Fish are indicative species with a relatively balanced ecosystem. Underwater target fish detection is of great significance to fishery resource investigations. Traditional investigation methods cannot meet the increasing requirements of environmental protection and investigation, and the existing target detection technology has few studies on the dynamic identification of underwater fish and small targets. To reduce environmental disturbances and solve the problems of many fish, dense, mutual occlusion and difficult detection of small targets, an improved CME-YOLOv5 network is proposed to detect fish in dense groups and small targets. First, the coordinate attention (CA) mechanism and cross-stage partial networks with 3 convolutions (C3) structure are fused into the C3CA module to replace the C3 module of the backbone in you only look once (YOLOv5) to improve the extraction of target feature information and detection accuracy. Second, the three detection layers are expanded to four, which enhances the model’s ability to capture information in different dimensions and improves detection performance. Finally, the efficient intersection over union (EIOU) loss function is used instead of the generalized intersection over union (GIOU) loss function to optimize the convergence rate and location accuracy. Based on the actual image data and a small number of datasets obtained online, the experimental results showed that the mean average precision (mAP@0.50) of the proposed algorithm reached 94.9%, which is 4.4 percentage points higher than that of the YOLOv5 algorithm, and the number of fish and small target detection performances was 24.6% higher. The results show that our proposed algorithm exhibits good detection performance when applied to densely spaced fish and small targets and can be used as an alternative or supplemental method for fishery resource investigation.

To understand the influence of fly ash (FA) content on the microstructure and strength of concrete under freeze–thaw cycles (FTCs), four groups of concrete with different FA contents (0–30%) were tested under FTC condition. Mass loss rate, relative dynamic modulus of elasticity (RDME), splitting tensile strength and other damage indicators were selected to describe the impact of macro properties. The micro physical changes, porosity and pore size distribution parameters were obtained through scanning electron microscope (SEM) and nuclear magnetic resonance (NMR) experiments. The influence mechanism of FA content on the frost resistance durability of concrete under FTC was discussed from the macro and micro perspectives. The results show that under the action of FTC, the addition of FA fills the pores, reduces the pore spacing, improves the strength of concrete, and makes the RDEM and splitting tensile strength of concrete increase first and then decrease. Among them, 20% FA concrete has the best frost resistance. The pore structure parameters show that the content of pores smaller than 100 nm has a great impact on the frost resistance durability of FA concrete, and increasing the content of these pores can improve the frost resistance durability of concrete.

Freeze–thaw cycles (FTCs) and steel bar corrosion (SBC) are the most common service conditions of hydraulic concrete and have significant impacts on its durability. Using pullout and microscopic tests of different FTC and SBC rates, we selected the mass loss rate, ultrasonic velocity, bond strength and bond slip in order to describe the changes in the macro-properties, and also selected the porosity and pore size distribution as micro-parameters in order to explore the influence of FTCs and SBC on the mechanical properties of hydraulic concrete. The results showed that the bond strength decreased as the FTCs increased due to the microstructure damage caused by FTC and SBC, which affects the mechanical properties. A corrosion rate of ≤3% offset the damage caused by 50 FTCs. FTCs and SBC resulted in superimposed damage effects on the concrete. In addition, we established a bond strength damage model based on the joint FTCs and SBC and quantitatively described the degradation law of the macro-mechanical properties. The analysis shows that the influence of FTCs on the bond strength was greater than that of the SBC. These research results can provide a reference and experimental support for the frost-resistant design and durability prediction of hydraulic concrete structures in cold environments.

The timely identification and detection of surface cracks in concrete dams, an important public safety infrastructure, is of great significance in predicting engineering hazards and ensuring dam safety. Due to their low efficiency and accuracy, manual detection methods are gradually being replaced by computer vision techniques, and deep learning semantic segmentation methods have higher accuracy and robustness than traditional image methods. However, the lack of data images and insufficient detection performance remain challenges in concrete dam surface crack detection scenarios. Therefore, this paper proposes an intelligent detection method for concrete dam surface cracks based on two-stage transfer learning. First, relevant domain knowledge is transferred to the target domain using two-stage transfer learning, cross-domain and intradomain learning, allowing the model to be fully trained with a small dataset. Second, the segmentation capability is enhanced by using residual network 50 (ResNet50) as a UNet model feature extraction network to enhance crack feature information extraction. Finally, multilayer parallel residual attention (MPR) is integrated into its jump connection path to improve the focus on critical information for clearer fracture edge segmentation. The results show that the proposed method achieves optimal mIoU and mPA of 88.3% and 92.7%, respectively, among many advanced semantic segmentation models. Compared with the benchmark UNet model, the proposed method improves mIoU and mPA by 4.6% and 3.2%, respectively, reduces FLOPs by 36.7%, improves inference speed by 48.9%, verifies its better segmentation performance on dam face crack images with a low fine crack miss detection rate and clear crack edge segmentation, and achieves an accuracy of over 85.7% in crack area prediction. In summary, the proposed method has higher efficiency and accuracy in concrete dam face crack detection, with greater robustness, and can provide a better alternative or complementary approach to dam safety inspections than the benchmark UNet model.

Measuring the water content of slopes is essential because the distribution and migration of water within slopes are important factors of landslide instability. In this study, the relationship between the resistivity, volumetric temperature water content and temperature of landslide soil was modelled. The model was validated by indoor landslide model tests and field tests in Baijiabao to investigate the effect of reservoir water levels on the water content of landslide slopes. Test results showed that, as the reservoir levels rose, the water content of the landslide soil increased. Moreover, a good correspondence between the measured results and the inversion results based on the resistivity data was obtained by using the high-density electrical method in combination with the developed model of the relationship between resistivity, volumetric water content and temperature, indicating that the proposed method is reliable and practicable in hydrodynamic landslide monitoring.

Bioinspired composites, capable of tailoring mechanical properties by the strategy of making full use of their advantages and bypassing their drawbacks, are vital for numerous engineering applications such as lightweight ultrahigh-strength, enhanced toughness, improved low-/high- velocity impact resistance, wave filtering, and energy harvesting. Helicoidal composites are examples of them. However, how to optimize the geometric structure to maximize the low-velocity impact resistance of helicoidal composites has been ignored, which is vital to the lightweight and high strength for aerospace, defense, ship, bridge, dam, vessel, and textile industries. Here, we combined experiments and numerical simulations to report the dynamic response of helicoidal composites subjected under low-velocity impact (0–10 m/s). Our helicoidal structures, inspired by the Stomatopod Dactyl club, are fabricated using polylactic acid (PLA) by FFF in a single-phase way. The helicoidal strategy aims to exploit, to a maximum extent, the axial tensile strength of filaments and simultaneously make up the shortage of inter-filament contact strength. We demonstrate experimentally that the low-velocity impact resistance has been enhanced efficiently as the helicoidal angle varies, and that the 15° helicoidal plate is better than others, which has also been confirmed by the numerical simulations. The findings reported here provide a new routine to design composites systems with enhanced impact resistance, offering a method to improve impact performance and expand the application of 3D printing.