Infrared and visible image fusion aims to produce images that highlight key targets and offer distinct textures, by merging the thermal radiation infrared images with the detailed texture visible images. Traditional auto encoder-decoder-based fusion methods often rely on manually designed fusion strategies, which lack flexibility across different scenarios. Addressing this limitation, we introduce EMAFusion, a fusion approach featuring an enhanced multiscale encoder and a learnable, lightweight fusion network. Our method incorporates skip connections, the convolutional block attention module (CBAM), and nest architecture within the auto encoder-decoder framework to adeptly extract and preserve multiscale features for fusion tasks. Furthermore, a fusion network driven by spatial and channel attention mechanisms is proposed, designed to precisely capture and integrate essential features from both image types. Comprehensive evaluations of the TNO image fusion dataset affirm the proposed method’s superiority over existing state-of-the-art techniques, demonstrating its potential for advancing infrared and visible image fusion.

Feedforward neural network (FNN) is an information processing system that simulates human brain function to a certain extent by referring to the structure of biological neural network and the working mechanism of biological neurons. As the most important part, the architecture of FNN essentially influences its application performance. This paper proposed a structure self-adaptive algorithm for feedforward neural network (SSAFNN) based on biological mechanism. Firstly, self-adaptive neuron growing index and pruning index are proposed based on the idea of biological neuron grow factor and neuron competition, respectively. The FNN structure is dynamically adjusted according to the growing and pruning index of hidden neurons. Secondly, the connect weights of FNN are automatically adjusted during the self-organizing process and trained by gradient descent method during the learning process. Thirdly, the theoretical analysis is given that this proposed algorithm not only optimized the network structure, but also ensured the convergence and performance of FNN. The proposed SSAFNN is tested on the benchmark problems in the field of classification and prediction, and applied in the engineering problems of anchor bolt non-destructive testing and wastewater effluent ammonia nitrogen predicting. The experiment results reveal the good performance and potentiality of SSAFNN in industrial applications.

As a critical supporting component in geotechnical engineering structures such as bridges, tunnels, and highways, the anchorage quality of bolts directly impacts their structural safety. The ultrasonic guided wave method is a popular method for the non-destructive testing of anchorage quality. However, noise from complex field environments, modal mixing caused by anchoring interface reflections, and dispersion effects make it challenging to directly extract defect features from guided wave signals in the time or frequency domains. To address these challenges, this study proposes a solution based on the combination of the guided wave time–frequency spectrum and the gated attention residual network (GA-ResNet). The GA-ResNet introduces a gating mechanism to balance spatial attention and channel attention, and it is used for anchoring model type recognition. Experiments were conducted on four types of anchorage models, and the time–frequency spectrum was selected to be the input feature. The results demonstrate that the GA-ResNet can effectively predict the anchorage bolt defect type and prevent potential safety accidents.

Multi-sensor fire detection technology has been widely adopted in practical applications; however, existing methods still suffer from high false alarm rates and inadequate adaptability in complex environments due to their limited capacity to capture deep time-series dependencies in sensor data. To enhance robustness and accuracy, this paper proposes a novel model named BiLSTM-LN-SA, which integrates a Bidirectional Long Short-Term Memory (BiLSTM) network with Layer Normalization (LN) and a Self-Attention (SA) mechanism. The BiLSTM module extracts intricate time-series features and long-term dependencies. The incorporation of Layer Normalization mitigates feature distribution shifts across different environments, thereby improving the model’s adaptability to cross-scenario data and its generalization capability. Simultaneously, the Self-Attention mechanism dynamically recalibrates the importance of features at different time steps, adaptively enhancing fire-critical information and enabling deeper, process-aware feature fusion. Extensive evaluation on a real-world dataset demonstrates the superiority of the BiLSTM-LN-SA model, which achieves a test accuracy of 98.38%, an F1-score of 0.98, and an AUC of 0.99, significantly outperforming existing methods including EIF-LSTM, rTPNN, and MLP. Notably, the model also maintains low false positive and false negative rates of 1.50% and 1.85%, respectively. Ablation studies further elucidate the complementary roles of each component: the self-attention mechanism is pivotal for dynamic feature weighting, while layer normalization is key to stabilizing the learning process. This validated design confirms the model’s strong generalization capability and practical reliability across varied environmental scenarios.

Rock bolts play an important supporting role in the construction of slopes, deep foundation pits and tunnels. As such, it is especially necessary to assess bolt anchorage quality. This paper proposes an identification model for bolt anchorage defects based on an Elman neural network (ElmanNN) optimised using an improved chicken swarm optimisation (CSO) algorithm and the frequency response function. First, the principal components of the frequency response functions of different anchorage bolts are used as the input within the model. Next, the weights and thresholds of the ElmanNN are optimised using an improved CSO algorithm based on chaotic disturbance and elite opposition-based learning. Finally, the model is used to identify bolt anchorage defects. The experimental results show that the model has a higher identification accuracy and faster convergence rate than other neural network models.Publisher Note: Following the original publication of this paper, a change was made to identify the corresponding authors on page 588. Full details can be found in the correction notice given after page 597 of the PDF for this article.