Source Domain Research Articles

Branded Bodies

This article investigates how the emotion of shame was theorized in ancient Greece. After a discussion of the conceptualization of shame in modern times in the first part of the article, in the second part I will analyse its theorization in antiquity more specifically. I will argue that the visual component of shame, which is crucial in our modern understanding, also played a prominent role in ancient Greek theories. More precisely, shame was experienced and manifested predominantly through the skin. Alterations in complexion were conceived as the visible representation of the feeling of shame, which stigmatizes the sufferers. Likewise, the resulting itch and its obsessive nature were employed to represent mental disturbance caused by shame. Interestingly for our purpose here, shame in antiquity was a gendered emotion: different behaviours were considered shameful for men and women. By reconsidering the conceptualization of shame in ancient Greece in light of modern psychological theories, this article will contribute to demonstrating how the use of psychological and neuroscientific studies can provide new interpretations of ancient sources on shame, with particular regard to its metaphorical source domains concerning the body.

Konceptualizacja pojęcia „nadzieja” we francuskich i polskich formach autonimicznych – analiza kognitywna

The method used to describe the conceptualization of the notion of hope in French and in Polish is cognitive, as represented by G. Lakoff, M. Johnson, W. R. Langacker and E. Rosch. The analyzed corpus consists of Polish and French quotations. The distinguished conceptual metaphors that describe the notion of hope refer to a number of cognitive models, including the Ideal Cognitive Model (ICM) of hope, as well as to the Typical Cognitive Model (TCM). Hypo-hyperonymic relations found between cognitive metaphors as well as semantic universals belonging to metaphor source domains in two compared languages have been identified. The prototypical proverb about hope in each language has been identified as well.

Sample-prototype optimal transport-based universal domain adaptation for remote sensing image classification

In recent years, there is a growing interest in domain adaptation for remote sensing image scene classification, particularly in universal domain adaptation, where both source and target domains possess their unique private categories. Existing methods often lack precision on remote sensing image datasets due to insufficient prior knowledge between the source and target domains. This study aims to effectively distinguish between common and private classes despite large intra-class sample discrepancies and small inter-class sample discrepancies in remote sensing images. To address these challenges, we propose Sample-Prototype Optimal Transport-Based Universal Domain Adaptation (SPOT). The proposed approach comprises two key components. Firstly, we utilize an unbalanced optimal transport algorithm along with a sample complement mechanism to identify common and private classes based on the optimal transport assignment matrix. Secondly, we leverage the optimal transport algorithm to enhance discriminability among different classes while promoting similarity within the same class. Experimental results demonstrate that SPOT significantly enhances classification accuracy and robustness in universal domain adaptation for remote sensing images, underscoring its efficacy in addressing the identified challenges.

Feature decoupling integrated domain generalization network for bearing fault diagnosis under unknown operating conditions

In real engineering scenarios, the complex and variable operating conditions of mechanical equipment lead to distributional differences between the collected fault data and the training data. This distribution difference can lead to the failure of deep learning-based diagnostic models. Extracting generalized diagnostic knowledge from the source domain in scenarios where the target domain is not visible is the key to solving this problem. To this end, in this paper, we propose a domain generalization network for diagnosing bearing faults under unknown operating conditions, i.e., Feature Decoupled Integrated Domain Generalization Network (FDIDG). First, we propose a “feature decoupling” algorithm to uncover generalized representations of fault features from multiple source domains. The algorithm aims to explore the generalized representations of fault features by shrinking the distribution of data from multiple source domains and further generalize the fault features to unknown domains to reduce the coupling between fault features and operating conditions. Second, the diagnostic accuracy of the model under unknown operating conditions is further improved by adopting a multi-expert integration strategy in the decision-making stage and utilizing domain-private features to reduce the negative impact of edge samples on diagnosis. We conducted several sets of cross-domain experiments on both public and private datasets, and the results show that FDIDG has excellent generalization capabilities.

A Transfer Learning Radiomics Nomogram to Predict the Postoperative Recurrence of Advanced Gastric Cancer.

In this study, a transfer learning (TL) algorithm was used to predict postoperative recurrence of advanced gastric cancer (AGC) and to evaluate its value in a small-sample clinical study. A total of 431 cases of AGC from three centers were included in this retrospective study. First, TL signatures (TLSs) were constructed based on different source domains, including whole slide images (TLS-WSIs) and natural images (TLS-ImageNet). Clinical model and non-TLS based on CT images were constructed simultaneously. Second, TL radiomic model (TLRM) was constructed by combining optimal TLS and clinical factors. Finally, the performance of the models was evaluated by ROC analysis. The clinical utility of the models was assessed using integrated discriminant improvement (IDI) and decision curve analysis (DCA). TLS-WSI significantly outperformed TLS-ImageNet, non-TLS, and clinical models (p < 0.05). The AUC value of TLS-WSI in training cohort was 0.9459 (95CI%: 0.9054, 0.9863) and ranged from 0.8050 (95CI%: 0.7130, 0.8969) to 0.8984 (95CI%: 0.8420, 0.9547) in validation cohorts. TLS-WSI and the nodular or irregular outer layer of gastric wall were screened to construct TLRM. The AUC value of TLRM in training cohort was 0.9643 (95CI%: 0.9349, 0.9936) and ranged from 0.8561 (95CI%: 0.7571, 0.9552) to 0.9195 (95CI%: 0.8670, 0.9721) in validation cohorts. The IDI and DCA showed that the performance of TLRM outperformed the other models. TLS-WSI can be used to predict postoperative recurrence in AGC, whereas TLRM is more effective. TL can effectively improve the performance of clinical research models with a small sample size.

A Lightweight Network with Domain Adaptation for Motor Imagery Recognition.

Brain-computer interfaces (BCI) are an effective tool for recognizing motor imagery and have been widely applied in the motor control and assistive operation domains. However, traditional intention-recognition methods face several challenges, such as prolonged training times and limited cross-subject adaptability, which restrict their practical application. This paper proposes an innovative method that combines a lightweight convolutional neural network (CNN) with domain adaptation. A lightweight feature extraction module is designed to extract key features from both the source and target domains, effectively reducing the model's parameters and improving the real-time performance and computational efficiency. To address differences in sample distributions, a domain adaptation strategy is introduced to optimize the feature alignment. Furthermore, domain adversarial training is employed to promote the learning of domain-invariant features, significantly enhancing the model's cross-subject generalization ability. The proposed method was evaluated on an fNIRS motor imagery dataset, achieving an average accuracy of 87.76% in a three-class classification task. Additionally, lightweight experiments were conducted from two perspectives: model structure optimization and data feature selection. The results demonstrated the potential advantages of this method for practical applications in motor imagery recognition systems.

Angle-based transfer support matrix machine for roller bearing fault diagnosis under limited labeled data

In real-world fault diagnosis, the process of acquiring fault vibration signals frequently faces various challenges, which lead to certain fault types being represented by a scarcity of vibration signal data. Although support matrix machine (SMM) exhibits excellent classification ability in matrix-based data, its ideal model is difficult to establish with limited labeled samples. To address this limitation, a novel angle-based transfer support matrix machine (ATSMM) is proposed, which integrates matrix-based classifier and transfer learning theory. ATSMM employs an angle-based transfer learning method, which aims to reduce the angle between the normals of the target domain hyperplane and the source domain hyperplane, enabling the transfer of structural information across different domains. Meantime, ATSMM effectively addresses the issue of negative transfer that frequently occurs in transfer learning, which adds the weights calculated by the maximum mean discrepancy method to further improve the transfer effectiveness of the target domain model. The superiority of ATSMM is verified by two different roller bearing datasets, and the outcomes of experiments exhibit that the proposed ATSMM has excellent classification performance compared to alternative methods, especially in scenarios where the labeled sample size is limited.

Metaphorical Construction of Chinese Female Images in Media

Introduction. Media is closely related to discourse. Female metaphor in the media is not a simple reflection of the objective world, but a selective and conscious construction of female image in social reality. This study explores the metaphorical construction of Chinese female images in the blessings for women on “Women’s Day” through the media.Methodology and sources. While some words similar to the meanings of “Women’s Day” were selected as keywords, the researcher searched the Wechat platform for articles with the title including these keywords. Then, a total of 306 female metaphors were manually identified and selected as a corpus. With word frequency statistics of the corpus, this study sorted out the categories of Chinese female metaphors and analyzed female image constructed in the media.Results and discussion. It was found that the source domains of female metaphors mainly include plant, animal, daily object, fictional character, nature and environment. The female metaphors can be analyzed from the following categories: Woman is a Flower; Woman is Water; Woman is Half the Heaven; Woman is Light; Woman is a Fairy; Woman is a Book; Woman is a Tiger; Woman is a Jewel. Among them, “Woman is a Flower” is the most frequent one and dominates the female metaphors, while the beauty of “flowers” reflects the beauty of women, the blooming of “flowers” reflects the youth of women, and the aroma of “flowers” reflects the charming of women. “Woman is Water”, “Woman is Half of the Heaven”, “Woman is Light” are also frequent female metaphors, constructing the tenderness, high social status, unique charm and outstanding contribution of female. Other female metaphors also play a certain role in the construction of holiness, wisdom, independence, confidence, strength, precious character of female image. These metaphors construct diversified images in terms of the appearance, character, ability and value of women, and most of the women constructed are beautiful and gentle.Conclusion. It can be seen that the construction of female image in the media becomes diversified on the one hand, on the other hand it still follows the traditional thinking pattern. The paper suggests that gender discourse in the media has been diversified and improved, but still needs to strengthen the construction of equal female image in the harmonious society.

A CRITICAL METAPHOR ANALYSIS OF ENVIRONMENTAL DISCOURSES IN THE PHILIPPINE SETTING

This research employed Critical Metaphor Analysis to examine the inaccuracies embedded in metaphors within the environmental discourses in the Philippines from a corpus of 30 documentary films. The study involved interviews with 5 science teachers and 5 senior high school students in the metaphor analyses. The pre-metaphor analysis level encompassed the discussions on pollution, climate change, biodiversity loss, and nature conservation. During the metaphor identification level, 35 conceptual metaphors containing inaccuracies were identified such as paradise, festival, treasure, refuge, landscape, society, patient, lungs, psychological disorder, sinking ship, waterworld, victims, battle, karma, last forest, symbols of freedom, burning, and simple solution domains. These are grouped into 12 conceptual keys such as ideal place, home, threat, interconnected system, living organism, monster, war, cause and effect, unrestricted movement, combustion, and simple solution source domains. At the metaphor interpretation phase, topics comprised conservation, preservation, and natural value; resource value and commodification of nature; waste and pollution; climate change and environmental degradation; and human influence and environmental ethics. The identified metaphors were characterized by orientational (5), ontological (17), and structural (13) mappings, with 23 metaphors conveying alarmist tones and 12 relational and hopeful tones. Purposes were found as ideological (9), empathetic (10), heuristic (7), and aesthetic (9). At the metaphor explanation level, ideologies such as human dominance, commodification, environmental ethics, sanctity of nature, and overwhelming burden were revealed reflecting modify (2), adapt (1), and depend (2) power dynamics. This study highlighted the cultural, ideological, and motivational underpinnings of metaphor usage in public media that can significantly shape public perception and environmental action. KEYWORDS: environmental discourses; documentary films; critical metaphor analysis; applied linguistics; Philippines

Unsupervised Domain Adaptation Semantic Segmentation of Remote Sensing Imagery with Scene Covariance Alignment

Remote sensing imagery (RSI) segmentation plays a crucial role in environmental monitoring and geospatial analysis. However, in real-world practical applications, the domain shift problem between the source domain and target domain often leads to severe degradation of model performance. Most existing unsupervised domain adaptation methods focus on aligning global-local domain features or category features, neglecting the variations of ground object categories within local scenes. To capture these variations, we propose the scene covariance alignment (SCA) approach to guide the learning of scene-level features in the domain. Specifically, we propose a scene covariance alignment model to address the domain adaptation challenge in RSI segmentation. Unlike traditional global feature alignment methods, SCA incorporates a scene feature pooling (SFP) module and a covariance regularization (CR) mechanism to extract and align scene-level features effectively and focuses on aligning local regions with different scene characteristics between source and target domains. Experiments on both the LoveDA and Yanqing land cover datasets demonstrate that SCA exhibits excellent performance in cross-domain RSI segmentation tasks, particularly outperforming state-of-the-art baselines across various scenarios, including different noise levels, spatial resolutions, and environmental conditions.

Soft sensor modeling method for Pichia pastoris fermentation process based on substructure domain transfer learning

BackgroundAiming at the problem that traditional transfer methods are prone to lose data information in the overall domain-level transfer, and it is difficult to achieve the perfect match between source and target domains, thus reducing the accuracy of the soft sensor model.MethodsThis paper proposes a soft sensor modeling method based on the transfer modeling framework of substructure domain. Firstly, the Gaussian mixture model clustering algorithm is used to extract local information, cluster the source and target domains into multiple substructure domains, and adaptively weight the substructure domains according to the distances between the sub-source domains and sub-target domains. Secondly, the optimal subspace domain adaptation method integrating multiple metrics is used to obtain the optimal projection matrices Ws and Wt that are coupled with each other, and the data of source and target domains are projected to the corresponding subspace to perform spatial alignment, so as to reduce the discrepancy between the sample data of different working conditions. Finally, based on the source and target domain data after substructure domain adaptation, the least squares support vector machine algorithm is used to establish the prediction model.ResultsTaking Pichia pastoris fermentation to produce inulinase as an example, the simulation results verify that the root mean square error of the proposed soft sensor model in predicting Pichia pastoris concentration and inulinase concentration is reduced by 48.7% and 54.9%, respectively.ConclusionThe proposed soft sensor modeling method can accurately predict Pichia pastoris concentration and inulinase concentration online under different working conditions, and has higher prediction accuracy than the traditional soft sensor modeling method.

Small-Sample Target Detection Across Domains Based on Supervision and Distillation

To address the issues of significant object discrepancies, low similarity, and image noise interference between source and target domains in object detection, we propose a supervised learning approach combined with knowledge distillation. Initially, student and teacher models are jointly trained through supervised and distillation-based approaches, iteratively refining the inter-model weights to mitigate the issue of model overfitting. Secondly, a combined convolutional module is integrated into the feature extraction network of the student model, to minimize redundant computational effort; an explicit visual center module is embedded within the feature pyramid network, to bolster feature representation; and a spatial grouping enhancement module is incorporated into the region proposal network, to mitigate the adverse effects of noise on the outcomes. Ultimately, the model undergoes a comprehensive optimization process that leverages the loss functions originating from both the supervised and knowledge distillation phases. The experimental results demonstrate that this strategy significantly boosts classification and identification accuracy on cross-domain datasets; when compared to the TFA (Task-agnostic Fine-tuning and Adapter), CD-FSOD (Cross-Domain Few-Shot Object Detection) and DeFRCN (Decoupled Faster R-CNN for Few-Shot Object Detection), with sample orders of magnitude 1 and 5, increased the detection accuracy by 1.67% and 1.87%, respectively.

Adaptive Centroid-Connected Structure Matching Network Based on Semi-Supervised Heterogeneous Domain

Heterogeneous domain adaptation (HDA) utilizes the knowledge of the source domain to model the target domain. Although the two domains are semantically related, the problem of feature and distribution differences in heterogeneous data still needs to be solved. Most of the existing HDA methods only consider the feature or distribution problem but do not consider the geometric semantic information similarity between the domain structures, which leads to a weakened adaptive performance. In order to solve the problem, a centroid connected structure matching network (CCSMN) approach is proposed, which firstly maps the heterogeneous data into a shared public feature subspace to solve the problem of feature differences. Secondly, it promotes the overlap of domain centers and nodes of the same category between domains to reduce the positional distribution differences in the internal structure of data. Then, the supervised information is utilized to generate target domain nodes, and the geometric structural and semantic information are utilized to construct a centroid-connected structure with a reasonable inter-class distance. During the training process, a progressive and integrated pseudo-labeling is utilized to select samples with high-confidence labels and improve the classification accuracy for the target domain. Extensive experiments are conducted in text-to-image and image-to-image HDA tasks, and the results show that the CCSMN outperforms several state-of-the-art baseline methods. Compared with state-of-the-art HDA methods, in the text-to-image transfer task, the efficiency has increased by 8.05%; and in the image-to-image transfer task, the efficiency has increased by about 2%, which suggests that the CCSMN benefits more from domain geometric semantic information similarity.

Breast Histopathological Image Classification Based on Auto-Encoder Reconstructed Domain Adaptation

As an effective computer-aided diagnostic tool, deep learning has been successfully applied to the classification of breast histopathological images. However, the performance of the deep model is data-driven, and it is difficult to obtain satisfied results when the number of histopathological images is small and labelling histopathological images is difficult. Moreover, in traditional deep learning methods, the representation of features is monotonous, which leads to the limitation of the classification performance of the model. This study proposes an auto-encoder reconstructed semi-supervised domain adaptation for a breast histopathological image classification algorithm. First, the model was pre-trained and transferred to extract high-level features of the sample images. Then, the encoding and decoding parts of the auto-encoder were used to reconstruct the feature representation learning and the sample feature reconstruction learning, respectively. This ensured that the useful information for the classification was purified and retained. At the same time, the domain discriminator was used to confuse the source and target domain features to enhance the learning ability of the model. Finally, the distribution difference of features at different depths of the auto-encoder was measured to minimize the discrepancy of feature distribution between domains, so as to complete the classification of histopathological images. Compared to the results of the comparative and ablation algorithms from the BreakHis to SNL datasets, the proposed method achieved the best results in terms of F1 score (93.40%), accuracy (95.24%), sensitivity (94.66%), and specificity (95.56%). The experimental results demonstrate that the proposed method achieves a remarkable classification performance.

Unsupervised Temporal Adaptation in Skeleton-Based Human Action Recognition

With deep learning approaches, the fundamental assumption of data availability can be severely compromised when a model trained on a source domain is transposed to a target application domain where data are unlabeled, making supervised fine-tuning mostly impossible. To overcome this limitation, the present work introduces an unsupervised temporal-domain adaptation framework for human action recognition from skeleton-based data that combines Contrastive Prototype Learning (CPL) and Temporal Adaptation Modeling (TAM), with the aim of transferring the knowledge learned from a source domain to an unlabeled target domain. The CPL strategy, inspired by recent success in contrastive learning applied to skeleton data, learns a compact temporal representation from the source domain, from which the TAM strategy leverages the capacity for self-training to adapt the representation to a target application domain using pseudo-labels. The research demonstrates that simultaneously solving CPL and TAM effectively enables the training of a generalizable human action recognition model that is adaptive to both domains and overcomes the requirement of a large volume of labeled skeleton data in the target domain. Experiments are conducted on multiple large-scale human action recognition datasets such as NTU RGB+D, PKU MMD, and Northwestern–UCLA to comprehensively evaluate the effectiveness of the proposed method.

Transfer Learning with a Graph Attention Network and Weighted Loss Function for Screening of Persistent, Bioaccumulative, Mobile, and Toxic Chemicals.

In silico methods for screening hazardous chemicals are necessary for sound management. Persistent, bioaccumulative, mobile, and toxic (PBMT) chemicals persist in the environment and have high mobility in aquatic environments, posing risks to human and ecological health. However, lack of experimental data for the vast number of chemicals hinders identification of PBMT chemicals. Through an extensive search of measured chemical mobility data, as well as persistent, bioaccumulative, and toxic (PBT) chemical inventories, this study constructed comprehensive data sets on PBMT chemicals. To address the limited volume of the PBMT chemical data set, a transfer learning (TL) framework based on graph attention network (GAT) architecture was developed to construct models for screening PBMT chemicals, designating the PBT chemical inventories as source domains and the PBMT chemical data set as target domains. A weighted loss (LW) function was proposed and proved to mitigate the negative impact of imbalanced data on the model performance. Results indicate the TL-GAT models outperformed GAT models, along with large coverage of applicability domains and interpretability. The constructed models were employed to identify PBMT chemicals from inventories consisting of about 1 × 106 chemicals. The developed TL-GAT framework with the LW function holds broad applicability across diverse tasks, especially those involving small and imbalanced data sets.

Data imbalanced fault diagnosis of gearbox transmission system under various speeds based on dynamic dual-scale normalized fusion network

Abstract Gearboxes play a pivotal role in industrial production, and their reliability and safety are essential for production safety and efficiency. However, gearboxes frequently encounter challenges such as variable rotational speeds and unknown operating conditions. Unfortunately, most existing traditional fault diagnosis methods face the following issues: (1) They heavily rely on expert experience and pre-existing knowledge bases, making them unable to tackle fault diagnosis in unknown working conditions. (2) While addressing various speed issues, they seldom consider the problem of data imbalance in real-world industrial environments. (3) Many transfer learning methods primarily focus on global distribution alignment and knowledge transfer between source and target domains, neglecting the importance of fine-grained distribution alignment between subdomains. To address these issues, a dynamic dual-scale normalization fusion network (DSNFNet) is proposed for cross-domain fault diagnosis under variable speed and data imbalance. Firstly, the two parallel graph convolution frameworks constructed are used to extract multi-scale fault features. Subsequently, a dual-scale normalization fusion module is adopted to integrate the global and local fault feature information within the subdomains of both the source and target domains, thereby aligning their fine-grained distributions to obtain domain-invariant features. Finally, a dynamic soft threshold feedback strategy is introduced, which assigns pseudo labels to the target domain data, enabling the model to give equal attention to each class of data samples, even under data imbalance conditions, thereby improving the recognition accuracy of minority fault classes. Validating the proposed method on two real cases, our method achieved the highest accuracy compared to other advanced approaches.

MDRN: Multi‐domain representation network for unsupervised domain generalization

AbstractIn deep neural networks, performance can degrade when test data distributions differ from training data. Unsupervised Domain Generalization (UDG) aims to improve generalization across unseen domains by leveraging multiple source domains without supervision. Traditional methods focus on extracting domain‐invariant features, potentially at the expense of feature space integrity and generalization potential. We presents a Multi‐Domain Representation Network (MDRN) for unsupervised multi‐domain learning. MDRN innovates by disentangling and preserving both domain‐invariant and domain‐specific features through an unsupervised cross‐domain reconstruction task. It employs content encoders for domain‐invariant features and multi‐domain style encoders for domain‐specific characteristics. By merging these features based on domain similarity, MDRN constructs a comprehensive feature space that enhances image reconstruction across domains. Additionally, MDRN integrates domain‐specific classifiers, which learn domain classification and provide weighted fusion of domain‐specific features. This design facilitates effective inter‐domain distance measurement and feature integration. Experiments on PACS and DomainNet show MDRN's superior performance over existing state‐of‐the‐art UDG approaches, highlighting its effectiveness in handling distribution shifts between source and target domains.

MSOTL: Multi-source online transfer learning-enabled performance degradation assessment of machinery

Performance degradation assessment is an important component of predictive maintenance for manufacturing systems and is crucial in determining whether they will persist to the overhaul check stage. However, different manufacturing systems’ feature space varies widely across service and degradation processes. Moreover, in real industrial scenarios, data arrives in real-time in a sequential manner. There is a bottleneck in how the model is being updated in real-time when the data is acquired online. To address the above issues, a novel degradation performance assessment method based on online transfer learning (OTL-DPA) is proposed. The initial transformed matrix is calculated between each source domain and partial target domain to obtain the mapped feature space, based on which the k-nearest neighbor (K-NN) classifier is constructed. When the sequential instances in the remaining target domain arrive, corresponding feature representations are obtained using the above-transformed matrices to reduce the distribution discrepancy. The final assessment results are obtained by integrating the multiple K-NN classifiers on mapped feature spaces. The key of OTL-DPA is to update the transformed matrices and classifier weights each time for a certain amount of samples. Experimental evaluation on fault type diagnosis and fault severity assessment cases indicates that OTL-DPA significantly outperforms non-transfer and general transfer learning methods in the bearing online assessment problems where data were obtained randomly or in a time sequence.

Fundus Vascular Segmentation Based on Data Enhancement and Invariant Feature Extraction

Deep neural networks have emerged as the predominant method for medical image segmentation, owing to their robust feature learning capabilities, enabling accurate automatic segmentation of target structures within intricate medical images. Fundus images, being crucial in diagnosing ophthalmic diseases, underscore the importance of effective segmentation techniques. However, fundus vascular images pose challenges due to their high complexity and subtle individual differences, necessitating improvement in existing segmentation methodologies for enhanced disease classification accuracy. This paper introduces a fundus blood vessel segmentation model, employing data augmentation and invariant feature extraction, to systematically tackle the core challenges in medical image processing, particularly in fundus blood vessel segmentation. These challenges include limited source domain samples and inadequate model domain generalization. The model adopts a dual-dimensional strategy. Firstly, it delves into data augmentation technology to enhance the diversity and representativeness of samples within the finite source domain. This is achieved through an image enhancement module based on Fourier transform, mitigating the impact of data scarcity on model training effectiveness. Secondly, the research focuses on interdomain invariant feature extraction, aiming to extract feature representations that consistently characterize fundus blood vessel structure and pathology across different data distributions, thereby enhancing model generalization performance in unfamiliar domains. Specifically, the paper designs a fundus image enhancement module based on Fourier transform in the data augmentation dimension. In the feature extraction dimension, it proposes a normalization module based on uncertainty theory, departing from traditional normalization methods. Experimental results demonstrate the efficacy of the proposed method, showcasing superior generalization performance compared to existing techniques in retinal blood vessel and OD/OC segmentation tasks. Experience validates the model-agnostic nature of the learned strategy, indicating its potential for seamless transferability to other models, thereby offering robust support for advancing medical image segmentation research and applications.