentation and pattern matching. If binary objects in the image are assumed to be separated, global features such as area, length of perimeter, or the ratio of the two can be used to recognize the objects in the image. However, if such an assumption is not valid, the global features can not be used but local features such as points or line segments should be used to recognize the objects. In this paper points with large curvature along the perimeter are chosen to be the feature points, and pairs of points selected from them are used as local features. Similarity of two local features are defined using elastic deformation energy for making the lengths and angles between gradient vectors at the end points same. Neighbour support value is defined and used for robust recognition of partially occluded binary objects. An experiment on Kimia-25 data showed that the proposed algorithm runs 4.5 times faster than the maximum clique algorithm with same recognition rate.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.2
• /
• pp.41-50
• /
• 2004

Since the MPEG-4 visual standard enables content-based functionalities, it is necessary to extract video objects from video sequences. Segmentation algorithms can largely be classified into two different categories: automatic segmentation and user-assisted segmentation. In this paper, we propose a new user-assisted image segmentation method based on the active contour. If we define a shape space as a set of all possible variations from the initial curve and we assume that the shape space is linear, it can be decomposed into the column space and the left null space of the shape matrix. In the proposed method, the shape space vector in the column space describes changes from the initial curve to the imaginary feature curve, and a dynamic graph search algorithm describes the detailed shape of the object in the left null space. Since we employ the shape matrix and the SUSAN operator to outline object boundaries, the proposed algorithm can ignore unwanted feature points generated by low-level image processing operations and is, therefore, applicable to images of complex background. We can also compensate for limitations of the shape matrix with a dynamic graph search algorithm.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.4
• /
• pp.349-358
• /
• 2019

This paper aimed at collecting sensor signals to extract characteristic parameter of the rotor. A vibration test rig has been developed to perform model tests. Signal characteristics were analyzed when driving normally. Envelope FFT Analysis is used to extract vibration components caused by periodic impacts from other vibration factors. Signal analysis was performed when load changes were given to speed sensors and vibration test rigs that show low frequency characteristics of the rotor and signal analysis according to rotational speed. The acceleration signal measured in the bearing housing has a small amplitude and produces only the rotational frequency component and harmonic component of the motor. As the number of rotations increases, the amplitude of acceleration can be seen. As the rotational speed increases, it can be seen that there is a difference in the shape of the original data and compared with the acceleration FFT graph, it can be seen that the noise is strong at low frequencies and the corresponding rotational frequency components are clearly represented. It can be seen that changing the load does not increase the main rotational frequency component.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.369-370
• /
• 2012

일반적으로 회귀분석의 최적화는 평균적인 개념을 확장하여 사용되어지고 있다. 평균은 관찰값들에 관한 모든 정보와 관련된 통계량으로써 많은 연구에 이용되어지고 있다. 정규분포를 이루는 모집단의 경우 평균을 사용한 추정이 바람직하지만, 이상치로 인한 분포의 꼬리가 두꺼워지는 경우 중위수(median)를 사용하는 것이 바람직하다고 알려져 있다. 강수량의 분포형태는 꼬리(tail)가 두꺼운 왜곡된 형태를 갖고 있으므로 robust 통계량인 Quantile을 이용한 강수량의 분석 및 평가를 실시하였다. 본 연구에서는 Quantile에 따른 회귀선의 변화를 이용하여 강수량의 경향성을 평가하고, 극치강수량의 변화를 보여줄 수 있는 Quantle값을 추출해 보고자 한다. 또한 bootstrap 방법을 이용하여 Quantile에 따른 회귀계수의 신뢰구간을 분석하여 회귀인자의 신뢰성을 평가하였다. 본 연구에서 적용한 Quantile Regression 기법은 회귀계수의 추정에 있어서 회귀인자의 신뢰성을 Quantile-회귀계수 그래프를 통해 분석할 수 있으며, 이상값의 영향을 저감시키는 평균과 달리 이상값의 영향을 효과적으로 분리 및 재현시킬 수 있어 극치값에 따른 변화를 효과적으로 평가할 수 있으며, robust 통계량의 특징인 분산이 적은 안정적인 추정량을 확보할 수 있다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.8
• /
• pp.1604-1611
• /
• 2007

In a process of insect footprint recognition, basic footprint segments should be extracted from a whole insect footprint image in order to find out appropriate features for classification. In this paper, we used a clustering method as a preprocessing stage for extraction of basic insect footprint segments. In general, sizes and strides of footprints may be different according to type and sire of an insect for recognition. Therefore we proposed an improved ART2 algorithm for extraction or basic insect footprint segments regardless of size and stride or footprint pattern. In the proposed ART2 algorithm, threshold value for clustering is determined automatically using contour shape of the graph created by accumulating distances between all the spots of footprint pattern. In the experimental results applying the proposed method to two kinds of insect footprint patterns, we could see that all the clustering results were accomplished correctly.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.5
• /
• pp.477-481
• /
• 2008

Place recognition is necessary for a mobile user to be provided with place-dependent information. This paper proposes real-time video based place recognition system that identifies users' current place while moving in the building. As for the feature extraction of a scene, there have been existing methods based on global feature analysis that has drawback of sensitive-ness for the case of partial occlusion and noises. There have also been local feature based methods that usually attempted object recognition which seemed hard to be applied in real-time system because of high computational cost. On the other hand, researches using statistical methods such as HMM(hidden Markov models) or bayesian networks have been used to derive place recognition result from the feature data. The former is, however, not practical because it requires huge amounts of efforts to gather the training data while the latter usually depends on object recognition only. This paper proposes a combined approach of global and local feature analysis for feature extraction to complement both approaches' drawbacks. The proposed method is applied to a mobile information system and shows real-time performance with competitive recognition result.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.2
• /
• pp.147-157
• /
• 2017

In recent years, Super-pixels have become very popular for use in computer vision applications. Super-pixel algorithm transforms pixels into perceptually feasible regions to reduce stiff features of grid pixel. In particular, super-pixels are useful to depth estimation, skeleton works, body labeling, and feature localization, etc. But, it is not easy to generate a good super-pixel partition for doing these tasks. Especially, super-pixels do not satisfy more meaningful features in view of the gestalt aspects such as non-sum, continuation, closure, perceptual constancy. In this paper, we suggest an advanced algorithm which combines simple linear iterative clustering with fuzzy clustering concepts. Simple linear iterative clustering technique has high adherence to image boundaries, speed, memory efficient than conventional methods. But, it does not suggest good compact and regular property to the super-pixel shapes in context of gestalt aspects. Fuzzy similarity measures provide a reasonable graph in view of bounded size and few neighbors. Thus, more compact and regular pixels are obtained, and can extract locally relevant features. Simulation shows that fuzzy similarity based super-pixel building represents natural features as the manner in which humans decompose images.