Wei-Chien Wang

National Central University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Ho Woo Lee

Chaoyang University of Technology

Yung-Chih Wang

National Defense Medical Center

Liang-Bi Chen

Hunan Normal University

Xiang-Rui Huang

National Penghu University of Science and Technology

Jia-Chen Xue

National Central University

Yishuo Huang

Chaoyang University of Technology

Yu Tsao

Research Center for Information Technology Innovation, Academia Sinica

Yu-Cheng Kan

Chaoyang University of Technology

Hoang Trung Hieu Duong

National Central University

Chung-Hao Wu

National Kaohsiung University of Science and Technology

Cooperative Institutions

Chaoyang University of Technology

National Central University

National Taiwan University

Chung Yuan Christian University

National Cheng Kung University

National Tsing Hua University

Southern Taiwan University of Science and Technology

National Yang Ming Chiao Tung University

National Penghu University of Science and Technology

Tunghai University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Mechanical Properties of High-Strength Pervious Concrete with Steel Fiber or Glass Fiber

Buildings (2022)

Ho Woo Lee Wei-Chien Wang Yung-Chih Wang Yi‐Cheng Hsieh Yung‐Chih Lin

Pervious concrete (also called porous concrete) is one of the most promising sustainable and green building materials today. This study examined high-strength pervious concrete and ordinary-strength pervious concrete reinforced with steel fiber or glass fiber. A total of fifteen mixtures of normal- and high-strength pervious concretes with steel fiber or glass fiber were used. The goal of high-strength pervious concrete is that the 28-day compressive strength be above 42 MPa and the porosity be as close to 15% as possible to achieve technical specifications. Both normal- and high-strength pervious concretes reinforced with steel fiber (1%, 2%) or glass fiber (0.25%, 0.5%) were investigated in water permeability, porosity, compressive strength, flexural strength, elastic modulus, and toughness tests. The test results show that in both high-strength pervious concrete and ordinary pervious concrete with steel fibers added, the porosity and permeability coefficient are increased compared with the control group. The coefficient of permeability for high-strength, fiber-reinforced pervious concretes with two aggregate sizes meets the requirements of the ACI specification for structural concrete. In addition, the high-strength pervious concrete specimen H1-S2 (2% steel fiber) has the highest compressive strength of 52.8 MPa at the age of 28 days. The flexural strength of pervious concrete also increases with age. However, the flexural strength of fiber-reinforced pervious concrete did not follow this trend due to the large variation in the quality control of different fiber mixtures. However, both steel fiber and glass fiber have a certain degree of improvement in the flexural toughness, and the effect is better with steel fiber. After the flexural strength reaches the peak value, there is still about 30% of the bearing capacity, and it gradually decreases until it is completely destroyed.

Pervious concrete

Fiber Reinforced Concrete

10.3390/buildings12050620

Cite

Citations (28)

A Novel Temporal Attentive-Pooling based Convolutional Recurrent Architecture for Acoustic Signal Enhancement

IEEE Transactions on Artificial Intelligence (2022)

Tassadaq Hussain Wei-Chien Wang Mandar Gogate Kia Dashtipour Yu Tsao

Removing background noise from acoustic observations to obtain clean signals is an important research topic regarding numerous real acoustic applications. Owing to their strong model capacity in function mapping, deep neural network-based algorithms have been successfully applied in target signal enhancement in acoustic applications. As most target signals carry semantic information encoded in a hierarchal structure in short- and long-term contexts, noise may distort such structures nonuniformly. In most deep neural network-based algorithms, such local and global effects are not explicitly considered in a modeling architecture for signal enhancement. In this article, we propose a temporal attentive pooling (TAP) mechanism combined with a conventional convolutional recurrent neural network (CRNN) model, called TAP-CRNN, which explicitly considers both global and local information for acoustic signal enhancement (ASE). In the TAP-CRNN model, we first use a convolution layer to extract local information from acoustic signals and a recurrent neural network (RNN) architecture to characterize temporal contextual information. Second, we exploit a novel attention mechanism to contextually process salient regions of noisy signals. We evaluate the proposed ASE system using an infant cry dataset. The experimental results confirm the effectiveness of the proposed TAP-CRNN, compared with related deep neural network models, and demonstrate that the proposed TAP-CRNN can more effectively reduce noise components from infant cry signals with unseen background noises at different signal-to-noise levels. We further tested the TAP-CRNN ASE system on a downstream infant cry detection (ICD) system, which determines whether a sound segment is involved in an infant cry event. Experimental results show that TAP-CRNN ASE can effectively reduce the noise components, thereby improving the performance of ICD under noisy conditions.

Pooling

SIGNAL (programming language)

Convolution (computer science)

10.1109/tai.2022.3169995

Cite

Citations (3)

A Novel Temporal Attentive-Pooling based Convolutional Recurrent Architecture for Acoustic Signal Enhancement

arXiv (Cornell University) (2022)

Tassadaq Hussain Wei-Chien Wang Mandar Gogate Kia Dashtipour Yu Tsao

In acoustic signal processing, the target signals usually carry semantic information, which is encoded in a hierarchal structure of short and long-term contexts. However, the background noise distorts these structures in a nonuniform way. The existing deep acoustic signal enhancement (ASE) architectures ignore this kind of local and global effect. To address this problem, we propose to integrate a novel temporal attentive-pooling (TAP) mechanism into a conventional convolutional recurrent neural network, termed as TAP-CRNN. The proposed approach considers both global and local attention for ASE tasks. Specifically, we first utilize a convolutional layer to extract local information of the acoustic signals and then a recurrent neural network (RNN) architecture is used to characterize temporal contextual information. Second, we exploit a novelattention mechanism to contextually process salient regions of the noisy signals. The proposed ASE system is evaluated using a benchmark infant cry dataset and compared with several well-known methods. It is shown that the TAPCRNN can more effectively reduce noise components from infant cry signals in unseen background noises at challenging signal-to-noise levels.

Pooling

Benchmark (surveying)

SIGNAL (programming language)

10.48550/arxiv.2201.09913

Cite

Citations (0)

Blind Monaural Source Separation on Heart and Lung Sounds Based on Periodic-Coded Deep Autoencoder

IEEE Journal of Biomedical and Health Informatics (2020)

Kun-Hsi Tsai Wei-Chien Wang Chui-Hsuan Cheng Chan‐Yen Tsai Jou‐Kou Wang

Auscultation is the most efficient way to diagnose cardiovascular and respiratory diseases. To reach accurate diagnoses, a device must be able to recognize heart and lung sounds from various clinical situations. However, the recorded chest sounds are mixed by heart and lung sounds. Thus, effectively separating these two sounds is critical in the pre-processing stage. Recent advances in machine learning have progressed on monaural source separations, but most of the well-known techniques require paired mixed sounds and individual pure sounds for model training. As the preparation of pure heart and lung sounds is difficult, special designs must be considered to derive effective heart and lung sound separation techniques. In this study, we proposed a novel periodicity-coded deep auto-encoder (PC-DAE) approach to separate mixed heart-lung sounds in an unsupervised manner via the assumption of different periodicities between heart rate and respiration rate. The PC-DAE benefits from deep-learning-based models by extracting representative features and considers the periodicity of heart and lung sounds to carry out the separation. We evaluated PC-DAE on two datasets. The first one includes sounds from the Student Auscultation Manikin (SAM), and the second is prepared by recording chest sounds in real-world conditions. Experimental results indicate that PC-DAE outperforms several well-known separations works in terms of standardized evaluation metrics. Moreover, waveforms and spectrograms demonstrate the effectiveness of PC-DAE compared to existing approaches. It is also confirmed that by using the proposed PC-DAE as a pre-processing stage, the heart sound recognition accuracies can be notably boosted. The experimental results confirmed the effectiveness of PC-DAE and its potential to be used in clinical applications.

Heart sounds

Auscultation

Monaural

Autoencoder

Spectrogram

Source Separation

10.1109/jbhi.2020.3016831

Cite

Citations (2)

Factors Affecting the High Early Strength Development and the Methods for Testing High Early Compressive Strength of the Rigid Pavement

Lecture notes in civil engineering (2022)

Wei-Chien Wang Shuei-Lien Fang Tien-Yu Chen Tai-Chiang Kao Liu Chang-chu

Hammer

10.1007/978-3-030-87379-0_30

Cite

Citations (0)

The Influence of Temperatures on the Early Age of Cement Hydration, and Behaviour of Cement with the Addition of Tea and Tipa

SSRN Electronic Journal (2022)

Wei-Chien Wang Hoang Trung Hieu Duong Wei‐Hsing Huang Chih-Chien Liu I-Chieh Lo

This paper aims to investigate the influence of temperatures on the early age of cement hydration, as well as the performance of cement paste with the addition of TEA and TIPA; hence, the mechanisms of TEA, TIPA, and curing temperature were explored. The results demonstrate that the addition of 0.05% and 0.1% TEA, and 0.05% and 0.1% TIPA retarded the setting time of cement pastes; however, this retardation depended on both the dosage and the curing temperature. At 23oC curing temperature, both TEA and TIPA could be used as a hardening accelerator at the dosage of 0.05% by cement weight. Adding 0.05% TEA into the cement paste strongly accelerated the hydration of C3A and increased the early compressive strength, which was more efficient than adding 0.05% and 0.1% TIPA, especially after 8 hours of curing. However, at 50oC curing temperature, as TIPA accelerated the formation of AFt, AFm, CH, and C-S-H, using TIPA was more effective than using TEA to increase the compressive strength of the cement paste. These findings provide theoretical guidance for the use of cement-based materials adding TEA and TIPA under hot and standard curing temperatures.

10.2139/ssrn.4105313

Cite

Citations (0)

The pH test technology suitable to evaluate content of hazardous electric arc furnace oxidizing slag in fine aggregate

Construction and Building Materials (2024)

Jia-Chen Xue Wei-Chien Wang Kao‐Hao Chang Shang-Feng Hong Her-Yung Wang

Oxidizing agent

Electric arc furnace

Slag (welding)

10.1016/j.conbuildmat.2024.136231

Cite

Citations (0)

A Machine Vision Based Automatic Optical Inspection System for Measuring Drilling Quality of Printed Circuit Boards

IEEE Access (2016)

Wei-Chien Wang Shang-Liang Chen Liang-Bi Chen Wan‐Jung Chang

In this paper, we develop and put into practice an automatic optical inspection (AOI) system based on machine vision to check the holes on a printed circuit board (PCB). We incorporate the hardware and software. For the hardware part, we combine a PC, the three-axis positioning system, a lighting device, and charge-coupled device cameras. For the software part, we utilize image registration, image segmentation, drill numbering, drill contrast, and defect displays to achieve this system. Results indicated that an accuracy of 5 μm could be achieved in errors of the PCB holes allowing comparisons to be made. This is significant in inspecting the missing, the multi-hole, and the incorrect location of the holes. However, previous work only focuses on one or other feature of the holes. Our research is able to assess multiple features: missing holes, incorrectly located holes, and excessive holes. Equally, our results could be displayed as a bar chart and target plot. This has not been achieved before. These displays help users to analyze the causes of errors and immediately correct the problems. In addition, this AOI system is valuable for checking a large number of holes and finding out the defective ones on a PCB. Meanwhile, we apply a 0.1-mm image resolution, which is better than others used in industry. We set a detecting standard based on 2-mm diameter of circles to diagnose the quality of the holes within 10 s.

Automated optical inspection

Machine Vision

Automated X-ray inspection

Feature (linguistics)

10.1109/access.2016.2631658

Cite

Citations (96)

Flexure Strengthening and Analysis Using CFRP Composite and Reactive Powder Concrete

Buildings (2023)

Ho Woo Lee Yishuo Huang Yu-Cheng Kan Wei-Chien Wang Yung-Chih Wang

There are many cases of carbon fiber-reinforced polymer (CFRP) and reactive powder concrete (RPC) in structural repair and reinforcement, but there are not many related theoretical and experimental discussions. Therefore, the purpose of this research is to focus on the experiment and theoretical calculation of the flexural strengthening of concrete beam specimens. The study was primarily separated into two parts. In the first part, the laboratory tests of flexure strengthening by using CFRP and two reactive powder concretes (RPC1 and RPC2) are used to evaluate the effect of strengthening concrete beams. Secondly, the test value of the maximum flexure failure load of the strengthening specimen using RPC or CFRP and its theoretical value calculated by the transform section method or the ultimate strength method are discussed and compared with their reinforced effect. The test results show that the RPC and CFRP display excellent repair and retrofit potential. The RPC reinforcement material with a thickness of 2 cm and a steel fiber content of 1.0% is approximately equal to the reinforcement effect of three layers of CFRP. In particular, the RPC reinforcement effect is good, and the bonding interface is not damaged. The transform section method could be used to analyze and calculate the maximum flexure failure load of the RPC strengthening concrete beam. The transform section method and ultimate strength method cannot accurately analyze the maximum flexure failure load of the CFRP-strengthening concrete beam.

Ultimate load

10.3390/buildings13081890

Cite

Citations (3)

Semantic Segmentation With Global Encoding and Dilated Decoder in Street Scenes

IEEE Access (2018)

Lei Fan Huifang Kong Wei-Chien Wang Jiapeng Yan

Semantic segmentation as a pixel-wise segmentation task provides rich object information, and it has been widely applied in many fields ranging from autonomous driving to medical image analysis. There are two main challenges on existing approaches: the first one is the obfuscation between objects resulted from the prediction of the network and the second one is the lack of localization accuracy. Hence, to tackle these challenges, we proposed global encoding module (GEModule) and dilated decoder module (DDModule). Specifically, the GEModule that integrated traditional dictionary learning and global semantic context information is to select discriminative features and improve performance. DDModule that combined dilated convolution and dense connection is used to decoder module and to refine the prediction results. We evaluated our proposed architecture on two public benchmarks, Cityscapes and CamVid data set. We conducted a series of ablation studies to exploit the effectiveness of each module, and our approach has achieved an intersection-over-union scores of 71.3% on the Cityscapes data set and 60.4% on the CamVid data set.

Discriminative model

Ranging

Convolution (computer science)

Margin (machine learning)

Data set

10.1109/access.2018.2868801

Cite

Citations (20)