ABSTRACT Object detection is an essential task in the domain of computer vision; however, its performance often deteriorates under adverse conditions such as low illumination and fog. To tackle these challenges, we propose SGPS‐YOLO, a lightweight and robust object detection framework built upon the YOLOv11 architecture. The proposed SharedPyramidConv module leverages dilated convolutions and shared kernel strategies to ensure multi‐scale semantic consistency while preserving fine‐grained spatial details. The designed GroupEfficientDetect module adopts a grouped convolutional architecture to effectively extract salient features from complex backgrounds while reducing computational overhead. Additionally, we incorporate the Powerful‐IoU loss function, which includes an adaptive penalty factor and gradient modulation mechanism to improve localization accuracy, and integrate the Shuffle Attention mechanism to enhance feature representation across scales. Results from experiments on the Complex VOC dataset demonstrate that SGPS‐YOLO achieves a 3.2% improvement in and a 4.2% boost in , while reducing the number of parameters by 9.4%, compared to YOLOv11s. Similar performance improvements and lightweight characteristics were also observed on the public datasets RTTS and ExDark.

Artificial neural network (ANN) also known simply as a neural network , is a branch of machine learning(ML), that is developed based on neuronal organization discovered by connectionism in the biological neural network in animal intelligence. In this manuscript, we invent the theory of pq-rung orthopair fuzzy linguistic (pq-ROFL) set and their valuable properties. Moreover, we expose the theory of pq-ROFL Muirhead mean (pq-ROFLMM), pq-ROFL weighted Muirhead mean (pq-ROFLWMM), pq-ROFL dual Muirhead mean (pq-ROFLDMM), and pq-ROFL dual weighted Muirhead mean (pq-ROFLDWMM) operators. Some effective and reliable properties of the invented theory are also derived. Additionally, we also evaluate the unkhnown weight vector of criteria by using analytical hierarchy process (AHP). Moreover, we discovered the best type of artificial neural network under the consideration of derived operators for pq-ROFL information. Finally, we illustrate some numerical examples in the environment of multi-attribute decision-making (MADM) and try to compare the proposed results with some prevailing results to show the reliability and supremacy of the invented approaches.

Under real-world conditions, faulty samples of key components (e.g., bearings and cutting tools, etc.) are typically limited and sparse. Additionally, their historical data is characterized by time-series and imbalance characteristics. In other words, the training samples are not only limited and noisy, but also exhibit both within-class and between-class imbalance. These factors present significant challenges in the realm of fault monitoring modeling. To tackle these challenges, this paper presents an innovative fault diagnosis method rooted in the extended NI-MWMOTE and LS-SVM. NI-MWMOTE stands as an advanced noise-immunity majority weighted minority oversampling technique, originally introduced in our prior research, and it has exhibited exceptional competitiveness in noisy imbalanced benchmark datasets. It champions an adaptive noise processing strategy leveraging the distribution characteristics of noisy imbalanced data and the essence of machine learning. Specifically, it employs Euclidean distance and neighbor density to differentiate between spurious noise and true noise, and it determines the optimal processing strategy based on misclassification error and iteration. Furthermore, it employs unsupervised aggregative hierarchical clustering, misclassification error, and majority-weighted minority oversampling in a collaborative manner to address both within-class and between-class imbalanced problems. The primary contribution of our paper lies in the context of the monitoring scenario mentioned above. We have expanded the hyper-parameter range of NI-MWMOTE, corrected and optimized its built-in noise function to enhance the interpretability of the model, and successfully applied it in conjunction with LS-SVM to this particular setting. Notably, this marks the pioneering endeavor within our established knowledge sphere into the domain of tool wear state monitoring. The results suggest that, when compared to 11 well-known algorithms, our framework demonstrates significant competitiveness in real-world scenarios characterized under data-limited and noise-imbalanced scenarios for bearings and cutting tools fault diagnosis. This establishes a solid theoretical and practical foundation for similar scenarios.