Unified Feature Research Articles

Deep Plug-and-Play Non-Iterative Cluster for 3D Global Feature Extraction

Efficient and accurate point cloud feature extraction is crucial for critical tasks such as 3D recognition and semantic segmentation. However, existing global feature extraction methods for 3D data often require designing different models for different input types (point clouds, voxels, and maps). This article proposes an efficient plug-and-play non-iterative clustering method (NICM) to establish a unified point cloud global feature extraction paradigm suitable for any input type to solve the above problems. The core idea of the NICM is to construct the connection between a single point and other global points based only on the cosine similarity between center points to achieve global feature extraction, which has linear complexity characteristics and can be combined with any existing feature extraction model. Additionally, to better integrate the features extracted by NICM and the original model, this article designs an adaptive feature fusion module is designed based on the gate unit, which retains similar features and effectively fuses dissimilar features based on their importance to downstream tasks. We have applied our method to downstream tasks such as point cloud recognition, part segmentation and scene segmentation. Sufficient experiments have proven that our method can provide comprehensive and robust features for the original model, and effectively improve the performance of downstream tasks.

Deep coupled registration and segmentation of multimodal whole-brain images.

Recent brain mapping efforts are producing large-scale whole-brain images using different imaging modalities. Accurate alignment and delineation of anatomical structures in these images are essential for numerous studies. These requirements are typically modeled as two distinct tasks: Registration and segmentation. However, prevailing methods, fail to fully explore and utilize the inherent correlation and complementarity between the two tasks. Furthermore, variations in brain anatomy, brightness and texture pose another formidable challenge in designing multi-modal similarity metrics. A high-throughput approach capable of overcomingthe bottleneck of multi-modal similarity metric design, while effective leveraging the highly correlated and complementary nature of two tasks is highly desirable. We introduce a deep learning framework for joint registration and segmentation of multi-modal brain images. Under this framework, registration and segmentation tasks are deeply coupled and collaborated at two hierarchical layers. In the inner layer, we establish a strong feature-level coupling between the two tasks by learning a unified common latent feature representation. In the outer layer, we introduce a mutually supervised dual-branch network to decouple latent features and facilitate task-level collaboration between registration and segmentation. Since the latent features we designed are also modality-independent, the bottleneck of designing multi-modal similarity metric is essentially addressed. Another merit offered by this framework is the interpretability of latent features, which allows intuitive manipulation of feature learning, thereby further enhancing network training efficiency and the performance of both tasks. Extensive experiments conducted on both multi-modal and mono-modal datasets of mouse and human brains demonstrate the superiority of our method. The code is available at https://github.com/tingtingup/DCRS. Supplementary data are available at Bioinformaticsonline.

Coastline target detection based on UAV hyperspectral remote sensing images

Timely and accurate monitoring of typical coastal targets using remote sensing technology is crucial for maintaining marine ecological stability. Hyperspectral target detection technology proves to be an effective tool in extracting various typical materials along the coastline. Traditional target detection methods using spectral domain information can effectively retain the intrinsic properties of the material. However, it is difficult to effectively recognize targets in homogeneous regions by using only spectral domain information, which may lead to insufficient utilization of spatial information. In this study, a detector based on signal-to-noise ratio fusion constrained energy minimization with low-rank sparse decomposition (SFLRSD) is proposed. This detector improves the separability of background and target by obtaining spatial domain information from hyperspectral images and fusing spectral domain information. First, total variation regularization and fractional Fourier transform are applied to process spatial and spectral domain information, respectively. The constrained energy minimization (CEM) detector is used to improve the separability between the target and background of the processed data. Then, the background and anomalies are represented as low-rank and sparse components, respectively, using low-rank sparse matrix factorization. This transforms the model solution into a covariance matrix problem, which is then solved using marginal distance difference (MDD) to isolate anomalous parts. Subsequently, the anomaly parts are fused with CEM detector results, weighted by their respective signal-to-noise ratios. This detection model leverages unified hyperspectral image features, enhancing spectral discreteness of anomalous targets and backgrounds. Finally, experiments on custom created hyperspectral dataset show that the proposed method outperforms other baseline methods in terms of visualization and quantitative performance. In this paper, we not only propose a new hyperspectral target detection method, but we also collect three typical marine litter of different materials by means of airborne hyperspectral remote sensing and construct four hyperspectral datasets in a real environment. All the simulation experiments in this paper are conducted in these four datasets.

Predicting the potential associations between circRNA and drug sensitivity using a multisource feature-based approach.

The unique non-coding RNA molecule known as circular RNA (circRNA) is distinguished from conventional linear RNA by having a longer half-life, greater degree of conservation and inherent solidity. Extensive research has demonstrated the profound impact of circRNA expression on cellular drug sensitivity and therapeutic efficacy. There is an immediate need for the creation of efficient computational techniques to anticipate the potential correlations between circRNA and drug sensitivity, as classical biological research approaches are time-consuming and costly. In this work, we introduce a novel deep learning model called SNMGCDA, which aims to forecast the relationships between circRNA and drug sensitivity. SNMGCDA incorporates a diverse range of similarity networks, enabling the derivation of feature vectors for circRNAs and drugs using three distinct calculation methods. First, we utilize a sparse autoencoder for the extraction of drug characteristics. Subsequently, the application of non-negative matrix factorization (NMF) enables the identification of relationships between circRNAs and drugs based on their shared features. Additionally, the multi-head graph attention network is employed to capture the characteristics of circRNAs. After acquiring the characteristics from these three separate components, we combine them to form a unified and inclusive feature vector for each cluster of circRNA and drug. Finally, the relevant feature vectors and labels are inputted into a multilayer perceptron (MLP) to make predictions. The outcomes of the experiment, obtained through 5-fold cross-validation (5-fold CV) and 10-fold cross-validation (10-fold CV), demonstrate SNMGCDA outperforms five other state-of-art methods in terms of performance. Additionally, the majority of case studies have predominantly confirmed newly discovered correlations by SNMGCDA, thereby emphasizing its reliability in predicting potential relationships between circRNAs and drugs.

Transforming Mexico’s Electric Load Infrastructure: A Quantile Transformer Network Deep Learning Approach, 2019-2020

To transform Mexico's electric load infrastructure, accurate electric load forecasts are required that are crucial to efficiently allocate resources, maintain system stability, and manage energy. The purpose of this study is to use the Quantile Transformer Network (QTN) as a novel approach for a deep learning framework for load forecasting, emphasizing its potential and practical consequences in enhancing the accuracy of load forecasting in real-world energy systems. Moreover, it is shown that QTN efficiently captures complex patterns, temporal relationships and interconnections among factors that influence electric load. The dataset utilized consists of past records of energy consumption in the Baja California System in Mexico. It includes several factors such as electricity demand, marginal prices, temporal characteristics, temperature-related variables, seasonal patterns, and holidays. Additionally, QTN is combined with the Rainbow Technique (RT) to manage categorical variables, resulting in the creation of a unified feature called category. In this case, RT examines the connections between descriptive phrases that reflect distinct combinations of categorical factors depending on the load values. Moreover, this research shows that the common Quantile Regression (QR) outperforms QTN in capturing dependencies in sequential data. Finally, several recommendations for promoting Mexico's Electric Load Infrastructure are provided.

DDBFusion: An unified image decomposition and fusion framework based on dual decomposition and Bézier curves

Existing image fusion algorithms mostly concentrate on the design of network architecture and loss functions, and using unified feature extraction strategies while neglecting the division of redundant and effective information. However, for complementary information, unified feature extractor may not appropriate. Thus, this paper presents a unified image fusion algorithm based on Bézier curves image augmentation and hierarchical decomposition, and a self-supervised learning task is constructed to learn the meaningful information. Where Bézier curves aim to simulate different image features and constructed special self-supervised learning samples, so our method does not require task specific data and can be easily trained on public natural image datasets. Meanwhile, our dual decomposition self-supervised training method can bring redundant information filtering capability to the model. During the decomposition stage, we classify and extract different features of the images and only utilize the extracted effective information in the fusion stage, and the decomposition ability of images provides a foundation for advanced visual tasks, such as image segmentation and object detection. Finally, more detailed and comprehensive fusion images are generated, and the existence of redundant information is effectively reduced. The validity of the proposed method is verified through qualitative and quantitative analysis of multiple image fusion tasks, and our algorithm gets the state-of-the-art results on multiple datasets of different image fusion tasks. The code of our fusion method is available at https://github.com/Yukarizz/DDBFusion.

MMDG-DTI: Drug–target interaction prediction via multimodal feature fusion and domain generalization

Recently, deep learning has become the essential methodology for Drug–Target Interaction (DTI) prediction. However, the existing learning-based representation methods embed the prior knowledge encapsulated by the training data and, as such, acquire redundant domain information irrelevant to the DTI prediction, compromising the generalization capability of a deep network to unfamiliar instances. To tackle this limitation, we propose a novel DTI prediction method using Multi-Modal feature fusion and Domain Generalization (MMDG-DTI). Specifically, we first use pre-trained Large Language Models (LLMs) to obtain generalized textual features covering nearly the whole biological text vocabulary, with the capacity to handle unseen samples and extract robust and discriminative features. In parallel, we propose a unified structural feature extractor based on a hybrid Graph Neural Network (GNN). The extractor improves the performance of MMDG-DTI by extracting features from another modality and discarding the representation of domain-specific prior substructures. Then, we integrate the complementary information of LLM and GNN features using an innovative classifier, while enhancing the generalization ability with the help of Domain Adversarial Training (DAT) and contrastive learning. Last, we propose a novel cross-domain evaluation protocol to enhance the predication fidelity in practical applications. The quantitative and visualization results obtained on several benchmarking datasets demonstrate that MMDG-DTI achieves state-of-the-art performance under both single-domain and cross-domain settings, confirming the superiority of the proposed method. The datasets and codes will be available on the project homepage: https://github.com/JU-HuaY/MMDG-DTI.

A novel CatractNetDetect deep learning model for effective cataract classification through data fusion of fundus images

Cataracts are common eye disorders characterized by the clouding of the lens, preventing light from passing through and impairing vision. Various factors, including changes in the lens’s hydration or alterations in its proteins, may contribute to their development. Regular eye examinations conducted by an ophthalmologist or optometrist are imperative for detecting cataracts and other ocular conditions early on. Manual checks by caregivers pose several problems, including subjectivity, human error, and a lack of expertise. Biomedical fusion involves combining or linking various characteristics specific to certain diseases from different medical imaging resources. The primary objectives of this approach in disease classification are to reduce the error rate and increase the number of retrieved features. The aim of this study is to evaluate the outcomes associated with fusing visual features related to left and right eye cataract characteristics. Additionally, we investigate the impact of limited variability in deep learning models, specifically in the classification of cataract fundus versus normal fundus images. To address this issue, this study introduces CataractNetDetect, an innovative multi-label deep learning classification system that fuses feature representations from pairs of fundus images (e.g., left and right eyes) for the automatic diagnosis of various ocular disorders. Our focus is on achieving improved performance by stacking discriminative deep feature representations to combine two fundus images into a unified feature representation. Several deep learning architectures are utilized as feature descriptors, including ResNet-50, DenseNet-121, and Inception-V3, enhancing the resilience and quality of representations. Fine-tuning of these DL architectures is conducted using the ImageNet dataset, followed by an integrated stacking approach combining ResNet-50, DenseNet-121, and Inception-V3 models. The model is trained on the publicly available ODIR-5k dataset, which includes 5000 left/right eye images depicting eight different ocular conditions, ranging from healthy states to uncommon ailments such as cataracts, glaucoma, age-related macular degeneration (AMD), diabetes, hypertension, and myopia abnormalities. Moreover, extensive preprocessing of the images is performed, including data augmentation, noise reduction, contrast enhancement, scaling, and circular border cropping. The CataractNetDetect system demonstrates F1-scores, AUC, and maximum validation scores of 98.0%, 97.9%, and 100%, respectively. This ensemble-based model distinguishes itself by surpassing the performance of conventional established methodologies, including ResNet-50, DenseNet-121, and Inception-V3, thereby underscoring its efficacy in diagnostic applications.

Embedded feature fusion for multi-view multi-label feature selection

With the explosive growth of data sources, multi-view multi-label learning (MVML) has garnered significant attention. However, the task of selecting informative features in MVML becomes more challenging as the dimensionality increase. Existing methods often extract information separately from the consensus part and the complementary part, potentially leading to noise attributed to ambiguous segmentation. In this paper, we propose an embedded feature selection model that combines with two aspects, which are the feature fusion between views and feature enhancement. Firstly, we calculate the adaptive weight of each view based on the local structure relations, and integrate it into one unified feature matrix. Subsequently, the mapping between unified feature matrix and ground-truth label matrix is established. Furthermore, a regularizer for the feature weight of each view is constructed to emphasize its characteristic, respectively. As a result, the relationship for inter-view and intra-view has been simultaneously considered, preserving comprehensive information of features by minimizing the difference between two types of feature weight. Experimental results demonstrate the superior performance of our method in coping with feature selection.

A Generalized Lightweight Intrusion Detection Model With Unified Feature Selection for Internet of Things Networks

ABSTRACTThe applicability of the Internet of Things (IoT) cutting across different domains has resulted in newer “things” acquiring IP connectivity. These things, technically known as IoT devices, are vulnerable to diverse security threats. Consequently, there has been an exponential increase in IoT malware over the past 5 years, and securing IoT devices from such attacks is a pressing concern in the current era. However, the traditional peripheral security measures do not comply with the lightweight security requirements of the IoT ecosystem. Considering this, we propose a lightweight intrusion detection model for IoT networks (LIDM‐IoT) that demonstrates similar efficiency in exposing malicious activities compared with the existing computationally expensive methods. The crux of the proposed model is that it provides efficient attack detection with lower computational requirements in IoT networks. LIDM‐IoT achieves the feat through a novel unified feature selection strategy that unifies filter‐based and embedded feature selection methods. The proposed feature selection strategy reduces the feature space by 94%. Also, we use only the records of a single attack type to build the model using the XGBoost algorithm. We have tested LIDM‐IoT with unseen attack types to ensure its generalized behavior. The results indicate that the proposed model exhibits efficient attack detection, with a reduced feature set, in IoT networks compared with the state‐of‐the‐art models.

Depth Prior-Guided 3D Voxel Feature Fusion for 3D Semantic Estimation from Monocular Videos

Existing 3D semantic scene reconstruction methods utilize the same set of features extracted from deep learning networks for both 3D semantic estimation and geometry reconstruction, ignoring the differing requirements of semantic segmentation and geometry construction tasks. Additionally, current methods allocate 2D image features to all voxels along camera rays during the back-projection process, without accounting for empty or occluded voxels. To address these issues, we propose separating the features for 3D semantic estimation from those for 3D mesh reconstruction. We use a pretrained vision transformer network for image feature extraction and depth priors estimated by a pretrained multi-view stereo-network to guide the allocation of image features within 3D voxels during the back-projection process. The back-projected image features are aggregated within each 3D voxel via averaging, creating coherent voxel features. The resulting 3D feature volume, composed of unified voxel feature vectors, is fed into a 3D CNN with a semantic classification head to produce a 3D semantic volume. This volume can be combined with existing 3D mesh reconstruction networks to produce a 3D semantic mesh. Experimental results on real-world datasets demonstrate that the proposed method significantly increases 3D semantic estimation accuracy.

INP_ESM: Neuropeptide Identification Based on Evolutionary Scale Modeling and Unified Representation Embedding Features.

Neuropeptides are biomolecules with crucial physiological functions. Accurate identification of neuropeptides is essential for understanding nervous system regulatory mechanisms. However, traditional analysis methods are expensive and laborious, and the development of effective machine learning models continues to be a subject of current research. Hence, in this research, we constructed an SVM-based machine learning neuropeptide predictor, iNP_ESM, by integrating protein language models Evolutionary Scale Modeling (ESM) and Unified Representation (UniRep) for the first time. Our model utilized feature fusion and feature selection strategies to improve prediction accuracy during optimization. In addition, we validated the effectiveness of the optimization strategy with UMAP (Uniform Manifold Approximation and Projection) visualization. iNP_ESM outperforms existing models on a variety of machine learning evaluation metrics, with an accuracy of up to 0.937 in cross-validation and 0.928 in independent testing, demonstrating optimal neuropeptide recognition capabilities. We anticipate improved neuropeptide data in the future, and we believe that the iNP_ESM model will have broader applications in the research and clinical treatment of neurological diseases.

Resilient leaderless and leader–follower consensus over random networks through [formula omitted]-hop communication

This paper studies the resilient consensus problems under ℓ-hop communication over directed random networks. We develop novel multi-hop protocols to tackle both leaderless and leader–follower consensus with a single leader when the multiagent system is under both attack from Byzantine agents and failure of communication edges. Our unified framework features a multiple-input multiple-output system design, and generalizes the weighted mean subsequence reduced algorithms, in which extremal values in the ℓ-hop neighborhood of a cooperative agent are censored based on the notion of minimum message cover. We establish conditions on dynamical network structures, under which secure consensus with vector states can be achieved in the sense of almost sure convergence.

Unified feature learning network for few-shot fault diagnosis

Few-shot fault diagnosis aims at diagnosing the state of mechanical signals with only a few training samples. Numerous contemporary approaches incorporate Time–Frequency Images (TFIs) derived from vibration signals to provide a thorough understanding of both the time and frequency domains. Current approaches often neglect the exploration of sample-level features, which hinders the enrichment and refinement of sample features. To this end, we propose a Unified Feature Learning Network (UFLN) designed to comprehensively model TFIs at two levels. We first present a Texture Enhancement Module (TEM) to amplify intricate details, aiding in the extraction of category-level features. Subsequently, we devise a dynamic Feature Selection Module (DFSM), tailored for the extraction of fault-related features. Finally, we develop an Intra-Diversity (ID) loss function to promote intra-class diversity, enriching the representation of each sample. Extensive experiments on three cases have demonstrated the effectiveness of the proposed UFLN approach.

Stories behind decisions: Towards interpretable malware family classification with hierarchical attention

Malware family classification is essential for understanding malware code and behavioral characteristics, significantly reducing analysis time and response time for emerging malware. However, existing malware family classification methods are limited for two reasons. (1) The feature representation learned from a single information source is not sufficiently discriminative for large-scale malware family classification. (2) The lack of interpretability of classification results makes it difficult to provide intuitive guidance for malware analysis. To tackle the issues, we propose MalAtt, a novel malware family classification method for interpretable and large-scale classification. Specifically, a unified feature engineering approach is proposed to obtain integrated embedding from heterogeneous static and dynamic information. Then, a hierarchical attention encoder module is employed to progressively extract semantic features from the static and dynamic embedding. A collaborative attention module is introduced to model the association between high-level static and dynamic features. Finally, MalAtt is pluggable for different classifiers, such as SVM and RF, to implement malware family classification. For evaluations, we carefully collected a dataset from 62 malware families containing over 20,070 in-the-wild samples. MalAtt is proved to be effective when classifying large-scale malware families, with an accuracy of 93.26%. Besides, the ablation study demonstrates the rationality of the proposed modules. Additionally, we provide a case study to show how MalAtt intuitively guides in-depth malware analysis by highlighting partial input features with high attention scores.

Dynamic YOLO for small underwater object detection

The practical application of object detection inevitably encounters challenges posed by small objects. In underwater object detection, a crucial method for marine exploration, the presence of small objects in underwater environments significantly hampers the performance of detection. In this paper, a dynamic YOLO detector is proposed as a solution to alleviate this problem. Specifically, a light-weight backbone network is first constructed based on deformable convolution v3, with some specialized designs for small object detection. Secondly, a unified feature fusion framework based on channel-wise, scale-wise, and spatial-aware attention is proposed to fuse feature maps from different scales. This is particularly critical for detecting small objects since it allows us to fully exploit the enhanced capabilities offered by our proposed backbone network. Finally, a simple but effective detection head is designed to handle the conflict between classification and localization by disentangling and aligning the two tasks. Extensive experiments are conducted on benchmark datasets to demonstrate the effectiveness of the proposed model. Without bells and whistles, dynamic YOLO outperforms the recent state-of-the-art methods by a large margin of +0.8\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$+\\,0.8$$\\end{document} AP and +1.8\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$+\\,1.8$$\\end{document} APS\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ ext {AP}_{S}$$\\end{document} on the DUO dataset. Experimental results on Pascal VOC and MS COCO datasets also demonstrate the superiority of the proposed method. At last, ablation studies are conducted on DUO dataset to validate the effectiveness and efficiency of each design in dynamic YOLO. Source code will be available at https://github.com/chenjie04/Dynamic-YOLO.

Unified structure-aware feature learning for Graph Convolutional Network

Graph Convolutional Network (GCN), as an effective technique for processing graph-structural data, has been widely used in various fields. Label Propagation Algorithm (LPA), also used as a way for message passing on a graph, broadcasts and integrates node labels over the edges. There have been several efforts to explore the joint learning of GCN and LPA for semi-supervised classification tasks. However, they fail to consider the following two points: (1) Capture precise connections and dispel heterophilous relationships between nodes under the mutual supervision of feature and label spaces; (2) Exploit high-order information so that it has varied weight values in affecting distinct nodes. In light of this, we propose a joint framework called structure-aware feature learning for graph convolutional network, which simultaneously takes into account the above two aspects. In specific, the proposed model consists of two modules: feature aggregation and label propagation, aiming to better utilize the information from both label and feature spaces to learn more accurate node representations. As a result, two modules interact with each other to flexibly assign edge weights. Meanwhile, we adopt a structure-aware manner to provide various weights for neighbors of each order of the target node, contributing to optionally capturing high-order information. Experiments indicate that the proposed method outperforms other compared semi-supervised classification models.

Enhancing vessel arrival time prediction: A fusion-based deep learning approach

The logistic community of shippers has struggled to predict the precise arrival time of the seagoing vessels with reliable certainty. While deep-learning approaches are promising, the existing methods fail to provide desirable results due to a shallow prediction architecture. This research work proposes a method to predict vessel arrival time that could eventually be incorporated into an intelligent decision support system that we call Vessel Arrival Time Prediction (VATP). VATP presents a hybrid architecture of Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), Attention mechanism, Dropout, and Dense layers to take advantage of the coarse-grained local features produced by CNN, the longitudinal nature of AIS (long-time dependencies) via LSTM and paying attention to the influence feature on the arrival time. While many prior approaches have relied solely on AIS data, and some incorporated a combination of AIS and vessel information, our method also integrates diverse data sources, including Automatic Identification System (AIS), Augmented information (Ain), and Maritime Weather Data (MWD). Furthermore, only a few methods have considered weather information, often using minimal weather-related features. In contrast, our approach involves a comprehensive range of weather-related features, and besides these data sources, we extracted specially crafted features.To verify the effectiveness of VATP, we conducted our experiment on large-scale datasets. The VATP model obtained a Root Mean Square Error (RMSE) of 10.63 and a Mean Absolute Percentage Error (MAPE) of 35.11 %. Our results demonstrate that VATP achieves significant performance. Furthermore, these positive results demonstrate that i) the accuracy of the VATP approach can be improved using the AIS, MWD, and Ain information; ii) learning from a unified feature set can result in a significant performance improvement compared to learning from a subset of the features; iii) we also obtained a superior performance in comparison with other well-known methods in the literature and various state-of-the-art baseline methods. Finally, our results illustrate the consistent performance of the VATP across different datasets.

TO-UGDA: target-oriented unsupervised graph domain adaptation

Graph domain adaptation (GDA) aims to address the challenge of limited label data in the target graph domain. Existing methods such as UDAGCN, GRADE, DEAL, and COCO for different-level (node-level, graph-level) adaptation tasks exhibit variations in domain feature extraction, and most of them solely rely on representation alignment to transfer label information from a labeled source domain to an unlabeled target domain. However, this approach can be influenced by irrelevant information and usually ignores the conditional shift of the downstream predictor. To effectively address this issue, we introduce a target-oriented unsupervised graph domain adaptive framework for graph adaptation called TO-UGDA. Particularly, domain-invariant feature representations are extracted using graph information bottleneck. The discrepancy between two domains is minimized using an adversarial alignment strategy to obtain a unified feature distribution. Additionally, the meta pseudo-label is introduced to enhance downstream adaptation and improve the model’s generalizability. Through extensive experimentation on real-world graph datasets, it is proved that the proposed framework achieves excellent performance across various node-level and graph-level adaptation tasks.

Development of Multimodal Fusion Technology for Tomato Maturity Assessment.

The maturity of fruits and vegetables such as tomatoes significantly impacts indicators of their quality, such as taste, nutritional value, and shelf life, making maturity determination vital in agricultural production and the food processing industry. Tomatoes mature from the inside out, leading to an uneven ripening process inside and outside, and these situations make it very challenging to judge their maturity with the help of a single modality. In this paper, we propose a deep learning-assisted multimodal data fusion technique combining color imaging, spectroscopy, and haptic sensing for the maturity assessment of tomatoes. The method uses feature fusion to integrate feature information from images, near-infrared spectra, and haptic modalities into a unified feature set and then classifies the maturity of tomatoes through deep learning. Each modality independently extracts features, capturing the tomatoes' exterior color from color images, internal and surface spectral features linked to chemical compositions in the visible and near-infrared spectra (350 nm to 1100 nm), and physical firmness using haptic sensing. By combining preprocessed and extracted features from multiple modalities, data fusion creates a comprehensive representation of information from all three modalities using an eigenvector in an eigenspace suitable for tomato maturity assessment. Then, a fully connected neural network is constructed to process these fused data. This neural network model achieves 99.4% accuracy in tomato maturity classification, surpassing single-modal methods (color imaging: 94.2%; spectroscopy: 87.8%; haptics: 87.2%). For internal and external maturity unevenness, the classification accuracy reaches 94.4%, demonstrating effective results. A comparative analysis of performance between multimodal fusion and single-modal methods validates the stability and applicability of the multimodal fusion technique. These findings demonstrate the key benefits of multimodal fusion in terms of improving the accuracy of tomato ripening classification and provide a strong theoretical and practical basis for applying multimodal fusion technology to classify the quality and maturity of other fruits and vegetables. Utilizing deep learning (a fully connected neural network) for processing multimodal data provides a new and efficient non-destructive approach for the massive classification of agricultural and food products.