Mask is considered as an important prior for fusion, which could selectively enhance specific regions to generate ideal fused images. However, masks used in the existing methods exhibit limitations in the precise representation of targets, and more importantly, these masks are generated from a single modality, which restricts the effective integration of multi-modal information. To address this issue, we propose a competitive mask-guidance fusion method for infrared and visible images. A multi-modal semantic-sensitive mask selection network is proposed to generate complementary-mask maps, which organically integrate advantageous target regions of different modalities by competitively comparing the qualities of masks. In this network, a pseudosiamese architecture is designed to obtain respective target masks, and specifically, a spatial-aligned-based feature aggregation module is devised to produce high-quality pseudo-labels which are served as references for the generation of the complementary-mask maps. Furthermore, we propose a bidirectional-collaboration region fusion strategy, which enhances the expression of advantageous target regions from each modality in foreground while suppressing the contribution of corresponding regions from the other modality in background. Compared to methods on public datasets, the results show that our method significantly enhances the description of semantic-sensitive targets in fused images, including the saliency and the integrity of structural information. Code are available at https://github.com/9DunDun9/FusionNet.git.

In industrial production, complex working conditions often result in poor generalization of fault diagnosis models. Consequently, we propose a method called Dual Invariant Feature Domain Generalization (DIFDG) for fault diagnosis of rolling bearings. This method is built upon the knowledge distillation framework, wherein phase information is extracted from one-dimensional signals using the Fourier transform. The phase information remains insensitive to domain changes and serves as the training data for the teacher network, enabling it to learn internally-invariant features. Simultaneously, employing the knowledge distillation technique, the student network learns internally-invariant features from one-dimensional time series signals. To prevent information loss, two-dimensional time-frequency diagram data is incorporated into the student network training. The student network is trained to acquire mutually-invariant features using a class-related loss function for data from various domains. This method trains a student network to extract dual invariant features from the data, enhancing the generalization to applications involving unknown data. Experimental results demonstrate that this approach outperforms other representative methods in both the CWRU and NEFU datasets.

The operational data of rolling bearings under different working conditions vary greatly, leading to poor generalization ability of fault diagnosis models under unknown working conditions. The current domain generalization methods used in vibration fault diagnosis have not yet solved the problem of extracting invariant features. This paper proposes a invariant feature purification method for domain generalization in rolling bearing fault diagnosis (IFPDG) to address this issue. This method iterates the model through two stages of game theory, ensuring that only invariant features exist in the feature space. During the global training phase, interference from domainrelated features is eliminated through a phase attention mechanism and feature decoupling loss. During the adversarial training phase, interference from inefficient features is eliminated through feature masks. The experimental verification of this method is conducted on the CWRU dataset and the NEFU_FDDG dataset. Especially in ablation experiments, it is confirmed that this method has advantages in generalization performance by comparing with four variants.