• Smart Structures and Systems
• /
• v.32 no.3
• /
• pp.135-151
• /
• 2023

Using 3D asphalt pavement surface data, a deep and multiscale network named CrackNet-M is proposed in this paper for pixel-level crack detection for improvements in both accuracy and robustness. The CrackNet-M consists of four function-specific architectural modules: a central branch net (CBN), a crack map enhancement (CME) module, three pooling feature pyramids (PFP), and an output layer. The CBN maintains crack boundaries using no pooling reductions throughout all convolutional layers. The CME applies a pooling layer to enhance potential thin cracks for better continuity, consuming no data loss and attenuation when working jointly with CBN. The PFP modules implement direct down-sampling and pyramidal up-sampling with multiscale contexts specifically for the detection of thick cracks and exclusion of non-crack patterns. Finally, the output layer is optimized with a skip layer supervision technique proposed to further improve the network performance. Compared with traditional supervisions, the skip layer supervision brings about not only significant performance gains with respect to both accuracy and robustness but a faster convergence rate. CrackNet-M was trained on a total of 2,500 pixel-wise annotated 3D pavement images and finely scaled with another 200 images with full considerations on accuracy and efficiency. CrackNet-M can potentially achieve crack detection in real-time with a processing speed of 40 ms/image. The experimental results on 500 testing images demonstrate that CrackNet-M can effectively detect both thick and thin cracks from various pavement surfaces with a high level of Precision (94.28%), Recall (93.89%), and F-measure (94.04%). In addition, the proposed CrackNet-M compares favorably to other well-developed networks with respect to the detection of thin cracks as well as the removal of shoulder drop-offs.

• Analytical Science and Technology
• /
• v.29 no.6
• /
• pp.277-282
• /
• 2016

British Pharmacopoeia (BP 2013), the United States Pharmacopoeia (USP 39) and the Korean Pharmacopoeia (KP XI) contain monographs for the quality control of raw diosmin using high performance liquid chromatography (HPLC). However, official monographs detailing pharmaceutical formulations for diosmin are not available in foreign pharmacopoeias. In the KP XI, ultraviolet-visible (UV-Vis) spectroscopy-which is less specific than HPLC-is reported for the testing of diosmin capsules. In this study, we present an alternative HPLC assay for such testing that is more specific than UV-Vis methods. Method validation was performed to determine linearity, precision, accuracy, system suitability, and robustness. The linearity of calibration curves in the desired concentration range was high ($r^2$>0.999), while the RSDs for intra- and inter-day precision were 0.15-0.29 % and 1.05-1.74%, respectively. Accuracies ranged from 101.2-103.2 %, while the retention time and peak area RSDs were 0.37 % and 0.06 %, respectively. Additionally, the plate number and asymmetry factor values for diosmin were 3591.293 and 1.35, respectively. Since the intermediate-precision and robustness of the assay were satisfactory, this method will be a valuable addition to the Korean Pharmacopoeia (KP XI).

• Journal of Computing Science and Engineering
• /
• v.7 no.2
• /
• pp.99-111
• /
• 2013

Approximating a divergence between two probability distributions from their samples is a fundamental challenge in statistics, information theory, and machine learning. A divergence approximator can be used for various purposes, such as two-sample homogeneity testing, change-point detection, and class-balance estimation. Furthermore, an approximator of a divergence between the joint distribution and the product of marginals can be used for independence testing, which has a wide range of applications, including feature selection and extraction, clustering, object matching, independent component analysis, and causal direction estimation. In this paper, we review recent advances in divergence approximation. Our emphasis is that directly approximating the divergence without estimating probability distributions is more sensible than a naive two-step approach of first estimating probability distributions and then approximating the divergence. Furthermore, despite the overwhelming popularity of the Kullback-Leibler divergence as a divergence measure, we argue that alternatives such as the Pearson divergence, the relative Pearson divergence, and the $L^2$-distance are more useful in practice because of their computationally efficient approximability, high numerical stability, and superior robustness against outliers.

• Smart Structures and Systems
• /
• v.32 no.5
• /
• pp.319-338
• /
• 2023

The study presents a new hybrid data-driven method by combining radial basis functions neural networks (RBF-NN) with the Jaya algorithm (JA) to provide effective structural health monitoring of arch dams. The novelty of this approach lies in that only one user-defined parameter is required and thus can increase its effectiveness and efficiency, as compared to other machine learning techniques that often require processing a large amount of training and testing model parameters and hyper-parameters, with high time-consuming. This approach seeks rapid damage detection in arch dams under dynamic conditions, to prevent potential disasters, by utilizing the RBF-NNN to seamlessly integrate the dynamic elastic modulus (DEM) and modal parameters (such as natural frequency and mode shape) as damage indicators. To determine the dynamic characteristics of the arch dam, the JA sequentially optimizes an objective function rooted in vibration-based data sets. Two case studies of hyperbolic concrete arch dams were carefully designed using finite element simulation to demonstrate the effectiveness of the RBF-NN model, in conjunction with the Jaya algorithm. The testing results demonstrated that the proposed methods could exhibit significant computational time-savings, while effectively detecting damage in arch dam structures with complex nonlinearities. Furthermore, despite training data contaminated with a high level of noise, the RBF-NN and JA fusion remained the robustness, with high accuracy.

• KSBB Journal
• /
• v.29 no.5
• /
• pp.361-371
• /
• 2014

Mycoplasma is well recognized as one of the most prevalent and serious microbial contaminants of biologic manufacturing processes. Conventional methods for mycoplasma testing, direct culture method and indirect indicator cell culture method, are lengthy, costly and less sensitive to noncultivable species. In this report, we describe a new TaqMan probe-based real-time PCR method for rapid and quantitative detection of mycoplasma contamination during manufacture of biologics. Universal mycoplasma primers were used for mycoplasma PCR and mycoplasma DNA was quantified by use of a specific TaqMan probe. Specificity, sensitivity, and robustness of the real-time PCR method was validated according to the European Pharmacopoeia. The validation results met required criteria to justify its use as a replacement for the culture method. The established real-time PCR assay was successfully applied to the detection of mycoplasma from human keratinocyte and mesenchymal stem cell as well as Vero cell lines artificially infected with mycoplasma. The overall results indicated that this rapid, specific, sensitive, and robust assay can be reliably used for quantitative detection of mycoplasma contamination during manufacture of biologics.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.17 no.9
• /
• pp.2242-2251
• /
• 1993

The mesh-based frictional contact model has been developed which does not rely on the spatial derivatives of the tool surface. Only points on the surface are evaluated from the description. which can then be simplified because of the relaxed demands placed on it. The surface tangents, normals, and corresponding derivatives at each finite-element node are evaluated directly from the finite-element mesh, in terms of the connecting nodal positions. The advantages accrue because there is no longer a need for a smooth tool surface to assure reasonable normals and derivatives. Furthermore, it can be shown that the equilibrium equations can only be properly written with a special normal derived from the mesh itself. The validity, accuracy, computation time, and stability of mesh-based contact model were discussed with the numerical examples of rounded flat-top and rough, flat-top rounded punch forming operations. Also, the forming process of a automobile inner panel section was simulated for testing the robustness of new contact model. In the discussion, the superiority of new model was examined, comparing with tool-based contact one.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.283-286
• /
• 2001

In this paper, we present a robust region segmentation method based on the watershed transformation of a pseudo-distance map (PDM). A usual approach for the segmentation of a gray-scale image with the watershed algorithm is to apply it to a gradient magnitude image or the Euclidean distance map (EDM) of an edge image. However, it is well known that this approach suffers from the oversegmentation of the given image due to noisy gradients or spurious edges caused by a thresholding operation. In this paper we show thor applying the watershed algorithm to the EDM, which is a regularized version of the EDM and is directly computed form the edgestrength function (ESF) of the input image, significantly reduces the oversegmentation, and the final segmentation results obtained by a simple region-merging process are more reliable and less noisy than those of the gradient-or EDM-based methods. We also propose a simple and efficient region-merging criterion considering both boundary strengths and inner intensities of regions to be merged. The robustness of our method is proven by testing it with a variety of synthetic and real images.

• Smart Structures and Systems
• /
• v.17 no.3
• /
• pp.523-539
• /
• 2016

The ambient vibration measurement is an output-data-only dynamic testing where natural excitations are represented, for instance, by winds and typhoons. The modal identification involving output-only measurements requires the use of specific modal identification techniques. This paper presents the application of a reliable method (the Stochastic Subspace Identification - SSI) implemented in a general purpose software. As a criterion toward the robustness of identified modes, a bio-inspired optimization algorithm, with a highly nonlinear objective function, is introduced in order to find the optimal deployment of a reduced number of sensors across a large civil engineering structure for the validation of its modal identification. The Ting Kau Bridge (TKB), one of the longest cable-stayed bridges situated in Hong Kong, is chosen as a case study. The results show that the proposed method catches eigenvalues and eigenvectors even for a reduced number of sensors, without any significant loss of accuracy.

• The KIPS Transactions:PartD
• /
• v.14D no.6
• /
• pp.649-656
• /
• 2007

Today, controller area network(CAN) is a field bus that is nowadays widespread in distributed embedded systems due to its electrical robustness, low price, and deterministic access delay. However, its use safety-critical applications has been controversial due to dependability limitation, such as those arising from its bus topology. Thus it is important to analyze the performance of the network in terms of load of data bus, maximum time delay, communication contention, and others during the design phase of the controller area network. In this paper, a simulation algorithm is introduced to evaluate the communication performance of the vehicle network and apply software base fault injection techniques. This can not only reduce any erratic implementation of the vehicle network but it also improves the reliability of the system.

• Journal of Advanced Marine Engineering and Technology
• /
• v.24 no.3
• /
• pp.88-98
• /
• 2000

In this paper, a robust controller using $H_{\infty}$ control theory has been designed for the load frequency control of interconnected 2-area power system. The main advantage of the proposed $H_{\infty}$ controller is that uncertainties of power system can be included at the stage of controller design. Representation of uncertainties is modeled by multiplicative uncertainly. In the mixed sensitivity problems, disturbance attenuation and uncertainty of the system is treated simultaneously. The robust stability and the performance of model uncertainties are represented by frequency weighted transfer function. The design of load frequency controller for each area was based on state-space approach. The comparative computer simulation results for the proposed controller and the conventional techniques such as the optimal control and the PID one were analyzed at the additions of various disturbances. Their deviation magnitude of frequency and tie line power flow at each area were mainly evaluated. Also the testing results of robustness for the cases that the perturbations of the all parameters of power system were amounted to about 20% were introduced. It was approved that the resultant performances of the proposed $H_{\infty}$ controller with mixed sensitivity were more robust and stable than the one of conventional controllers.