• ETRI Journal
• /
• v.40 no.4
• /
• pp.499-510
• /
• 2018

A vector space based augmented structural kinematic (VSASK) feature descriptor is proposed for human activity recognition. An action descriptor is built by integrating the structural and kinematic properties of the actor using vector space based augmented matrix representation. Using the local or global information separately may not provide sufficient action characteristics. The proposed action descriptor combines both the local (pose) and global (position and velocity) features using augmented matrix schema and thereby increases the robustness of the descriptor. A multiclass support vector machine (SVM) is used to learn each action descriptor for the corresponding activity classification and understanding. The performance of the proposed descriptor is experimentally analyzed using the Weizmann and KTH datasets. The average recognition rate for the Weizmann and KTH datasets is 100% and 99.89%, respectively. The computational time for the proposed descriptor learning is 0.003 seconds, which is an improvement of approximately 1.4% over the existing methods.

• Structural Monitoring and Maintenance
• /
• v.9 no.2
• /
• pp.179-200
• /
• 2022

This paper proposes a data-driven methodology for online early damage identification under changing environmental conditions. The proposed method relies on two data analysis methods: feature-based method and hybrid principal component analysis (PCA) and kernel PCA to separate damage from environmental influences. First, spectral sub-band features, namely, spectral sub-band centroids (SSCs) and log spectral sub-band energies (LSSEs), are proposed as damage-sensitive features to extract damage information from measured structural responses. Second, hybrid modeling by integrating PCA and kernel PCA is performed on the spectral sub-band feature matrix for data normalization to extract both linear and nonlinear features for nonlinear procedure monitoring. After feature normalization, suppressing environmental effects, the control charts (Hotelling T² and SPE statistics) is implemented to novelty detection and distinguish damage in structures. The hybrid PCA-KPCA technique is compared to KPCA by applying support vector machine (SVM) to evaluate the effectiveness of its performance in detecting damage. The proposed method is verified through numerical and full-scale studies (a Bridge Health Monitoring (BHM) Benchmark Problem and a cable-stayed bridge in China). The results demonstrate that the proposed method can detect the structural damage accurately and reduce false alarms by suppressing the effects and interference of environmental variations.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.28 no.3
• /
• pp.74-82
• /
• 2024

In this study, feature point detection and displacement measurement performance depending on feature extraction algorithms were compared and analyzed according to environmental changes and target types in the feature point-based displacement measurement algorithm. A three-story frame structure was designed for performance evaluation, and the displacement response of the structure was digitized into FHD (1920×1080) resolution. For performance analysis, the initial measurement distance was set to 10m, and increased up to 40m with an increment of 10m. During the experiments, illuminance was fixed to 450lux or 120lux. The artificial and natural targets mounted on the structure were set as regions of interest and used for feature point detection. Various feature detection algorithms were implemented for performance comparisons. As a result of the feature point detection performance analysis, the Shi-Tomasi corner and KAZE algorithm were found that they were robust to the target type, illuminance change, and increase in measurement distance. The displacement measurement accuracy using those two algorithms was also the highest. However, when using natural targets, the displacement measurement accuracy is lower than that of artificial targets. This indicated the limitation in extracting feature points as the resolution of the natural target decreased as the measurement distance increased.

• Speech Sciences
• /
• v.6
• /
• pp.103-117
• /
• 1999

This paper suggests that the onset-coda asymmetry found in languages like Korean and others should be dealt with in terms of one acoustic feature rather than other articulatory features, claiming that the acoustic feature involved here is [noise], i.e., 'aperiodic waveform energy'. It determines the structural well-formedness of the languages in question whether a coda ends in [noise] or not, regardless of the intensity, the frequency, and the time duration of the [noise]. Fricatives, affricates, aspirated stops, tense stops, and released stops are all disallowed in the coda position due to the acoustic feature [noise] they, commonly end with if they were, posited in the coda. The proposal implies that the three seemingly separate prohibitions of consonants in the coda position -- i) no fricatives/affricates, ii) no aspirated/tense stops, and iii) no released stops -- are directly correlated with each other. Incorporation of the one acoustic feature [noise] in the feature theory enables us to see that the aspects of onset-coda asymmetry are derived from one single source: ban, of [noise] in the coda.

• Proceedings of the IEEK Conference
• /
• 2002.07b
• /
• pp.1252-1255
• /
• 2002

In this paper, face recognition algorithm, by using skin color information of HSI color coordinate collected from face images, elliptical mask, fratures of face including eyes, nose and mouth, and geometrical feature vectors of face and facial angles, is proposed. The proposed algorithm improved face region extraction efficacy by using HSI information relatively similar to human's visual system along with color tone information about skin colors of face, elliptical mask and intensity information. Moreover, it improved face recognition efficacy with using feature information of eyes, nose and mouth, and Θ1(ACRED), Θ2(AMRED) and Θ 3(ANRED), which are geometrical face angles of face. In the proposed algorithm, it enables exact face reading by using color tone information, elliptical mask, brightness information and structural characteristic angle together, not like using only brightness information in existing algorithm. Moreover, it uses structural related value of characteristics and certain vectors together for the recognition method.

• Smart Structures and Systems
• /
• v.2 no.3
• /
• pp.253-274
• /
• 2006

The structural monitoring of multi-wire strands is of importance to prestressed concrete structures and cable-stayed or suspension bridges. This paper addresses the monitoring of strands by ultrasonic guided waves with emphasis on the signal processing and automatic defect classification. The detection of notch-like defects in the strands is based on the reflections of guided waves that are excited and detected by magnetostrictive ultrasonic transducers. The Discrete Wavelet Transform was used to extract damage-sensitive features from the detected signals and to construct a multi-dimensional Damage Index vector. The Damage Index vector was then fed to an Artificial Neural Network to provide the automatic classification of (a) the size of the notch and (b) the location of the notch from the receiving sensor. Following an optimization study of the network, it was determined that five damage-sensitive features provided the best defect classification performance with an overall success rate of 90.8%. It was thus demonstrated that the wavelet-based multidimensional analysis can provide excellent classification performance for notch-type defects in strands.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2006.04a
• /
• pp.240-247
• /
• 2006

Piezoelectric sensor-based health monitoring technique using a two-step support vector machine (SYM) classifier is discussed for damage identification of a railroad track. An active sensing system composed of two PZT patches was investigated in conjunction with both impedance and guided wave propagation methods to detect two kinds of damage of the railroad track (one is a hole damage of 0.5cm in diameter at web section and the other is a transverse cut damage of 7.5cm in length and 0.5cm in depth at head section). Two damage-sensitive features were extracted one by one from each method; a) feature I: root mean square deviations (RMSD) of impedance signatures and b) feature II: wavelet coefficients for $A_0$ mode of guided waves. By defining damage indices from those damage-sensitive features, a two-dimensional damage feature (2-D DF) space was made. In order to minimize a false-positive indication of the current active sensing system, a two-step SYM classifier was applied to the 2-D DF space. As a result, optimal separable hyper-planes were successfully established by the two-step SYM classifier: Damage detection was accomplished by the first step-SYM, and damage classification was also carried out by the second step-SYM. Finally, the applicability of the proposed two-step SYM classifier has been verified by thirty test patterns.

• Journal of Korean Society for Quality Management
• /
• v.36 no.2
• /
• pp.36-44
• /
• 2008

In this paper a novel feature extraction and selection is carried out in order to improve the discriminating capability between healthy and damaged structure using vibration signals. Although many feature extraction and selection algorithms have been proposed for vibration signals, most proposed approaches don't consider the discriminating ability of features since they are usually in unsupervised manner. We proposed a novel feature extraction and selection algorithm selecting few wavelet coefficients with higher class discriminating capability for damage detection and class visualization. We applied three class separability measures to evaluate the features, i.e. T test statistics, divergence, and Bhattacharyya distance. Experiments with vibration signals from truss structure demonstrate that class separabilities are significantly enhanced using our proposed algorithm compared to other two algorithms with original time-based features and Fourier-based ones.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2001.05a
• /
• pp.821-826
• /
• 2001

This paper presents four types of strut-tie model design approaches of structural concrete through the anchorage zone design of a post-tensioned concrete I-beam. The differences and distinctive feature of each approach in terms of structural type of selected strut-tie model, external force acting on strut-tie model, effective strength of concrete strut, and strut-tie model design procedure are analyzed and compared. The outcomes of present study enable structural designers to understand the merits and demerits of each strut-tie model design approach, and thus to conduct reasonable and accurate design of structural concrete.

• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.673-687
• /
• 2023

Convolutional neural network-based deep learning technology is the most commonly used in image identification, but it requires large-scale data for training. Therefore, application in specific fields in which data acquisition is limited, such as in the military, may be challenging. In particular, the identification of ground weapon systems is a very important mission, and high identification accuracy is required. Accordingly, various studies have been conducted to achieve high performance using small-scale data. Among them, the ensemble method, which achieves excellent performance through the prediction average of the pre-trained models, is the most representative method; however, it requires considerable time and effort to find the optimal combination of ensemble models. In addition, there is a performance limitation in the prediction results obtained by using an ensemble method. Furthermore, it is difficult to obtain the ensemble effect using models with imbalanced classification accuracies. In this paper, we propose a transfer learning-based feature fusion technique for heterogeneous models that extracts and fuses features of pre-trained heterogeneous models and finally, fine-tunes hyperparameters of the fully connected layer to improve the classification accuracy. The experimental results of this study indicate that it is possible to overcome the limitations of the existing ensemble methods by improving the classification accuracy through feature fusion between heterogeneous models based on transfer learning.