Computer vision techniques, in conjunction with acquisition through remote cameras and unmanned aerial vehicles (UAVs), offer promising non-contact solutions to civil infrastructure condition assessment. The ultimate goal of such a system is to automatically and robustly convert the image or video data into actionable information. This paper provides an overview of recent advances in computer vision techniques as they apply to the problem of civil infrastructure condition assessment. In particular, relevant research in the fields of computer vision, machine learning, and structural engineering are presented. The work reviewed is classified into two types: inspection applications and monitoring applications. The inspection applications reviewed include identifying context such as structural components, characterizing local and global visible damage, and detecting changes from a reference image. The monitoring applications discussed include static measurement of strain and displacement, as well as dynamic measurement of displacement for modal analysis. Subsequently, some of the key challenges that persist towards the goal of automated vision-based civil infrastructure and monitoring are presented. The paper concludes with ongoing work aimed at addressing some of these stated challenges.
A full-scale composite floor plate was tested to investigate the flexural behavior and in-plane effects of the floor slab in a grillage of composite beams that reduces the tendency for longitudinal splitting of the concrete slab along the line of the primary beams. This is important in cases where the steel decking is discontinuous when it is orientated parallel to the beams. In this case, it is important to demonstrate that the amount of transverse reinforcement required to transfer local forces from the shear connectors can be reduced relative to the requirements of Eurocode 4. The mechanism under study involved in-plane compression forces being developed in the slab due to the restraining action of the floor plate, which was held in position by the peripheral composite beams; while the secondary beams acted as transverse ties to resist the forces in the floor plate that would otherwise lead to splitting of the slab along the line of the primary beams. The tendency for cracking along the center line of the primary beam and at the peripheral beams was closely monitored. This is the first large floor plate test that has been carried out under laboratory conditions since the Cardington tests in the early 1990s, although those tests were not carried out to failure. This floor plate test was designed so that the longitudinal force transferred by the primary beams was relatively high (i.e., it was designed for full shear connection), but the transverse reinforcement was taken as the minimum of 0.2% of the concrete area. The test confirmed that the primary beams reached their plastic bending resistance despite the discontinuous decking and transverse reinforcement at the minimum percentage given in Eurocode 4. Based on this test, a reduction factor due to shear connectors at edge beams without U-bars is proposed.
Structural health monitoring (SHM) is a multi-discipline field that involves the automatic sensing of structural loads and response by means of a large number of sensors and instruments, followed by a diagnosis of the structural health based on the collected data. Because an SHM system implemented into a structure automatically senses, evaluates, and warns about structural conditions in real time, massive data are a significant feature of SHM. The techniques related to massive data are referred to as data science and engineering, and include acquisition techniques, transition techniques, management techniques, and processing and mining algorithms for massive data. This paper provides a brief review of the state of the art of data science and engineering in SHM as investigated by these authors, and covers the compressive sampling-based data-acquisition algorithm, the anomaly data diagnosis approach using a deep learning algorithm, crack identification approaches using computer vision techniques, and condition assessment approaches for bridges using machine learning algorithms. Future trends are discussed in the conclusion.
Structural analysis of steel frames is typically performed using beam elements. Since these elements are unable to explicitly capture the local buckling behavior of steel cross-sections, traditional steel design specifications use the concept of cross-section classification to determine the extent to which the strength and deformation capacity of a cross-section are affected by local buckling. The use of plastic design methods are restricted to Class 1 cross-sections, which possess sufficient rotation capacity for plastic hinges to develop and a collapse mechanism to form. Local buckling prevents the development of plastic hinges with such rotation capacity for cross-sections of higher classes and, unless computationally demanding shell elements are used, elastic analysis is required. However, this article demonstrates that local buckling can be mimicked effectively in beam elements by incorporating the continuous strength method (CSM) strain limits into the analysis. Furthermore, by performing an advanced analysis that accounts for both geometric and material nonlinearities, no additional design checks are required. The positive influence of the strain hardening observed in stocky cross-sections can also be harnessed, provided a suitably accurate stress–strain relationship is adopted; a quad-linear material model for hot-rolled steels is described for this purpose. The CSM strain limits allow cross-sections of all slenderness to be analyzed in a consistent advanced analysis framework and to benefit from the appropriate level of load redistribution. The proposed approach is applied herein to individual members, continuous beams, and frames, and is shown to bring significant benefits in terms of accuracy and consistency over current steel design specifications.
Concrete is a continuously evolving material, and even the definition of high-performance concrete has changed over time. In this paper, high-performance characteristics of concrete material are considered to be those that support the desirable durability, resilience, and sustainability of civil infrastructure that directly impact our quality of life. It is proposed that high-performance material characteristics include tensile ductility, autogenous crack-width control, and material “greenness.” Furthermore, smart functionalities should be aimed at enhancing infrastructure durability, resilience, and sustainability by responding to changes in the surrounding environment of the structure in order to perform desirable functions, thus causing the material to behave in a manner more akin to certain biological materials. Based on recent advances in engineered cementitious composites (ECC), this paper suggests that concrete embodying such high-performance characteristics and smart multifunctionalities can be designed, and holds the potential to fulfill the expected civil infrastructure needs of the 21st century. Highlights of relevant properties of ECC are provided, and directions for necessary future research are indicated.
Ultrasound (US) has become one of the most commonly performed imaging modalities in clinical practice. It is a rapidly evolving technology with certain advantages and with unique challenges that include low imaging quality and high variability. From the perspective of image analysis, it is essential to develop advanced automatic US image analysis methods to assist in US diagnosis and/or to make such assessment more objective and accurate. Deep learning has recently emerged as the leading machine learning tool in various research fields, and especially in general imaging analysis and computer vision. Deep learning also shows huge potential for various automatic US image analysis tasks. This review first briefly introduces several popular deep learning architectures, and then summarizes and thoroughly discusses their applications in various specific tasks in US image analysis, such as classification, detection, and segmentation. Finally, the open challenges and potential trends of the future application of deep learning in medical US image analysis are discussed.
Maintaining sustained attention during a prolonged cognitive task often comes at a cost: high levels of mental fatigue. Heuristically, mental fatigue refers to a feeling of tiredness or exhaustion, and a disengagement from the task at hand; it manifests as impaired cognitive and behavioral performance. In order to effectively reduce the undesirable yet preventable consequences of mental fatigue in many real-world workspaces, a better understanding of the underlying neural mechanisms is needed, and continuous efforts have been devoted to this topic. In comparison with conventional univariate approaches, which are widely utilized in fatigue studies, convergent evidence has shown that multivariate functional connectivity analysis may lead to richer information about mental fatigue. In fact, mental fatigue is increasingly thought to be related to the deviated reorganization of functional connectivity among brain regions in recent studies. In addition, graph theoretical analysis has shed new light on quantitatively assessing the reorganization of the brain functional networks that are modulated by mental fatigue. This review article begins with a brief introduction to neuroimaging studies on mental fatigue and the brain connectome, followed by a thorough overview of connectome studies on mental fatigue. Although only a limited number of studies have been published
thus far, it is believed that the brain connectome can be a useful approach not only for the elucidation of underlying neural mechanisms in the nascent field of neuroergonomics, but also for the automatic detection and classification of mental fatigue in order to address the prevention of fatigue-related human error in the near future.
Engineering the biosynthesis of plant-derived natural products in microbes presents several challenges, especially when the expression and activation of the plant cytochrome P450 enzyme is required. By recruiting two enzymes—HpaB and HpaC—from several bacteria, we constructed functional 4-hydroxyphenylacetate 3-hydroxylase (4HPA3H) in Saccharomyces cerevisiae to take on a role similar to that of the plant-derived cytochrome P450 enzyme and produce caffeic acid. Along with a common tyrosine ammonia lyase (TAL), the different combinations of HpaB and HpaC presented varied capabilities in producing the target product, caffeic acid, from the substrate, L-tyrosine. The highest production of caffeic acid was obtained with the enzyme combination of HpaB from Pseudomonas aeruginosa and HpaC from Salmonella enterica, which yielded up to 289.4 ± 4.6 mg·L−1 in shake-flask cultivation. The compatibility of heterologous enzymes within a yeast chassis was effectively improved, as the caffeic acid production was increased by 40 times from the initial yield. Six key amino acid residues around the flavin adenine dinucleotide (FAD) binding domain in HpaB from Pseudomonas aeruginosa were distinguished from those other HpaBs, and might play critical roles in affecting enzyme activity. We have thus established an effective approach to construct a highly efficient yeast system to synthesize non-native hydroxylated phenylpropanoids.
Mercury and its derivatives are hazardous environmental pollutants and could affect the aquatic ecosystems and human health by biomagnification. Lake sediments can provide important historical information regarding changes in pollution levels and thus trace anthropogenic or natural influences. This research investigates the 100-year history of mercury (Hg) deposition in sediments from Chao Lake, a shallow eutrophic lake in China. The results indicate that the Hg deposition history can be separated into three stages (pre-1960s, 1960s–1980s, and post-1980s) over then last 100 years. Before the 1960s, Hg concentrations in the sediment cores varied little and had no spatial difference. Since the 1960s, the concentration of Hg began to increase gradually, and showed a higher concentration of contamination in the western half of the lake region than in the eastern half of the lake region due to all kinds of centralized human-input sources. The influences of anthropogenic factors and hydrological change are revealed by analyzing correlations between Hg and heavy metals (Fe, Co, Cr, Cu, Mn, Pb, and Zn), stable carbon and nitrogen isotopes (δ13C and δ15N), nutrients, particle sizes, and meteorological factors. The results show that Hg pollution intensified after the 1960s, mainly due to hydrological change, rapid regional development and urbanization, and the proliferation of anthropogenic Hg sources. Furthermore, the temperature, wind speed, and evaporation are found to interactively influence the environmental behaviors and environmental fate of Hg.
An increasing number of drivers are relying on digital map navigation systems in vehicles or mobile phones to select optimal driving routes in order to save time and improve safety. In the near future, digital map navigation systems are expected to play more important roles in transportation systems. In order to extend current navigation systems to more applications, two fundamental problems must be resolved: the lane-level map model and lane-level route planning. This study proposes solutions to both problems. The current limitation of the lane-level map model is not its accuracy but its flexibility; this study proposes a novel seven-layer map structure, called as Tsinghua map model, which is able to support autonomous driving in a flexible and efficient way. For lane-level route planning, we propose a hierarchical route-searching algorithm to accelerate the planning process, even in the presence of complicated lane networks. In addition, we model the travel costs allocated for lane-level road networks by analyzing vehicle maneuvers in traversing lanes, changing lanes, and turning at intersections. Tests were performed on both a grid network and a real lane-level road network to demonstrate the validity and efficiency of the proposed algorithm.
Bearings are the most important component of nearly all mechanical equipment, as they guarantee the steady running of the equipment, which is especially important for high-end equipment such as high-speed trains and shield tunneling machines. Requirements regarding the quality of bearings are increasing with the rapid development in technology. A country’s bearings manufacturing level directly reflects the level of that country’s steel metallurgy and machinery manufacturing. The performance of the bearing steel is the critical factor that determines the quality of a bearing. The development of new bearing steel with higher performance is the ambition of material researchers and the expectation of the manufacturing industry. Many famous bearing manufacturing enterprises are competing to develop the new generation of bearing steel. Nanostructured bainitic bearing steel (NBBS), which is a newly developed bearing steel, not only possesses high strength and toughness, but also exhibits excellent wear resistance and rolling contact fatigue (RCF) resistance. In recent years, relevant achievements in NBBS in China have led to significant progress in this field. NBBS was first used in China to manufacture large bearings for wind turbines and heavy-duty bearings, with excellent performance. As a result, NBBS and its corresponding heat-treatment process have been included in the national and industry standards for the first time. The bearing industry considers the exploitation of NBBS to be epoch-making, and has termed this kind of bearing as the second generation of bainitic bearing. In this paper, the development of NBBS is reviewed in detail, including its advantages and disadvantages. Further research directions for NBBS are also proposed.
The internal flow field study of car compartments is an important step in railroad vehicle design and optimization. The flow field profile has a significant impact on the temperature distribution and passenger comfort level. Experimental studies on flow field can yield accurate results but carry a high time and computational cost. In contrast, the numerical simulation method can yield an internal flow field profile in less time than an experimental study. This study aims to improve the computational efficiency of numerical simulation by adapting two simplified models—the porous media model and the porous jump face model—to study the internal flow field of a railroad car compartment. The results provided by both simplified models are compared with the original numerical simulation model and with experimental data. Based on the results, the porous media model has a better agreement with the original model and with the experimental results. The flow field parameters (temperature and velocity) of the porous media model have relatively small numerical errors, with a maximum numerical error of 4.7%. The difference between the numerical results of the original model and those of the porous media model is less than 1%. By replacing the original numerical simulation model with the porous media model, the flow field of subway car compartments can be calculated with a reduction of about 25% in computing resources, while maintaining good accuracy.
Engineering designs for mountainous highways emphasize compliance checking to ensure safety. However, relying
solely on compliance checking may lead designers to minimize costs at the expense of high risk indicators, since the overall risk level of the highway design is unknown to the designers. This paper describes a method for the simultaneous consideration of traffic safety risks and the associated cost burden related to the appropriate planning and design of a mountainous highway. The method can be carried out in four steps: First, the highway design is represented by a new parametric framework to extract the key design variables that affect not only the life-cycle cost but also the operational safety. Second, the relationship between the life-cycle cost and the operational safety risk factors is established in the cost-estimation functions. Third, a fault tree analysis (FTA) is introduced to identify the traffic risk factors from the design variables. The safety performance of the design solutions is also assessed by the generalized linear-regression model. Fourth, a theory of acceptable risk analysis is introduced to the traffic safety assessment, and a computing algorithm is proposed to solve for a cost-efficient optimal solution within the range of acceptable risk, in order to help decision-makers. This approach was applied and examined in the Sichuan-Tibet Highway engineering project, which is located in a complex area with a large elevation gradient and a wide range of mountains. The experimental results show that the proposed approach significantly improved both the safety and cost performance of the project in the study area.