Abstract ：

Tunnel cracks are thin and narrow linear targets, and their pixel proportions in images are usually very low, less than 6%; therefore, a method is needed to better detect small crack targets. In this study, a crack detection method based on crack characteristics and an anchor-free framework is investigated. First, the characteristics of cracks are analyzed to obtain the real crack texture, interference noise texture, and targets appearing near each crack as the context information for the model to filter and remove noise. We discuss the crack detection performance of anchor-based and anchor-free algorithms. Then, an optimized anchor-free algorithm is proposed in this paper for crack detection. Based on the advantages of YOLOX-x, we add a semantic enhancement module to better use contextual information. The experimental results show that the anchor-free algorithm performs slightly better than other algorithms in crack detection situations. In addition, the proposed method displays better detection performance for slender and inconspicuous cracks, with an average precision of 0.858.

Keyword ：

Metro tunnel Anchor-free Cracks Optimized YOLOX-x

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Recently, deep learning-based fruit detection applications have been widely used in the modern fruit industry; however, the training data labeling process remains a time-consuming and labor-intensive process. Auto labeling can provide a convenient and efficient data source for constructing smart orchards based on deep-learning technology. In our previous study, based on a labeled source domain fruit dataset, we used a generative adversarial network and a fruit detection model to achieve auto labeling of unlabeled target domain fruit images. However, since the current method uses one species source domain fruit to label multiple species target domain fruits, there is a problem of the domain gap in both the foreground and the background between the training data (retaining the source domain fruit label information) and the application data (target domain fruit images) of the fruit detection model. Therefore, we propose a domain-adaptive anchor-free fruit detection model, DomAda-FruitDet, and apply it to the previously proposed fruit labeling method to further improve the accuracy. It consists of 2 design aspects: (a) With a foreground domain-adaptive structure based on double prediction layers, an anchor-free method with multiscale detection capability is constructed to generate adaptive bounding boxes that overcome the foreground domain gap; (b) with a background domain-adaptive strategy based on sample allocation, we enhance the ability of the model to extract foreground object features to overcome the background domain gap. As a result, the proposed method can label actual apple, tomato, pitaya, and mango datasets, with an average precision of 90.9%, 90.8%, 88.3%, and 94.0%, respectively. In conclusion, the proposed DomAdaFruitDet effectively addressed the problem of the domain gap and improved effective auto labeling for fruit detection tasks.

Cite：

Copy from the list or Export to your reference management。

Abstract ：

The development of cost-effective catalysts with excellent chlorine resistance and harmful by-products inhibition is important for the environmentally friendly purification of multi-component volatile organic compounds (VOCs and chlorine-containing VOCs (CVOCs)). In this work, the Sn-doped Silicalite-1-supported Ru (Ru@Silicalite-1Sn-x, and x is the molar ratio of Si/Sn) samples were prepared using a hydrothermal strategy, and catalytic activities of these materials were investigated for the oxidative removal of mixed VOCs (dichloromethane (DCM) and toluene). The Ru@Silicalite-1-Sn-50 sample with tightly coupled redox and acidic sites exhibited high catalytic activity (T90% = 287 degrees C for toluene oxidation and T90% = 361 degrees C for DCM oxidation at a space velocity of 40,000 mL/(g h); specific reaction rate and turnover frequency (TOFRu) for toluene oxidation at 170 degrees C were 9.67 mu mol/(gcat h) and 0.98 x 10-3 s-1, and specific reaction rate and TOFRu for DCM oxidation at 200 degrees C were 3.84 mu mol/(gcat h) and 0.46 x 10-3 s-1, respectively), excellent catalytic stability (within 100 h of on-stream oxidation at 380 degrees C), and effective inhibition of toxic chlorine-containing by-products formation in the oxidation of (DCM and toluene). The doping of Sn could effectively anchor the Ru atoms to result in single-atom dispersion of Ru and generate oxygen vacancies, and optimized the synergistic interaction between Lewis acid sites and Br & oslash;nsted acid sites. The high concentration of oxygen vacancies and enriched Br & oslash;nsted acid sites promoted the cleavage of C-Cl bonds in DCM and accelerated the desorption of Cl species as inorganic chlorine. In the meanwhile, the strong electron transfer within the Sn-O-Si bond increased the Lewis acidity, which promoted the deep oxidation of dechlorinated intermediates/other intermediates over Ru@Silicalite-1-Sn-50. We believe that the present work provides a feasible and promising strategy for the design of efficient catalysts for the destruction of multicomponent VOCs and CVOCs in an industrial scale.

Keyword ：

Supported Ru single-atom catalyst Catalytic oxidation Toluene Tin-doped Silicalite-1 Dichloromethane Volatile organic compound

Cite：

Copy from the list or Export to your reference management。

Abstract ：

The number of surface defects in steel has always been a key measure of the quality of steel production. Traditional detection methods, such as the strobe method, have the problems of slow detection reaction and low accuracy. The object detection technology based on deep learning can effectively improve the detection performance by virtue of its strong real-time performance and high accuracy. However, in the actual production process, complex production environments and other industrial factors may affect the efficiency of object detection technology. We used an improved You Only Look Once (YOLO) framework to identify defects on the steel surface. This improved version of the YOLO model optimizes the original architecture to enhance the detection of small size and low contrast defects. As a real-time object detection system, YOLO is able to predict the locations and category probability of defects directly from the input image by integrating convolutional neural networks. In addition, we have adjusted the loss function and anchor frame strategy of the model to improve the detection accuracy and response speed in complex environments. The performance of the model in different types of defect identification was analyzed, and the applicability and efficiency of the model in the actual steel defect dataset NEU-DET were discussed. © 2024 IEEE.

Keyword ：

Convolutional neural networks Steel research

Cite：

Copy from the list or Export to your reference management。

Abstract ：

In recent years, with the promotion and application of prefabricated buildings in China, prefabricated building structures have received extensive attention from researchers. Grouting sleeve is the main connection method of prefabricated buildings, and the construction quality is closely related to the safety and reliability of prefabricated buildings. Piezoelectric wave analysis method is a new nondestructive structural health detection technology. In this paper, ABAQUS was used to simulate the grouting sleeve with steel bars anchorage defects based on piezoelectric wave analysis method, and the signals of steel bars with different defects were compared, and a piezoelectric ceramics layout scheme was put forward, which was suitable for the engineering. © 2024 ACM.

Keyword ：

Bars (metal) Concretes Mortar Anchor bolts Buildings

Cite：

Copy from the list or Export to your reference management。

Abstract ：

We conducted a study to evaluate the potential and robustness of gradient boosting algorithms in rock burst assessment, established a variational autoencoder (VAE) to address the imbalance rock burst dataset, and proposed a multilevel explainable artificial intelligence (XAI) tailored for tree-based ensemble learning. We collected 537 data from real-world rock burst records and selected four critical features contributing to rock burst occurrences. Initially, we employed data visualization to gain insight into the data's structure and performed correlation analysis to explore the data distribution and feature relationships. Then, we set up a VAE model to generate samples for the minority class due to the imbalanced class distribution. In conjunction with the VAE, we compared and evaluated six state-of-theart ensemble models, including gradient boosting algorithms and the classical logistic regression model, for rock burst prediction. The results indicated that gradient boosting algorithms outperformed the classical single models, and the VAE-classifier outperformed the original classifier, with the VAE-NGBoost model yielding the most favorable results. Compared to other resampling methods combined with NGBoost for imbalanced datasets, such as synthetic minority oversampling technique (SMOTE), SMOTE-edited nearest neighbours (SMOTE-ENN), and SMOTE-tomek links (SMOTE-Tomek), the VAE-NGBoost model yielded the best performance. Finally, we developed a multilevel XAI model using feature sensitivity analysis, Tree Shapley Additive exPlanations (Tree SHAP), and Anchor to provide an in-depth exploration of the decision-making mechanics of VAE-NGBoost, further enhancing the accountability of treebased ensemble models in predicting rock burst occurrences.

Keyword ：

VAE Gradient boosting Explainable artificial intelligence (XAI) Rock burst Ensemble learning

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Prefabricated cantilevered beam -column structures have gained popularity in mountainous road construction. However, a notable research gap exists regarding the beam -column connections. To fill this gap, a new bolted connection method was introduced for the cantilevered beam -column structure. The mechanical performance of this novel structure was systematically investigated through reduced -scale experiments and finite element analysis. The structure ' s response under three critical loading conditions was experimentally examined, revealing the efficacy of bolted connections in preserving structural integrity and highlighting pronounced strain concentration at the connection region of the cap beam. Furthermore, to enhance structural design, a finite element model of this beam -column structure was established to simulate the influence of various materials and structural parameters on structural response and damage. Simulation results indicated that the anchor rod angle and the number of bolts had a minor impact on load -bearing capacity and damage modes. In contrast, the choice of concrete material and steel reinforcement ratio significantly influenced load -bearing capacity. The insights derived from this study offer valuable guidance for optimizing and extending the application of this structural design.

Keyword ：

Finite element model Load-bearing capacity Cantilevered beam-column structure Bolted connection

Cite：

Copy from the list or Export to your reference management。

Abstract ：

本发明公开了一种基于无人机航拍图像的早期森林火灾检测方法，首先采用双边滤波算法和暗通道先验去雾算法来提升林火图像质量；然后针对无人机航拍早期林火图像中火灾面积过小的问题，选用精度更高的两阶段网络Faster‑RCNN作为基础检测网络；然后通过分析早期林火目标的面积尺寸分布，根据其存在面积较小的林火目标和长宽比差异较大的林火目标的特点，采用K‑means算法优化Faster‑RCNN的Anchor框；其次针对早期林火目标像素过小的问题，采用特征融合的方法突出早期林火图像特征，从而提高检测精度；最后针对无人机航拍图像中早期森林火灾背景复杂的问题，通过引入注意力机制降低图像复杂背景对早期林火目标识别模型的干扰，从而提高模型的学习效率和检测精度。

Cite：

Copy from the list or Export to your reference management。

Abstract ：

本发明公开了一种改进的RefineDet的管道缺陷检测方法。首先，RefineDet的主干网络修改为Swin‑transformer；其次，把经过Neck模块的多尺度特征融合后输出的特征用作anchor细化模块的anchors的分类和微调，这样使得anchor细化模块输出的Refinedanchors更准确，进而使得目标检测模块能更好的进行目标的定位；最后，使用warmup和cosine的学习率调整策略，并且使用Adamw优化器训练改进的RefineDet网络。本发明的基于RefineDet的改进方法能够有效提高管道缺陷检测的平均准确率。

Cite：

Copy from the list or Export to your reference management。

Abstract ：

The stability evaluation of deep foundation pit slopes is crucial, including the solution of the factor of safety and the location of the critical slip surface. Prestressed anchor cables and underground diaphragm walls are commonly used to reinforce pit slopes. By means of the dimensional increase technique and the global analysis approach of slope stability, a nonlinear optimization problem is defined for the stability analysis of pit slopes. In the optimization problem, the ordinates of discrete points on the slip surface and the factor of safety are both the decision variables, the objective function is the factor of security, and the constraints are the equilibrium equations and the convexity of the slip surface. Because the objective function is linear and the constraint functions are polynomials of degree three at most, the optimization problem is a classical optimization one that conventional optimization techniques can solve without recourse to modern optimization techniques such as the AI technique. Examples suggest that the proposed procedure is far more efficient and stable than the optimization models based on the methods of slides. At last, applications are demonstrated to ongoing deep foundation pit slopes situated at Tong Zhou - Beijing sub-center.

Keyword ：

Methods of slices Critical sliding surface Nonlinear optimization Foundation pit slopes Factor of safety Global analysis approach of slope stability

Cite：

Copy from the list or Export to your reference management。