Crack detection on dam surfaces is an important task for safe inspection of hydropower stations. More and more object detection methods based on deep learning are being applied to crack detection. However, most of the methods can only achieve the classification and rough location of cracks. Pixel-level crack detection can provide more intuitive and accurate detection results for dam health assessment. To realize pixel-level crack detection, a method of crack detection on dam surface (CDDS) using deep convolution network is proposed. First, we use an unmanned aerial vehicle (UAV) to collect dam surface images along a predetermined trajectory. Second, raw images are cropped. Then crack regions are manually labelled on cropped images to create the crack dataset, and the architecture of CDDS network is designed. Finally, the CDDS network is trained, validated and tested using the crack dataset. To validate the performance of the CDDS network, the predicted results are compared with ResNet152-based, SegNet, UNet and fully convolutional network (FCN). In terms of crack segmentation, the recall, precision, F-measure and IoU are 80.45%, 80.31%, 79.16%, and 66.76%. The results on test dataset show that the CDDS network has better performance for crack detection of dam surfaces.

An increasing number of detection methods based on computer vision are applied to detect cracks in water conservancy infrastructure. However, most studies directly use existing feature extraction networks to extract crack information, which are proposed for open-source datasets. As the crack distribution and pixel features are different from these data, the extracted crack information is incomplete. In this paper, a deep learning-based network for dam surface crack detection is proposed, which mainly addresses the semantic segmentation of cracks on the dam surface. Particularly, we design a shallow encoding network to extract features of crack images based on the statistical analysis of cracks. Further, to enhance the relevance of contextual information, we introduce an attention module into the decoding network. During the training, we use the sum of cross-entropy and Dice loss as the loss function to overcome data imbalance. The quantitative crack information is extracted by the imaging principle after using morphological algorithms to extract the morphological features of the predicted result. We built a manual annotation dataset containing 1577 images to verify the effectiveness of the proposed method. This method achieves state-of-the-art performance on our dataset. Specifically, the precision, recall, Intersection of Union (IoU), F1_measure, and accuracy are 90.81%, 81.54%, 75.23%, 85.93%, 99.76%, respectively, and the quantification error of cracks is less than 4%.

This article investigates the underwater navigation control of a water–air amphibious multirotor vehicle. We use active disturbance rejection control (ADRC) to construct a tandem-level ADRC motion controller for the water–air multirotor vehicle and introduce particle swarm optimization (PSO) to quickly tune controller parameters. First, the amphibious multirotor vehicle's governing kinematic and dynamic equations are derived. Then, the hydrodynamics of the underwater navigation process is analyzed and estimated. Accordingly, ADRC-based position and attitude controllers are designed and compared with a traditional proportional–integral–derivative (PID) controller and a sliding mode controller (SMC). In addition, PSO is introduced to adjust the gain parameters of the PID, SMC controller, and the parameters of the nonlinear state error feedback law and extended state observer of the ADRC controller. Finally, to verify the stability and robustness of the ADRC controller, simulations are performed under strong external disturbances with a water–air multirotor vehicle. The results demonstrate that controller performance can be improved by introducing PSO to tune the controller parameters and that it is more beneficial for the self-adjacent controller with many control parameters and strong interparameter nonlinearity. ADRC responds faster, rejects external disturbances better, and is more robust than SMC and PID, which permits it to meet the performance requirements of the controller in complex underwater environments.

Underwater structure inspections are essential for infrastructure maintenance, such as hydraulic facilities, bridges, and ports. Due to the influence of turbidity, dark light, and distortion, the traditional methods cannot satisfy the requirements of on-site inspection applications. This paper proposed a methodology of the point cloud data capture in the turbid underwater environment. The method consisted of an acquisition device, a distortion correction algorithm, and a parameter optimization approach. The acquisition device was designed by composing a silt-removing module, a structured light camera module, and a clear water replacement module, which can integrate with an underwater inspection robot. The underwater multi-medium plane refraction distortion model was established through analysis, and a refraction correction algorithm was provided to correct the distortion. To obtain the maximum field of view of the point cloud, the nonlinear optimization approach was used to select the medium material and thickness. After the real experiments using the Intel RealSense sr300 depth camera, maximum measuring distance could range up to 253 mm in water, the accuracy of the point cloud of the underwater target objects was ±3.77 mm, and the maximum error was 8.76%. Compared with other methods, this method was more suitable for 3D point cloud capture in the turbidity environment.

Concrete defect detection is a critical task to ensure the safe and stable operation of concrete infrastructure. In order to effectively solve the problems of low efficiency, high cost and poor performance of existing methods, a high-precision concrete defect detection method YOLOv8-CDD (Concrete Defect Detection) combining convolutional neural network and transformer is proposed in this paper. Firstly, based on the features with a large span of concrete defect features, a bot-transformer module that can effectively extract the global information of defect features is proposed to improve the network’s ability to extract global features. Secondly, in order to further strengthen the interaction between defect feature channels and spatial information, a convolutional triplet attention module is introduced into the feature enhancement network to effectively integrate the information of different dimensions of defect features and improve the model detection accuracy. Additionally, in order to enhance the learning of samples with different degrees of difficulty, the introduction of Focaler-CIoU instead of the original boundary regression loss function can optimize the model training process. Finally, the dataset was collected and organized in concrete scenarios from bridge towers, dams, and tunnel corridors, and our method achieved 0.898 average precision, 0.893 average recall, 0.031 average FPR, 0.895 average F1 score, 0.929 mAP50, and 0.731 mAP50:95 on the dataset. The experimental results show that the proposed method achieves the best performance in concrete defect detection.

Spinning Light Detection And Ranging (LiDAR) is a range scanning device with a larger field of view, which is formed by continuous spinning of a LiDAR with a rotating mechanism. Precise and robust calibration of the relative pose between LiDAR and rotating mechanism is critical for surveying, mapping, localization, etc. However, existing methods generally require ad-hoc tools, well-designed targets, or regular calibrated scenes. For improving calibration in unprepared scenes, we propose a Filtered Grid Random Sample Consensus (FGRSC) calibration approach, constituted by a Filtered Grid (FG) strategy and a Grid Random Sample Consensus (GRSC) method. Based on dividing the whole scene points into small grids when sufficient scanning is received. Firstly, FG is designed to calculate point curvature upon receiving each frame of spinning LiDAR scanning. Then FG normalizes the number of low curvature points to get an approximate probability of containing planes for every grid. Finally, given the probabilities provided by FG, GRSC preselects grids to extract and associate planes to estimate parameters with a weighted optimization. The experimental results derived from indoor and outdoor unprepared scenes show that our FGRSC outperforms the state-of-the-art methods.