• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.14 no.4
• /
• pp.249-255
• /
• 2014

We present a simple method to watermark three-dimensional (3D) triangular meshes that have been generated from the depth data of the Kinect sensor. In contrast to previous methods, which maintain the shape of 3D triangular meshes and decide the embedding place, requiring calculations of vertices and their neighbors, our method is based on selecting one of the coordinate axes. To maintain shape, we use discrete wavelet transform and constant regularization. We know that the watermarking system needs the information to be embedded; we used a text to provide that information. We used geometry attacks such as rotation, scales, and translation, to test the performance of this watermarking system. Performance parameters in this paper include the vertices error rate (VER) and bit error rate (BER). The results from the VER and BER indicate that using a correction term before the extraction process makes our system robust to geometry attacks.

• Journal of information and communication convergence engineering
• /
• v.21 no.2
• /
• pp.139-144
• /
• 2023

Engineers prefer deep neural networks (DNNs) for solving computer vision problems. However, DNNs pose two major problems. First, neural networks require large amounts of well-labeled data for training. Second, the covariate shift problem is common in computer vision problems. Domain adaptation has been proposed to mitigate this problem. Recent work on adversarial-learning-based unsupervised domain adaptation (UDA) has explained transferability and enabled the model to learn robust features. Despite this advantage, current methods do not guarantee the distinguishability of the latent space unless they consider class-aware information of the target domain. Furthermore, source and target examples alone cannot efficiently extract domain-invariant features from the encoded spaces. To alleviate the problems of existing UDA methods, we propose the mixup regularization in adversarial discriminative domain adaptation (ADDA) method. We validated the effectiveness and generality of the proposed method by performing experiments under three adaptation scenarios: MNIST to USPS, SVHN to MNIST, and MNIST to MNIST-M.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.1B
• /
• pp.141-147
• /
• 2000

In the restoration of degraded noisy motion blurred image, we have trade-off problem between smoothing the noise and restoration of the edge region. While the noise is smoothed, die edge or details will be corrupted. On the other hand, restoring the edge will amplify the noise. To solve this problem we propose an adaptive algorithm which uses I- H regularization operator for flat region and Laplacian regularization operator for edge region. Through the experiments, we verify that the proposed method shows better results in the suppression of the noise amplification in flat region, introducing less ringing artifacts in edge region and better ISNR than those of the conventional ones.

• Journal of the Semiconductor & Display Technology
• /
• v.18 no.3
• /
• pp.66-71
• /
• 2019

In shape from focus (SFF) methods, the quality of image focus volume plays a vital role in the quality of 3D shape reconstruction. Traditionally, a linear 2D filter is applied to each slice of the image focus volume to rectify the noisy focus measurements. However, this approach is problematic because it also modifies the accurate focus measurements that should ideally remain intact. Therefore, in this paper, we propose to enhance the focus volume adaptively by applying 3-dimensional weighted least squares (3D-WLS) based regularization. We estimate regularization weights from the guidance volume extracted from the image sequences. To solve 3D-WLS optimization problem efficiently, we apply a technique to solve a series of 1D linear sub-problems. Experiments conducted on synthetic and real image sequences demonstrate that the proposed method effectively enhances the image focus volume, ultimately improving the quality of reconstructed shape.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.4
• /
• pp.1724-1731
• /
• 2018

The conventional polynomial neural network (PNN) is a classical flexible neural structure and self-organizing network, however it is not free from the limitation of overfitting problem. In this study, we propose a space search-optimized polynomial neural network (ssPNN) structure to alleviate this problem. Ranking selection is realized by means of ranking selection-based performance index (RS_PI) which is combined with conventional performance index (PI) and coefficients based performance index (CPI) (viz. the sum of squared coefficient). Unlike the conventional PNN, L2-norm regularization method for estimating the polynomial coefficients is also used when designing the ssPNN. Furthermore, space search optimization (SSO) is exploited here to optimize the parameters of ssPNN (viz. the number of input variables, which variables will be selected as input variables, and the type of polynomial). Experimental results show that the proposed ranking selection-based polynomial neural network gives rise to better performance in comparison with the neuron fuzzy models reported in the literatures.

• Journal of the Korean Society of Radiology
• /
• v.11 no.7
• /
• pp.679-685
• /
• 2017

Computed tomography has widely been used to diagnose patient disease, and patient dose also increase rapidly. To reduce the patient dose by CT, various techniques have been applied. The iterative reconstruction is used in view of image reconstruction. Image quality of the reconstructed section image through algebraic reconstruction technique, one of iterative reconstruction methods, was examined by the normalized root mean square error. The computer program was written with the Visual C++ under the parallel beam geometry, Shepp-Logan head phantom of $512{\times}512$ size, projections of 360, and detector-pixels of 1,024. The forward and backward projection was realized by Joseph method. The minimum NRMS of 0.108 was obtained after 10 iterations in the regularization parameter of 0.09-0.12, and the optimum image was obtained after 8 and 6 iterations for 0.1% and 0.2% noise. Variation of optimum value of the regularization parameter was observed according to the phantom used. If the ART was used in the reconstruction, the optimal value of the regularization parameter should be found in the case-by-case. By finding the optimal regularization parameter in the algebraic reconstruction technique, the reconstruction time can be reduced.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.6
• /
• pp.886-898
• /
• 1995

One of the ways to measure the distance in the computer vision is to use the focus and defocus. There are two methods in this way. The first method is caculating the distance from the focused images in a point (MMDFP: the method measuring the distance to the focal plane). The second method is to measure the distance from the difference of the camera parameters, in other words, the apertures of the focal planes, of two images with having the different parameters (MMDCI: the method to measure the distance by comparing two images). The problem of the existing methods in MMDFP is to decide the thresholding vaue on detecting the most optimally focused object in the defocused image. In this case, it could be solved by comparing only the error energy in 3x3 window between two images. In MMDCI, the difficulty is the influence of the deflection effect. Therefor, to minimize its influence, we utilize two differently focused images instead of different aperture images in this paper. At the first, the amount of defocusing between two images is measured through the introduction of regularization and then the distance from the camera to the objects is caculated by the new equation measuring the distance. In the results of simulation, we see the fact to be able to measure the distance from two differently defocused images, and for our approach to be robuster than the method using the different aperture in the noisy image.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.13 no.1
• /
• pp.321-339
• /
• 2021

Response Surface Method (RSM) has been widely used for flammable cloud size prediction as it can reduce computational intensity for further Explosion Risk Analysis (ERA) especially during the early design phase of offshore platforms. However, RSM encounters the overfitting problem under very limited simulations. In order to overcome the disadvantage of RSM, Bayesian Regularization Artificial Neural (BRANN)-based model has been recently developed and its robustness and efficiency have been widely verified. However, for ERA during the early design phase, there seems to be room to further reduce the computational intensity while ensuring the model's acceptable accuracy. This study aims to develop an integrated method, namely the combination of Center Composite Design (CCD) method with Bayesian Regularization Artificial Neural Network (BRANN), for flammable cloud size prediction. A case study with constant and transient leakages is conducted to illustrate the feasibility and advantage of this hybrid method. Additionally, the performance of CCD-BRANN is compared with that of RSM. It is concluded that the newly developed hybrid method is more robust and computational efficient for ERAs during early design phase.

• Korean Journal of Remote Sensing
• /
• v.28 no.5
• /
• pp.561-571
• /
• 2012

The MTF(Modulation Transfer Function) is one of quality assesment factors to evaluate the performance of satellite images. Image restoration is needed for MTF compensation, but it is an ill-posed problem and doesn't have a certain solution. Lots of filters were suggested to solve this problem, such as Inverse Filter(IF), Pseudo Inverse Filter(PIF) and Wiener Filter(WF). The most commonly used filter is a WF, but it has a limitation on distinguishing signal and noise. The L-curve-based Modified Wiener Filter(MWF) is a solution technique using a Tikhonov regularization method. The L-curve is used for estimating an optimal regularization parameter. The image restoration was performed with Dubaisat-1 images for PIF, WF, and MWF. It is found that the image restored with MWF results in more improved MTF by 20.93% and 10.85% than PIF and WF, respectively.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.15 no.1
• /
• pp.299-309
• /
• 1995

A geometrical inverse heat conduction problem is solved for the development of Infrared Computerized-Axial-Tomography (IR CAT) Scan by using a boundary element method in conjunction with regularization procedure. In this problem, an overspecified temperature condition by infrared scanning is provided on the surface, and is used together with other conditions to solve the position of an unknown boundary (cavity). An auxiliary problem is introduced in the solution of this problem. By defining a hypothetical inner boundary for the auxiliary problem domain, the cavity is located interior to the domain and its position is determined by solving a potential problem. Boundary element method with regularization procedure is used to solve this problem, and the effects of regularization on the inverse solution method are investigated by means of numerical analysis.