• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.12
• /
• pp.6028-6042
• /
• 2019

Since Sparse Nonnegative Matrix Factorization (SNMF) method can control the sparsity of the decomposed matrix, and then it can be adopted to control the sparsity of facial feature extraction and recognition. In order to improve the accuracy of SNMF method for facial feature recognition, new additive iterative rules based on the improved iterative step sizes are proposed to improve the SNMF method, and then the traditional multiplicative iterative rules of SNMF are transformed to additive iterative rules. Meanwhile, to further increase the sparsity of the basis matrix decomposed by the improved SNMF method, a threshold-sparse constraint is adopted to make the basis matrix to a zero-one matrix, which can further improve the accuracy of facial feature recognition. The improved SNMF method based on the additive iterative rules and threshold-sparse constraint is abbreviated as ASNMF, which is adopted to recognize the ORL and CK+ facial datasets, and achieved recognition rate of 96% and 100%, respectively. Meanwhile, from the results of the contrast experiments, it can be found that the recognition rate achieved by the ASNMF method is obviously higher than the basic NMF, traditional SNMF, convex nonnegative matrix factorization (CNMF) and Deep NMF.

• Smart Structures and Systems
• /
• v.29 no.6
• /
• pp.767-775
• /
• 2022

Metal additive manufacturing (AM), also known as metal three-dimensional (3D) printing, produces 3D metal products by repeatedly adding and solidifying metal materials layer by layer. During the metal AM process, products experience repeated local melting and cooling using a laser or electron beam, resulting in product defects, such as warpage, cracks, and internal pores. Such defects adversely affect the final product. This paper proposes the in situ monitoring-based warpage prediction of metal AM products with experimental feature extraction. The temperature profile of the metal AM substrate during the process was experimentally collected. Time-domain features were extracted from the temperature profile, and their relationships to the warpage mechanism were investigated. The standard deviation showed a significant linear correlation with warpage. The findings from this study are expected to contribute to optimizing process parameters for metal AM warpage reduction.

• Speech Sciences
• /
• v.11 no.1
• /
• pp.7-20
• /
• 2004

In this paper, we propose a two-step noise compensation algorithm in feature extraction for achieving robust speech recognition. The proposed method frees us from requiring a priori information on noisy environments and is simple to implement. First, in frequency domain, the Harmonics-based Spectral Subtraction (HSS) is applied so that it reduces the additive background noise and makes the shape of harmonics in speech spectrum more pronounced. We then apply a judiciously weighted variance Feature Vector Normalization (FVN) to compensate for both the channel distortion and additive noise. The weighted variance FVN compensates for the variance mismatch in both the speech and the non-speech regions respectively. Representative performance evaluation using Aurora 2 database shows that the proposed method yields 27.18% relative improvement in accuracy under a multi-noise training task and 57.94% relative improvement under a clean training task.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.828-831
• /
• 2005

In feature-based multi-resolution modeling, the features are rearranged according to a criterion for the levels of detail (LOD) of multi-resolution models. In this paper, two different LOD criteria are investigated and discussed. The one is the volumes of subtractive features, together with the precedence of additive features over subtractive features. The other is the volumes of features, regardless of whether the feature types are subtractive or additive. In addition, the algorithms to define and extract the LOD models based on the criteria are also described. The criterion of the volumes of features can be used for a wide range of applications in CAD and CAE in virtue of its generality.

• ETRI Journal
• /
• v.26 no.3
• /
• pp.273-276
• /
• 2004

We propose a novel feature processing technique which can provide a cepstral liftering effect in the log-spectral domain. Cepstral liftering aims at the equalization of variance of cepstral coefficients for the distance-based speech recognizer, and as a result, provides the robustness for additive noise and speaker variability. However, in the popular hidden Markov model based framework, cepstral liftering has no effect in recognition performance. We derive a filtering method in log-spectral domain corresponding to the cepstral liftering. The proposed method performs a high-pass filtering based on the decorrelation of filter-bank energies. We show that in noisy speech recognition, the proposed method reduces the error rate by 52.7% to conventional feature.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.54 no.6
• /
• pp.417-422
• /
• 2005

This paper presents an EMG pattern classification method to identify motion commands for the control of the artificial arm by SOM-TVC(self organizing map - tracking Voronoi cell) based on neural network with a feature parameter. The eigenvalue is extracted as a feature parameter from the EMG signals and Voronoi cells is used to define each pattern boundary in the pattern recognition space. And a TVC algorithm is designed to track the movement of the Voronoi cell varying as the condition of additive noise. Results are presented to support the efficiency of the proposed SOM-TVC algorithm for EMG pattern recognition and compared with the conventional EDM and BPNN methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.3
• /
• pp.613-620
• /
• 2010

Additive noise and channel distortion strongly degrade the performance of speaker verification systems, as it introduces distortion of the features of speech. This distortion causes a mismatch between the training and recognition conditions such that acoustic models trained with clean speech do not model noisy and channel distorted speech accurately. This paper presents a phase-related feature extraction method in order to improve the robustness of the speaker verification systems. The instantaneous frequency is computed from the phase of speech signals and features from the histogram of the instantaneous frequency are obtained. Experimental results show that the proposed technique offers significant improvements over the standard techniques in both clean and adverse testing environments.

• Journal of Mechanical Science and Technology
• /
• v.19 no.2
• /
• pp.558-566
• /
• 2005

We propose a feature-based multi-resolution representation of B-rep solid models using history-based Boolean operations based on the merge-and-select algorithm. Because union and subtraction are commutative in the history-based Boolean operations, the integrity of the models at various levels of detail (LOD) is guaranteed for the reordered features regardless of whether the features are subtractive or additive. The multi-resolution solid representation proposed in this paper includes a non-manifold topological merged-set model of all feature primitives as well as a feature-modeling tree reordered consistently with a given LOD criterion. As a result, a B-rep solid model for a given LOD can be provided quickly, because the boundary of the model is evaluated without any geometric calculation and extracted from the merged set by selecting the entities contributing to the LOD model shape.

• Korean Journal of Computational Design and Engineering
• /
• v.7 no.2
• /
• pp.121-130
• /
• 2002

The concept of Level Of Detail (LOD) was introduced and has been used to enhance display performance and to carry out certain engineering analysis effectively. We would like to use an adequate complexity level for each geometric model depending on specific engineering needs and purposes. Solid modeling systems are widely used in industry, and are applied to advanced applications such as virtual assembly. In addition, as the demand to share these engineering tasks through networks is emerging, the problem of building a solid model of an appropriate resolution to a given application becomes a matter of great necessity. However, current researches are mostly focused on triangular mesh models and various operators to reduce the number of triangles. So we are working on the multi-resolution of the solid model itself, rather than that of the triangular mesh model. In this paper, we propose multi-resolution representation of B-rep model by reordering and converting design features into an enclosing volume and subtractive features.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.544-551
• /
• 2023

Speech enhancement used to improve the perceptual quality and intelligibility of noise speech has been studied as a method using a complex-valued spectrum that can improve both magnitude and phase in a method using a magnitude spectrum. In this paper, a study was conducted on how to apply attention mechanism to complex-valued spectrum-based speech enhancement systems to further improve the intelligibility and quality of noise speech. The attention is performed based on additive attention and allows the attention weight to be calculated in consideration of the complex-valued spectrum. In addition, the global average pooling was used to consider the importance of the feature map. Complex-valued spectrum-based speech enhancement was performed based on the Deep Complex U-Net (DCUNET) model, and additive attention was conducted based on the proposed method in the Attention U-Net model. The results of the experiments on noise speech in a living room environment showed that the proposed method is improved performance over the baseline model according to evaluation metrics such as Source to Distortion Ratio (SDR), Perceptual Evaluation of Speech Quality (PESQ), and Short Time Object Intelligence (STOI), and consistently improved performance across various background noise environments and low Signal-to-Noise Ratio (SNR) conditions. Through this, the proposed speech enhancement system demonstrated its effectiveness in improving the intelligibility and quality of noisy speech.