• Phonetics and Speech Sciences
• /
• v.6 no.4
• /
• pp.127-132
• /
• 2014

This paper describes a robust speech recognition technique by reconstructing spectral components mismatched with a training environment. Although the cluster-based reconstruction method can compensate the unreliable components from reliable components in the same spectral vector by assuming an independent, identically distributed Gaussian-mixture process of training spectral vectors, the presented method exploits the temporal dependency of speech to reconstruct the components by introducing a hidden-Markov-model prior which incorporates an internal state transition plausible for an observed spectral vector sequence. The experimental results indicate that the described method can provide temporally consistent reconstruction and further improve recognition performance on average compared to the conventional method.

• Journal of Institute of Convergence Technology
• /
• v.4 no.2
• /
• pp.1-9
• /
• 2014

This paper examines feature-based reconstruction algorithm using feature-based modeling and based on topology optimization technology, which aims to achieve a minimal volume weight and to satisfy user-defined constraints such as stress, deformation related conditions. The finite element model after topology optimization allows us to remove some region of a solid model for predefined volume requirement. The stress or deformation distribution resulted from finite element analysis enables us to add some material to the solid model for a robust structure. For this purpose, we propose a feature-based redesign algorithm which inserts negative features to the solid model for material removal and positive features for material addition, and we introduce a bisection method which searches an optimal structure by iteratively applying the feature-based redesign algorithm. Several examples are considered to illustrate the proposed algorithms and to demonstrate the effectiveness of the present approach.

• Journal of Biomedical Engineering Research
• /
• v.18 no.4
• /
• pp.365-373
• /
• 1997

In this paper, a new method of the 3D(dimensional) image reconstruction is proposed to build up the 3D image from 2D images using digital image processing techniques and computer graphics. First, the new feature extraction algorithm that doesn't need various input parameters and is not affected by threshold is adopted This new algorithm extracts feature points by eliminating some undesirable points on the ground of the connectivity. Second, as the cast function to reconstruct surfaces using extracted feature points, the minimum distance measure between two plane images has been adopted According to this measure, the surface formation algorithm doesn't need complex calculation and takes the form of triangle or trapezoid To investigate usefulness, this approach has been applied to a head CT image and compared with other methods. Experimental comparisons show that the suggested algorithm yields better performance on feature extraction than others. In contrast with the other methods, the complex calculation for surface formation in the proposed algorithm is not necessary.

• Journal of Biomedical Engineering Research
• /
• v.15 no.2
• /
• pp.201-208
• /
• 1994

Many three-dimensional object modeling and display methods for computer graphics and computer vision have been developed. Recently, with the help of medical imaging devices such as computerized tomography, magnetic resonance image, etc., some of those object modeling and display methods have been widely used for capturing the shape, structure and other properties of real objects in many medical applications. In this paper, we propose the reconstruction and display method of the three-dimensional object from a series of the cross sectonal image. It is implemented by using the automatic threshold selection method and the contour following algorithm. The combination of curvature and distance, we select feature points. Those feature points are the candidates for the tiling method. As a results, it is proven that this proposed method is very effective and useful in the comprehension of the object's structure. Without the technician's responce, it can be automated.

• Journal of Ubiquitous Convergence Technology
• /
• v.1 no.1
• /
• pp.9-17
• /
• 2007

Feature based stereo matching is an effective way to perform 3D building reconstruction. However, in urban scene, the cluttered background and various building structures may interfere with the performance of building reconstruction. In this paper, we propose a novel method to robustly reconstruct buildings on the basis of rectangle regions. Firstly, we propose a multi-scale linear feature detector to obtain the salient line segments on the object contours. Secondly, candidate rectangle regions are extracted from the salient line segments based on their local information. Thirdly, stereo matching is performed with the list of matching line segments, which are boundary edges of the corresponding rectangles from the left and right image. Experimental results demonstrate that the proposed method can achieve better accuracy on the reconstructed result than pixel-level stereo matching.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.2
• /
• pp.78-87
• /
• 2006

In this paper. we propose an effective mask estimation scheme for missing-feature reconstruction in order to achieve robust speech recognition under unknown noise environments. In the previous work. colored noise is used for training the mask classifer, which is generated from the entire frequency Partitioned signals. However it gives a limited performance under the restricted number of training database. To reflect the spectral events of more various background noise and improve the performance simultaneously. a new Bayesian classifier for mask estimation is proposed, which works independent of other frequency bands. In the proposed method, we employ the colored noise which is obtained by combining colored noises generated from each frequency band in order to reflect more various noise environments and mitigate the 'sparse' database problem. Combined with the cluster-based missing-feature reconstruction. the performance of the proposed method is evaluated on a task of noisy speech recognition. The results show that the proposed method has improved performance compared to the Previous method under white noise. car noise and background music conditions.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.169-171
• /
• 2019

This paper proposes an algorithm to detect persons in bitstream domain by skipping a reconstruction picture process in HEVC decoding. A new 3-channel feature extraction map is introduced in this paper by modelling the relationship between bits per CU density, average PU shape in CU, and total transform coefficients in CU from syntax elements. A state-of-the-art of YOLOv3 detection algorithm is used to detect and localize person on extracted feature maps. Based on the experimental results, the proposed person detection framework can achieve mAP of 0.68 and be able to find persons on feature maps. In addition, the proposed person detection can save decoding time about 60% by removing reconstruction picture process.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.2
• /
• pp.544-564
• /
• 2022

Underwater images often suffer from color distortion, blurring and low contrast, which is caused by the propagation of light in the underwater environment being affected by the two processes: absorption and scattering. To cope with the poor quality of underwater images, this paper proposes a multiscale fusion underwater image enhancement method based on channel attention mechanism and local binary pattern (LBP). The network consists of three modules: feature aggregation, image reconstruction and LBP enhancement. The feature aggregation module aggregates feature information at different scales of the image, and the image reconstruction module restores the output features to high-quality underwater images. The network also introduces channel attention mechanism to make the network pay more attention to the channels containing important information. The detail information is protected by real-time superposition with feature information. Experimental results demonstrate that the method in this paper produces results with correct colors and complete details, and outperforms existing methods in quantitative metrics.

• Journal of the Korea Computer Graphics Society
• /
• v.23 no.1
• /
• pp.57-65
• /
• 2017

In this paper, we present a dictionary learning method for reducing errors in point cloud models and reconstructing their geometry. For this, 3D feature information is extracted from the models which have a similar shape characteristic as the target model. Then a dictionary is constructed and the geometry is reconstructed using the dictionary. The presented method in this paper consists of the following three steps. First, a geometric patch is constructed from a similar model. Second, a morphological 3D feature of the acquired patch is learned. Third, a geometry reconstruction is performed using the learned dictionary. Finally, the error between the original model and the reconstruction result is calculated, and the accuracy of the reconstruction result is checked.

• The KIPS Transactions:PartB
• /
• v.17B no.3
• /
• pp.183-188
• /
• 2010

A factorization-based 3D reconstruction system is realized to recover 3D scene from an image sequence. The image sequence is captured from uncalibrated perspective camera from several views. Many matched feature points over all images are obtained by feature tracking method. Then, these data are supplied to the 3D reconstruction module to obtain the projective reconstruction. Projective reconstruction is converted to Euclidean reconstruction by enforcing several metric constraints. After many triangular meshes are obtained, realistic reconstruction of 3D models are finished by texture mapping. The developed system is implemented in C++, and Qt library is used to implement the system user interface. OpenGL graphics library is used to realize the texture mapping routine and the model visualization program. Experimental results using synthetic and real image data are included to demonstrate the effectiveness of the developed system.