Capsule Neural Network Research Articles

Design of automatic sorting system based on capsule neural network

Abstract. This paper proposes a vegetable disease classification system based on Capsule Neural Network (CapsNet), which uses the visual recognition mechanism of deep learning algorithms to improve the efficiency and accuracy of agricultural disease recognition. The CapsNet algorithm outperforms traditional Convolutional Neural Network (CNN) in few-shot learning and generalization due to its unique capsule structure and dynamic routing mechanism. Experimental results show that the proposed system performs well in the accuracy, robustness, and model interpretation of image recognition, and can accurately identify diseases even in the case of poor image quality or noise and occlusion. In addition, the embedded system of this experiment includes an image acquisition module, an algorithm core module, and a logic control module, which realizes the automation of the whole process from image acquisition to recognition result display. This study also discusses the scalability and adaptability of the system, as well as the future application prospects in agricultural intelligence and automation. It is expected that CapsNet will play a more critical role in artificial intelligence, especially in agrarian automation and smart agriculture.

SENTINEL: Securing Indoor Localization Against Adversarial Attacks With Capsule Neural Networks

SENTINEL: Securing Indoor Localization Against Adversarial Attacks With Capsule Neural Networks

Comparison of deep learning-based models for detection of diseased trees using an image compression algorithm

The object of the research is the application of deep learning algorithms using an improved mathematical lossless image compression method for recognizing and identifying dead trees in aerospace images. The main problem that has been solved is the archiving of images due to their large volume on disk and the possibility of their further processing by deep learning methods such as convolutional and capsule neural networks, which have shown high efficiency and accuracy in image recognition and classification tasks using the proposed new image compression method. The article presents a comparative analysis of the performance of three YOLO (You Only Look Once) models with different types of architectures, such as YOLOv5, YOLOv7 and YOLOv8, to assess the effectiveness of their work for the task of recognizing aerospace tree images obtained from satellites, drones, and aircrafts. Comprehensive analysis of YOLO models presents that model YOLO v8 turned out to be most effective with a positive accuracy of 88.2 %, a recall of 77.4 %, and a mAP50 score of 87.2 %. Moreover, the average detection time was only 0.052 seconds for each image, even though the model size remains very small – 21.5 MB. These results suggest a much better usage of time and precise identification of dead trees, and classified targets with high efficiency. From the research, there is significant prospects of global forest management especially on forest reduction and protection of ecosystems through accurate assessment on the health of forestry. The proposed approach is universal and can be used in real life conditions, providing a good compromise of the speed, accuracy and resources required for forest monitoring and management

Object Detection Using Capsule Neural Network: An Overview

Appropriate remedies for tasks like language translation, object identification, object segmentation, picture recognition, and natural language processing are needed for today's computer vision tasks. This paper explores the use of Capsule neural Networks (Caps Nets) in object identification and offers a thorough analysis of their developments and uses. It presents an analysis of the transformative impact of Caps Nets on object identification tasks. This distinctive the architecture of neural network is presented as an alternative to convolutional neural networks (CNNs). The paper highlights that how Caps Nets are unique in capturing spatial correlations and hierarchical patterns in visualization data by analyzing the fundamental ideas of the technology. Additionally, it demonstrates how Caps net out perform CNNs in terms of generalization, interpretability, and last in the resistance to spatial distortions, confirming their goodness at object detection. By integrating the Caps networks with the latest scientific findings and advances, the paper shows the current status and potential future paths for object detection methods that use this leading neural network architecture.

A Spatio-Temporal Capsule Neural Network with Self-Correlation Routing for EEG Decoding of Semantic Concepts of Imagination and Perception Tasks.

Decoding semantic concepts for imagination and perception tasks (SCIP) is important for rehabilitation medicine as well as cognitive neuroscience. Electroencephalogram (EEG) is commonly used in the relevant fields, because it is a low-cost noninvasive technique with high temporal resolution. However, as EEG signals contain a high noise level resulting in a low signal-to-noise ratio, it makes decoding EEG-based semantic concepts for imagination and perception tasks (SCIP-EEG) challenging. Currently, neural network algorithms such as CNN, RNN, and LSTM have almost reached their limits in EEG signal decoding due to their own short-comings. The emergence of transformer methods has improved the classification performance of neural networks for EEG signals. However, the transformer model has a large parameter set and high complexity, which is not conducive to the application of BCI. EEG signals have high spatial correlation. The relationship between signals from different electrodes is more complex. Capsule neural networks can effectively model the spatial relationship between electrodes through vector representation and a dynamic routing mechanism. Therefore, it achieves more accurate feature extraction and classification. This paper proposes a spatio-temporal capsule network with a self-correlation routing mechaninsm for the classification of semantic conceptual EEG signals. By improving the feature extraction and routing mechanism, the model is able to more effectively capture the highly variable spatio-temporal features from EEG signals and establish connections between capsules, thereby enhancing classification accuracy and model efficiency. The performance of the proposed model was validated using the publicly accessible semantic concept dataset for imagined and perceived tasks from Bath University. Our model achieved average accuracies of 94.9%, 93.3%, and 78.4% in the three sensory modalities (pictorial, orthographic, and audio), respectively. The overall average accuracy across the three sensory modalities is 88.9%. Compared to existing advanced algorithms, the proposed model achieved state-of-the-art performance, significantly improving classification accuracy. Additionally, the proposed model is more stable and efficient, making it a better decoding solution for SCIP-EEG decoding.

Evaluating the performance of a non-uniform squash function in Capsule networks for early diabetic retinopathy detection using fundus image analysis

Diabetic retinopathy, known as one of the most crucial neurovascular barriers to diabetes, has affected the retina, leading to a potential risk of blindness. The harm is not isolated to personal health, with negative effects felt at the economic and social level in entire communities. The consequences of vision loss secondary to diabetic retinopathy on quality of life, independent living, and employment can be significant. It is necessary to adopt a comprehensive approach for effective management of underlying diabetes, which includes early detection, timely intervention, supportive care, and management of comorbid conditions. Critical to this is a precise classification of its stages for disease analysis. A manual diagnosis, on the other hand, requires qualified individuals to recognize the subtle nuances of the condition, making it a lengthy process. This study aims to develop a computer-assisted system for analyzing retinal fundus images to identify and classify different stages of diabetic retinopathy. A dataset of eye fundus images was used by the proposed network to classify the presence and severity of blindness and its stages. The proposed model uses a non-uniform squash function in capsule neural networks. The motive behind employing a non-uniform squash function was to encourage the model to allocate greater attention and stability to intricate images throughout the training phase. The proposed model's efficiency is assessed meticulously with pre-trained CNN methods like InceptionV3, VGG16, VGG19, and Xception. The proposed capsule network model has been revamped to address the shortcomings of prior research with care. The method stands out with an impressive 99.84 % training accuracy and 98.68 % validation accuracy. Diabetic retinopathy, though rare in occurrence, poses significant challenges due to the high cost of diagnosis, rendering it inaccessible to rural and remote communities with limited access to tests and check-ups.

Multimodal ischemic stroke recurrence prediction model based on the capsule neural network and support vector machine.

Ischemic stroke (IS) has a high recurrence rate. Machine learning (ML) models have been developed based on single-modal biochemical tests, and imaging data have been used to predict stroke recurrence. However, the prediction accuracy of these models is not sufficiently high. Therefore, this study aimed to collect biochemical detection and magnetic resonance imaging (MRI) data to establish a dataset and propose a high-performance heterogeneous multimodal IS recurrence prediction model based on deep learning. This is a retrospective cohort study. Data were retrospectively collected from 634 IS patients in Zhuhai, China, a 12-month follow-up was conducted to determine stroke recurrence. We propose the ischemic stroke multi-group learning (ISGL) model, an integrated model for predicting the recurrence risk of multimodal IS in patients, based on a capsule neural network and a linear support vector machine (SVM). Two capsule neural network prediction models based on T1 and T2 signals in the MRI data and a SVM prediction model based on biochemical test data were established. Finally, a vote was conducted on the final judgment of the integrated model. The ISGL model was compared with 6 classical ML and deep learning models: k-nearest neighbors, SVM, logistic regression, random forest, eXtreme Gradient Boosting, and visual geometry group. The results revealed that the accuracy, specificity, sensitivity and the area under the curve of the ISGL model were 95%, 96%, 94%, and 95%, respectively. Among the comparison models, the visual geometry group method exhibited the best performance, but it much lower than those of the ISGL model. Analysis of the importance of biochemical test data revealed that low-density lipoprotein, smoking, and heart disease history were the positively correlated factors, and total cholesterol, high-density lipoprotein, and diabetes were and the negatively correlated factors. This study proposes the ISGL model can be used simultaneously with MRI and biochemical data to predict IS recurrence. This combination resulted in higher rate of performance than that of the other ML models. Additionally, this study found related risk factors affected recurrence, which can be used to intervene in high-risk patients' recurrence as early as possible and promote the development of secondary prevention of stroke.

Deep learning based capsule networks for breast cancer classification using ultrasound images

Purposes: Breast cancer (BC) is a disease in which the breast cells multiply uncontrolled. Breast cancer is one of the most often diagnosed malignancies in women worldwide. Early identification of breast cancer is critical for limiting the impact on affected people's health conditions. The influence of technology and artificial intelligence approaches (AI) in the health industry is tremendous as technology advances. Deep learning (DL) techniques are used in this study to classify breast lumps. Materials and Methods: The study makes use of two distinct breast ultrasound images (BUSI) with binary and multiclass classification. To assist the models in understanding the data, the datasets are exposed to numerous preprocessing and hyperparameter approaches. With data imbalance being a key difficulty in health analysis, due to the likelihood of not having a condition exceeding that of having the disease, this study applies a cutoff stage to impact the decision threshold in the datasets data augmentation procedures. The capsule neural network (CapsNet), Gabor capsule network (GCN), and convolutional neural network (CNN) are the DL models used to train the various datasets. Results: The findings showed that the CapsNet earned the maximum accuracy value of 93.62% while training the multiclass data, while the GCN achieved the highest model accuracy of 97.08\% when training the binary data. The models were also evaluated using a variety of performance assessment parameters, which yielded consistent results across all datasets. Conclusion: The study provides a non-invasive approach to detect breast cancer; and enables stakeholders, medical practitioners, and health research enthusiasts a fresh view into the analysis of breast cancer detection with DL techniques to make educated judgements.

A Text-Inception-Based Natural Language Processing Model for Sentiment Analysis of Drug Experiences

The study of sentiment in Natural Language Processing (NLP) is among the most successful research areas because of the availability of millions of user opinions online since the turn of the century. The economic, political, and medical fields are just some of the many that have benefited from studies of sentiment research. While numerous studies have examined more mainstream topics like consumer electronics, movies, and restaurants, relatively few have examined health and medical concerns. Considerable insight into where to direct efforts to improve public health might be gained by a study of how people feel about healthcare as a whole and of individual drug experiences in particular. When it comes to medicine, automatic analysis of online user evaluations paves the way for sifting through massive amounts of user feedback to find information regarding medications' efficacy and side effects that might be used to enhance pharmacovigilance programs. Simple rules-based methods have given way to more complex machine learning approaches like deep learning, which is developing as a technology for many natural language processing jobs. The opensource datasets have been analyzed with models that use word embeddings and term frequency-inverse document frequency (TF-IDF). A feature-enhanced text-inception model for sentiment classification was presented to work in tandem with this approach. The model first employed a cutting-edge text-inception module to glean useful shallow features from the text. K-MaxPooling was subsequently employed to reduce the dimensionality of its shallow and deep includes as well as enhance the generalization of characteristics, and a deep feature extraction module was formed using the bidirectional gated recurrent unit (Bi-GRU) and the capsule neural network to comprehend the text's semantic data. By combining traditional methods with cutting-edge artificial intelligence techniques, this hybrid approach can revolutionize public health initiatives, decision-making, and pharmacovigilance in the healthcare industry. This model achieved an exceptional accuracy rate of 99%, underscoring its effectiveness in sentiment classification and demonstrating its potential to significantly contribute to advancing healthcare and medical research.

Attribute Feature Perturbation-Based Augmentation of SAR Target Data.

Large-scale, diverse, and high-quality data are the basis and key to achieving a good generalization of target detection and recognition algorithms based on deep learning. However, the existing methods for the intelligent augmentation of synthetic aperture radar (SAR) images are confronted with several issues, including training instability, inferior image quality, lack of physical interpretability, etc. To solve the above problems, this paper proposes a feature-level SAR target-data augmentation method. First, an enhanced capsule neural network (CapsNet) is proposed and employed for feature extraction, decoupling the attribute information of input data. Moreover, an attention mechanism-based attribute decoupling framework is used, which is beneficial for achieving a more effective representation of features. After that, the decoupled attribute feature, including amplitude, elevation angle, azimuth angle, and shape, can be perturbed to increase the diversity of features. On this basis, the augmentation of SAR target images is realized by reconstructing the perturbed features. In contrast to the augmentation methods using random noise as input, the proposed method realizes the mapping from the input of known distribution to the change in unknown distribution. This mapping method reduces the correlation distance between the input signal and the augmented data, therefore diminishing the demand for training data. In addition, we combine pixel loss and perceptual loss in the reconstruction process, which improves the quality of the augmented SAR data. The evaluation of the real and augmented images is conducted using four assessment metrics. The images generated by this method achieve a peak signal-to-noise ratio (PSNR) of 21.6845, radiometric resolution (RL) of 3.7114, and dynamic range (DR) of 24.0654. The experimental results demonstrate the superior performance of the proposed method.

STCPU-Net: advanced U-shaped deep learning architecture based on Swin transformers and capsule neural network for brain tumor segmentation

STCPU-Net: advanced U-shaped deep learning architecture based on Swin transformers and capsule neural network for brain tumor segmentation

Hybrid framework for membrane protein type prediction based on the PSSM

Membrane proteins are considered the major source of drug targets and are indispensable for drug design and disease prevention. However, traditional biomechanical experiments are costly and time-consuming; thus, many computational methods for predicting membrane protein types are gaining popularity. The position-specific scoring matrix (PSSM) method is an excellent method for describing the evolutionary information of protein sequences. In this study, we propose an improved capsule neural network (ICNN) model based on a capsule neural network to acquire sufficient relevant information from the PSSM. Furthermore, accounting for the complementarity between traditional machine learning and deep learning, we propose a hybrid framework that combines both approaches to predict protein types. This framework trains 41 baseline models based on the PSSM. The optimal subset features, selected after traversal, are fused using a two-level decision-level feature fusion approach. Subsequently, comparisons are made using three combined strategies within an ensemble learning framework. The experimental results demonstrate that solely relying on PSSM input, the proposed method not only surpasses the optimal methods by 1.52%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document}, 2.26%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} and 2.67%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} on Dataset1, Dataset2, and Datasets3, respectively, but also exhibits superior generalizability. Furthermore, the code and dataset can be free download at https://github.com/ruanxiaoli/membrane-protein-types.

Modified osprey algorithm for optimizing capsule neural network in leukemia image recognition

The diagnosis of leukemia is a serious matter that requires immediate and accurate attention. This research presents a revolutionary method for diagnosing leukemia using a Capsule Neural Network (CapsNet) with an optimized design. CapsNet is a cutting-edge neural network that effectively captures complex features and spatial relationships within images. To improve the CapsNet's performance, a Modified Version of Osprey Optimization Algorithm (MOA) has been utilized. Thesuggested approach has been tested on the ALL-IDB database, a widely recognized dataset for leukemia image classification. Comparative analysis with various machine learning techniques, including Combined combine MobilenetV2 and ResNet18 (MBV2/Res) network, Depth-wise convolution model, a hybrid model that combines a genetic algorithm with ResNet-50V2 (ResNet/GA), and SVM/JAYA demonstrated the superiority of our method in different terms. As a result, the proposed method is a robust and powerful tool for diagnosing leukemia from medical images.

A Security Domain Feature Pattern Recognition Method Based on Capsule Neural Network

Abstract In this paper, a security domain feature pattern recognition method based on capsule neural network is proposed. Firstly, Halton sampling is performed on the active output of all generators except for the balancing machine to obtain a uniform sample in the coverage space. Then, the support vector machine based recursive feature elimination (SVM-RFE) method is used to select the critical pre fault feature quantities of the system. Finally, based on the selected active features, one-dimensional data is converted into two-dimensional data and input into the capsule neural network. Dynamic security domain boundaries are constructed in both two-dimensional and three-dimensional spaces, which can divide the feature space into stable and unstable subspaces.

Prediction of drought hydrological and water scarcity based on optimal artificial intelligence by developing a metaheuristic optimization algorithm

This study presents a novel approach to predicting water stress in Yichang City, China, through an optimized Capsule Neural Network (CapsNN) model enhanced by the Improved Aquila Optimization (IAO) algorithm. The model, applied to monthly water stress data from 2000 to 2022, is gauged using the Streamflow Drought Index (SDI). When compared to traditional metaheuristic models, our CapsNN/IAO framework demonstrates superior accuracy in drought forecasting across various time scales with significant reductions in The Mean Absolute Error is 0.1021, the Root Mean Square Error is 0.1824, and good the Nash-Sutcliffe Efficiency is 0.75, the coefficient of determination R2 is 0.89, and the Willmott's Index is 0.93. Notably, the model predicts short-term water shortages with greater frequency at the SDI3 scale, while identifying the most severe drought conditions at the SDI6 scale. These insights are critical for enhancing water resource management.

The Fault Diagnosis of Rolling Bearings Is Conducted by Employing a Dual-Branch Convolutional Capsule Neural Network.

Currently, many fault diagnosis methods for rolling bearings based on deep learning are facing two main challenges. Firstly, the deep learning model exhibits poor diagnostic performance and limited generalization ability in the presence of noise signals and varying loads. Secondly, there is incomplete utilization of fault information and inadequate extraction of fault features, leading to the low diagnostic accuracy of the model. To address these problems, this paper proposes an improved dual-branch convolutional capsule neural network for rolling bearing fault diagnosis. This method converts the collected bearing vibration signals into grayscale images to construct a grayscale image dataset. By fully considering the types of bearing faults and damage diameters, the data are labeled using a dual-label format. A multi-scale convolution module is introduced to extract features from the data and maximize feature information extraction. Additionally, a coordinate attention mechanism is incorporated into this module to better extract useful channel features and enhance feature extraction capability. Based on adaptive fusion between fault type (damage diameter) features and labels, a dual-branch convolutional capsule neural network model for rolling bearing fault diagnosis is established. The model was experimentally validated using both Case Western Reserve University's bearing dataset and self-made datasets. The experimental results demonstrate that the fault type branch of the model achieves an accuracy rate of 99.88%, while the damage diameter branch attains an accuracy rate of 99.72%. Both branches exhibit excellent classification performance and display robustness against noise interference and variable working conditions. In comparison with other algorithm models cited in the reference literature, the diagnostic capability of the model proposed in this study surpasses them. Furthermore, the generalization ability of the model is validated using a self-constructed laboratory dataset, yielding an average accuracy rate of 94.25% for both branches.

A novel decoupling method of compound faults with incomplete dataset of rotating machinery

Abstract The occurrence of fault in rotating machinery is random and complex, and the diagnosis of the compound faults has been a challenge in industrial production. Accurate diagnosis of the compound faults can be of significant help to practical maintenance and management. However, most existing intelligent diagnostic methods typically require abundant data for training, which is often difficult to collect for compound faults. In this paper, a novel method called impact feature-based decoupling capsule network (IFDCN) is proposed for diagnosing compound faults with incomplete datasets. In this model, an improved Laplace wavelet kernel capsule neural network is proposed to extract and enhance the impact features of vibration signal. A decoupling classifier is designed to decouple the compound faults in the diagnostic process so as to identify the sub-faults contained in the compound faults. In using this proposed model for incomplete datasets, the compound fault data is not trained and is not labeled. Through training on single-fault data, the proposed model is capable of classifying and decoupling the fault types. The feature extraction capability of the network is visualized by heat maps, and the physical significance of feature extraction is explained by deep learning network. The effectiveness of IFDCN is verified through different experimental of gear and bearing and the experiment results indicate that the proposed model has higher identifying precision and can accurately decouple the compound faults without compound fault samples.

Novel hybrid classifier based on fuzzy type-III decision maker and ensemble deep learning model and improved chaos game optimization

This paper presents a unique hybrid classifier that combines deep neural networks with a type-III fuzzy system for decision-making. The ensemble incorporates ResNet-18, Efficient Capsule neural network, ResNet-50, the Histogram of Oriented Gradients (HOG) for feature extraction, neighborhood component analysis (NCA) for feature selection, and Support Vector Machine (SVM) for classification. The innovative inputs fed into the type-III fuzzy system come from the outputs of the mentioned neural networks. The system’s rule parameters are fine-tuned using the Improved Chaos Game Optimization algorithm (ICGO). The conventional CGO’s simple random mutation is substituted with wavelet mutation to enhance the CGO algorithm while preserving non-parametricity and computational complexity. The ICGO was evaluated using 126 benchmark functions and 5 engineering problems, comparing its performance with well-known algorithms. It achieved the best results across all functions except for 2 benchmark functions. The introduced classifier is applied to seven malware datasets and consistently outperforms notable networks like AlexNet, ResNet-18, GoogleNet, and Efficient Capsule neural network in 35 separate runs, achieving over 96% accuracy. Additionally, the classifier’s performance is tested on the MNIST and Fashion-MNIST in 10 separate runs. The results show that the new classifier excels in accuracy, precision, sensitivity, specificity, and F1-score compared to other recent classifiers. Based on the statistical analysis, it has been concluded that the ICGO and propose method exhibit significant superiority compared to the examined algorithms and methods. The source code for ICGO is available publicly at https://nimakhodadadi.com/algorithms-%2B-codes.Graphical abstract

A Novel Optimization Approach for Revolutionizing Architectural Design in Chinese Cultural Heritage

Chinese cultural heritage architecture, a blend of traditional and modern construction methods, showcases the country's technological, artistic, and cultural advancements, yet its preservation faces challenges due to structural decay and aesthetic degeneration. Using Microtrans Maryland 4-1000, Dazu Rock Carvings, Nanchan Temple and Foguang Temple smokescreen were captured; surface images and ground photos were captured with smokescreen resolution of 5192 x 4153 pixels. The goal of the research is to use an Ensemble Ant Colony Fused Convolutional Capsule Neural Network (EAC-CCNN) to enhance fault analysis in Chinese cultural heritage structures images, then the combination of Augmented Reality (AR), and Building Information Modeling (BIM) to enhance the designing model for the safety management and decision making. The process entails gathering and annotating a variety of information, creating a hybrid EAC-CCNN model to investigate the architectural building's problem, training it, integrating it with BIM, doing on-site inspections, and utilizing AR-enhanced BIM models to analyze the flaws found. The findings demonstrate how this integrated method improves the precision of flaw diagnosis, fosters teamwork, and aids in the upkeep and preservation of cultural heritage assets. Metrics including accuracy and F1 score are used to evaluate the machine learning model recognizes and categorizes flaws in Chinese cultural heritage architecture. The defect identifying and safety management model of architectural designing outcome of the accuracy is 93.29% and F1 Score is 95.47%. During the training, validation, and testing phases, performance is evaluated by project goals. With the help of this method, machine learning models can be trained to identify patterns, identify flaws, and generate well-informed predictions in a variety of circumstances.

MES-CTNet: A Novel Capsule Transformer Network Base on a Multi-Domain Feature Map for Electroencephalogram-Based Emotion Recognition.

Emotion recognition using the electroencephalogram (EEG) has garnered significant attention within the realm of human-computer interaction due to the wealth of genuine emotional data stored in EEG signals. However, traditional emotion recognition methods are deficient in mining the connection between multi-domain features and fitting their advantages. In this paper, we propose a novel capsule Transformer network based on a multi-domain feature for EEG-based emotion recognition, referred to as MES-CTNet. The model's core consists of a multichannel capsule neural network(CapsNet) embedded with ECA (Efficient Channel Attention) and SE (Squeeze and Excitation) blocks and a Transformer-based temporal coding layer. Firstly, a multi-domain feature map is constructed by combining the space-frequency-time characteristics of the multi-domain features as inputs to the model. Then, the local emotion features are extracted from the multi-domain feature maps by the improved CapsNet. Finally, the Transformer-based temporal coding layer is utilized to globally perceive the emotion feature information of the continuous time slices to obtain a final emotion state. The paper fully experimented on two standard datasets with different emotion labels, the DEAP and SEED datasets. On the DEAP dataset, MES-CTNet achieved an average accuracy of 98.31% in the valence dimension and 98.28% in the arousal dimension; it achieved 94.91% for the cross-session task on the SEED dataset, demonstrating superior performance compared to traditional EEG emotion recognition methods. The MES-CTNet method, utilizing a multi-domain feature map as proposed herein, offers a broader observation perspective for EEG-based emotion recognition. It significantly enhances the classification recognition rate, thereby holding considerable theoretical and practical value in the EEG emotion recognition domain.