• Journal of Broadcast Engineering
• /
• v.1 no.1
• /
• pp.7-13
• /
• 1996

MASF is a kind of temporal filter proposed for noise reduction and temporal band limitation. MASF uses motion vectors to extract temporal information in spatial domain. Therefore, inaccurate motion information causes some distortions in MASF operation. Currently, bilinear interpolation after BMA(Block Matching Algorithm) is used for the motion estimation sheme of MASF. But, this method results in unreliable estimation when the object in image sequence has larger movement than the maximum displacement assumed in BMA or the input images are severely corrupted with noise. In order to i:;olve this problem, we analyse the effect of inaccurate motion on MASF and propose a hierarchical motion estimation algorithm based on the analysis results. Experimental results show that the proposed method produces reliable output under large motion and noisy situations.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.43 no.8
• /
• pp.748-756
• /
• 2015

In this paper, we developed a miniaturized real time video compression system satisfying the military environment using ADV212 and FPGA. We present an efficient hardware design scheme for the weight reduction of the device and also a software solution to deal with noisy image signals. Experimental results show that the frame delay is reduced by a factor of 2 or 3 and the device's weight is decreased by a factor of 6 to 7. In order to prove the reliability for the military usage of this development, we examine the environmental test (MIL-STD-810G) and EMI test (MIL-STD-461F). Experimental results show that the developed system satisfies the requirements.

• Journal of Biomedical Engineering Research
• /
• v.16 no.4
• /
• pp.425-430
• /
• 1995

The correlation technique has been widely used in ctRl data processing. The proposed CLT (clus- tering threshold) technique is a modified CCT (correlation coefficient threshold) technique and has many advantages compared with the conventional CCT technique. The CLT technique is explained by the following two steps. First, once the correlation coefficient map above the proper TH value is obtained using the CCT technique which is discrete and includes splash noise data, then the spurious pixels are rejected and the real neural activity pixels extracted using an nxn matrix box. Second, a clustering operation is performed by the two correction rules. The real neuronal activated pixels can be clustered and the false spurious pixels can be suppressed by the proposed CLT technique. The proposed CLT technique used in the post processing in ctRl has advantages over other existing techniques. It is especially proved to be robust in noisy environment.

• Journal of Digital Convergence
• /
• v.14 no.8
• /
• pp.233-243
• /
• 2016

Due to speech recognition problems in noisy environment, Audio Visual Speech Recognition (AVSR) system, which combines speech information and visual information, has been proposed since the mid-1990s,. and lip reading have played significant role in the AVSR System. This study aims to enhance recognition rate of utterance word using only lip shape detection for efficient AVSR system. After preprocessing for lip region detection, Convolution Neural Network (CNN) techniques are applied for utterance period detection and lip shape feature vector extraction, and Hidden Markov Models (HMMs) are then used for the recognition. As a result, the utterance period detection results show 91% of success rates, which are higher performance than general threshold methods. In the lip reading recognition, while user-dependent experiment records 88.5%, user-independent experiment shows 80.2% of recognition rates, which are improved results compared to the previous studies.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.37 no.3
• /
• pp.98-109
• /
• 2000

While the bispectrum has desirable properties in itself and therefore has a lot of potential to be applied to signal and Image restoration. few real-world application results have appeared in literature The major problem with this IS the difficulty In realizing the expectation operator of the true bispectrum, due to the lack of realizations. In this paper, the true bispectrum is defined as the expectation of the sample bispectrum, which IS the Fourier representation of the triple correlation given one realization The characteristics of the sample bispectrum are analyzed and a way to obtain an estimate of the true bispectrum without stochastic expectation, using the generalized theory of weighted regularization is shown. The bispectrum estimated by the proposed algorithm is experimentally demonstrated to be useful for signal recovery under blurred noisy condition.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.2
• /
• pp.99-105
• /
• 2017

Generally, stereo matching methods are used to estimate depth information based on color and spatial similarity. However, most depth estimation methods suffer from the occlusion region because occlusion regions cause inaccurate depth information. Moreover, they do not consider the temporal dimension when estimating the disparity. In this paper, we propose a temporal stereo matching method, considering occlusion and disregarding inaccurate temporal depth information. First, we apply a global stereo matching algorithm to estimate the depth information, we segment the image to occlusion and non-occlusion regions. After occlusion detection, we fill the occluded region with a reasonable disparity value that are obtained from neighboring pixels of the current pixel. Then, we apply a temporal disparity estimation method using the reliable information. Experimental results show that our method detects more accurate occlusion regions, compared to a conventional method. The proposed method increases the temporal consistency of estimated disparity maps and outperforms per-frame methods in noisy images.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.28 no.1
• /
• pp.85-93
• /
• 2018

Biometric user authentications, day after day, propagate more to human life instead of traditional systems which use passwords and ID cards. However, most of these systems have many problems for given biometric information such noisy data, low-quality data, a limitation of recognition rate, and so on. To deal with these problems, I used Dual QML which is non-linear classification for classifying correctly the real-world data and then proposed preprocessing method for increasing recognition rate and performance by segmenting a specific region on an image. The previous published Dual QML used face, palmprint, ear for the experiment. In this paper, I used iris for experiment and then proved excellence of Dual QML at iris recognition. Finally I demonstrated these results (e.g. increasing recognition rate and performance, suitability for iris recognition) through experiments.

• Korean Journal of Remote Sensing
• /
• v.23 no.4
• /
• pp.273-285
• /
• 2007

Most stereo matching methods use intensity values in small image patches to measure the correspondence between two points. If the noisy pixels are used in computing the corresponding point, the matching performance becomes low. For this reason, the noise plays a critical role in determining the matching performance. In this paper, we propose a method for combining intensity and edge filters robust to the noise in order to improve the performance of stereo matching using high resolution satellite imagery. We used intensity filters such as Mean, Median, Midpoint and Gaussian filter and edge filters such as Gradient, Roberts, Prewitt, Sobel and Laplacian filter. To evaluate the performance of intensity and edge filters, experiments were carried out on both synthetic images and satellite images with uniform or gaussian noise. Then each filter was ranked based on its performance. Among the intensity and edge filters, Median and Sobel filter showed best performance while Midpoint and Laplacian filter showed worst result. We used Ikonos satellite stereo imagery in the experiments and the matching method using Median and Sobel filter showed better matching results than other filter combinations.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.9
• /
• pp.99-112
• /
• 2019

Cracks on pavement surfaces are critical signs and symptoms of the degradation of pavement structures. Image-based pavement crack detection is a challenging problem due to the intensity inhomogeneity, topology complexity, low contrast, and noisy texture background. In this paper, we address the problem of pavement crack detection and segmentation at pixel-level based on a Deep Neural Network (DNN) using gray-scale images. We propose a novel DNN architecture which contains a modified U-net network and a high-level features network. An important contribution of this work is the combination of these networks afforded through the fusion layer. To the best of our knowledge, this is the first paper introducing this combination for pavement crack segmentation and detection problem. The system performance of crack detection and segmentation is enhanced dramatically by using our novel architecture. We thoroughly implement and evaluate our proposed system on two open data sets: the Crack Forest Dataset (CFD) and the AigleRN dataset. Experimental results demonstrate that our system outperforms eight state-of-the-art methods on the same data sets.

• Smart Structures and Systems
• /
• v.33 no.6
• /
• pp.399-414
• /
• 2024

This study discusses the long-term deformation monitoring and shape sensing of bridge girder surfaces with an automated extraction scheme for point clouds in the Region Of Interest (ROI), invariant to the position of a Light Detection And Ranging system (LiDAR). Advanced smart construction necessitates continuous monitoring of the deformation and shape of bridge girders during the construction phase. An automated scheme is proposed for reconstructing geometric model of ROI in the presence of noisy non-stationary background. The proposed scheme involves (i) denoising irrelevant background point clouds using dimensions from the design model, (ii) extracting the outer boundaries of the bridge girder by transforming and processing the point cloud data in a two-dimensional image space, (iii) extracting topology of pre-defined targets using the modified Otsu method, (iv) registering the point clouds to a common reference frame or design coordinate using extracted predefined targets placed outside ROI, and (v) defining the bounding box in the point clouds using corresponding dimensional information of the bridge girder and abutments from the design model. The surface-fitted reconstructed geometric model in the ROI is superposed consistently over a long period to monitor bridge shape and derive deflection during the construction phase, which is highly correlated. The proposed scheme of combining 2D-3D with the design model overcomes the sensitivity of 3D point cloud registration to initial match, which often leads to a local extremum.