Feature Extraction Mechanism Research Articles

Visual SLAM system based on dynamic semantic features

Aiming at the problem that dynamic objects (such as pedestrians, vehicles, and animals) in real scenes affect the positioning and mapping accuracy of visual SLAM (Simultaneous Localization and Mapping), a YOLOv3-ORB-SLAM3 algorithm is proposed based on ORB-SLAM3 (Oriented FAST and Rotated BRIEF-Simultaneous Localization and Mapping 3). The algorithm adds a semantic thread on the basis of ORB-SLAM3, and adopts a dual-thread mechanism for dynamic and static scene feature extraction: the semantic thread uses YOLOv3 to perform semantic recognition target detection on dynamic objects in the scene, and removes outliers from the extracted dynamic area feature points; the tracking thread extracts scene area features through ORB features, and obtains static scene features combined with semantic information and sends them to the backend, thereby eliminating the interference of dynamic scenes on the system and improving the positioning accuracy of the visual SLAM algorithm. Verified using the TUM (Technical University of Munich) dataset, the results show that the ATE (Average Treatment Effect) index of the YOLOv3-ORB-SLAM3 algorithm in the monocular mode is about 30% lower than that of the ORB-SLAM3 algorithm, and the ATE index of the dynamic sequence in the RGB-D (Red, Green and Blue-Depth) mode is reduced 10%, while the static sequence has no significant decrease.

AI-based automated breast cancer segmentation in ultrasound imaging based on Attention Gated Multi ResU-Net

Breast cancer is a leading cause of death among women worldwide, making early detection and diagnosis critical for effective treatment and improved patient outcomes. Ultrasound imaging is a common diagnostic tool for breast cancer, but interpreting ultrasound images can be challenging due to the complexity of breast tissue and the variability of image quality. This study proposed an Attention Gated Multi ResU-Net model for medical image segmentation tasks, that has shown promising results for breast cancer ultrasound image segmentation. The model’s multi-scale feature extraction and attention-gating mechanism enable it to accurately identify and segment areas of abnormality in the breast tissue, such as masses, cysts, and calcifications. The model’s quantitative test showed an adequate degree of agreement with expert manual annotations, demonstrating its potential for improving early identification and diagnosis of breast cancer. The model’s multi-scale feature extraction and attention-gating mechanism enable it to accurately identify and segment areas of abnormality in the breast tissue, such as masses, cysts, and calcifications, achieving a Dice coefficient of 0.93, sensitivity of 93%, and specificity of 99%. These results underscore the model’s high precision and reliability in medical image analysis.

Advancing research on odor-induced sweetness enhancement: A EEG local-global fusion transformer network for sweetness quantification combined with EEG technology

Reducing sugar intake is crucial for health, and odor sweetening enhances food enjoyment and quality perception. Current research relies on subjective manual sensory evaluations, which are poorly reproducible. Traditional methods also fail to capture dynamic neural responses to odor-induced sweetness. We propose an electroencephalogram local-global fusion transformer network (EEG-LGFNet) model to decode this impact objectively. Electroencephalogram data were collected from 16 subjects under different odor and sucrose stimuli. The model captures complex neural signals by integrating local and global feature extraction mechanisms. Its performance was validated across three-time windows, demonstrating efficacy over various temporal ranges. Analysis of the coefficient of determination across brain regions confirmed the importance of the frontal, central, and parietal areas of sweetness perception. The EEG-LGFNet model excelled in quantifying odor-enhanced sweetness, significantly outperforming state-of-the-art models. This research offers new insights into odor sweetening, with applications in food development, personalized nutrition, and neuroscience.

Fractional Gradient Optimized Explainable Convolutional Neural Network for Alzheimer's Disease Diagnosis

Alzheimer’s is one of the brain syndromes that steadily effects the brain memory. The early stage of Alzheimer’s disease referred as mild-cognitive-impairment (MCI) and the growth of Alzheimer’s is not certain in the patients with MCI. The premature Alzheimer’s diagnosis detection is crucial for the healthy brains functionality by avoiding memory loss. Different multi-neural network architectures have been suggested by the researchers for efficient and accurate Alzheimer’s disease (AD) detection. The absence of improved feature extraction mechanisms and unexplored efficient optimizers in complex benchmark architectures lead to an inefficient and inaccurate AD classification. Moreover, the standard Convolutional Neural Network (CNN)-based architectures for Alzheimer’s diagnosis lack interpretability in their predictions. An interpretable, simplified yet effective deep learning model is required for accurate classification of Alzheimer’s disease. In this study, a generalized fractional order-based CNN classifier with explainable artificial intelligence (XAI) capabilities is proposed for accurate, efficient and interpretable classification of AD diagnosis. The proposed study (a) classifies AD accurately by incorporating unexplored pooling technique with enhanced feature extraction mechanism (b) provides fractional order-based optimization approach for adaptive learning and fast convergence speed (c) suggests an interpretable method for proving the transparency of the model. The proposed model outclasses complex benchmark architectures with regard to accuracy using standard ADNI dataset. The suggested fractional order-based CNN classifier achieves an improved accuracy of 99% as compared to the state-of-the-art models.

Introducing a novel dataset for facial emotion recognition and demonstrating significant enhancements in deep learning performance through pre-processing techniques

Facial expression recognition (FER) plays a pivotal role in various applications, ranging from human-computer interaction to psychoanalysis. To improve the accuracy of facial emotion recognition (FER) models, this study focuses on enhancing and augmenting FER datasets. It comprehensively analyzes the Facial Emotion Recognition dataset (FER13) to identify defects and correct misclassifications. The FER13 dataset represents a crucial resource for researchers developing Deep Learning (DL) models aimed at recognizing emotions based on facial features. Subsequently, this article develops a new facial dataset by expanding upon the original FER13 dataset. Similar to the FER + dataset, the expanded dataset incorporates a wider range of emotions while maintaining data accuracy. To further improve the dataset, it will be integrated with the extended Cohn-Kanade (CK+) dataset.This paper investigates the application of modern DL models to enhance emotion recognition in human faces. By training a new dataset, the study demonstrates significant performance gains compared with its counterparts. Furthermore, the article examines recent advances in FER technology and identifies critical requirements for DL models to overcome the inherent challenges of this task effectively. The study explores several DL architectures for emotion recognition in facial image datasets, with a particular focus on convolutional neural networks (CNNs). Our findings indicate that complex architecture, such as EfficientNetB7, outperforms other DL architectures, achieving a test accuracy of 78.9 %. Notably, the model surpassed the EfficientNet-XGBoost model, especially when used with the new dataset. Our approach leverages EfficientNetB7 as a backbone to build a model capable of efficiently recognizing emotions from facial images. Our proposed model, EfficientNetB7-CNN, achieved a peak accuracy of 81 % on the test set despite facing challenges such as GPU memory limitations. This demonstrates the model's robustness in handling complex facial expressions. Furthermore, to enhance feature extraction and attention mechanisms, we propose a new hybrid model, CBAM-4CNN, which integrates the convolutional block attention module (CBAM) with a custom 4-layer CNN architecture. The results showed that the CBAM-4CNN model outperformed existing models, achieving higher accuracy, precision, and recall metrics across multiple emotion classes. The results highlight the critical role of comprehensive and diverse data in enhancing model performance for facial emotion recognition.

Research on fault diagnosis and feature extraction mechanism visualization of rotating machinery based on improved 1D CNN

With the wide application of intelligent equipment in modern industrial production, it is particularly important to study how to intelligently sense the faults of rotating machinery, improve the diagnosis efficiency and enhance the interpretation ability of the diagnosis process. Although the traditional 1-D CNN performs well in fault diagnosis, it has limitations in capturing subtle changes and complex patterns of fault signals, and its interpretability needs to be improved. Therefore, based on the improved ELCNN model, this paper discusses its diagnosis mechanism in depth, aiming at providing a new idea for intelligent fault diagnosis of rotating machinery. The functions of convolutional layer and S-GAP layer in ELCNN are studied and analyzed. Through single-layer linear convolution, ELCNN can adaptively learn the frequency domain features of the signal and realize the lightweight of the model. At the same time, the S-GAP layer enhances the ability of ELCNN to capture the main peak frequency of fault signals through feature sparseness. The experimental results show that the accuracy of ELCNN in frequency domain feature extraction is more than 80 %. The main peak frequency extracted can effectively help engineers understand the basis of model judgment and improve the reliability of the model.

FocalNeXt: A ConvNeXt augmented FocalNet architecture for lung cancer classification from CT-scan images

Early and accurate diagnosis of lung cancer, a life-threatening disease, is critical to the successful treatment of patients with the disease. On the other hand, it is well known that the integration of computer-aided diagnosis (CAD) systems into the diagnostic workflow facilitates the process and improves the diagnostic results of automated systems by reducing the difficulties associated with human observation. In this paper, we present FocalNeXt, a new ConvNeXt-augmented FocalNet architecture specifically designed for automated lung cancer detection from computed tomography (CT) scan images. By combining the strong attention mechanism of FocalNet and the feature extraction mechanism of ConvNeXt within the vision transformer paradigm, FocalNeXt aims to improve diagnostic accuracy. Our comprehensive evaluation of a publicly available Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases (IQ-OTH/NCCD) CT-scan dataset includes rigorous comparisons and an ablation study. FocalNeXt achieved an accuracy of 99.81%, outperforming the state-of-the-art methods. The model also excelled in sensitivity (99.78%), recall (99.36%), and F1-score (99.56%), positioning FocalNeXt as a leading model for lung cancer detection. The ablation study further demonstrated its efficacy and underscored the robustness of FocalNeXt in different configurations. The results underline its potential to contribute to advances in medical imaging and personalized healthcare.

Advanced drone-based weed detection using feature-enriched deep learning approach

This research addresses the pressing challenge of weed identification in agriculture, crucial for ensuring food security in anticipation of a global population exceeding 9.7 billion by 2050. Utilizing drone imagery, we collected a dataset and proposed a customized model to achieve optimal performance. Our proposed model uses strategically modified backbone, neck, and head components, leveraging elements such as Ghost Convolution, BottleNeckCSP, and ECA (Efficient Channel Attention) layers. These modifications enhance the model’s capability to discern intricate patterns in drone imagery, ultimately leading to improved precision in weed detection. We introduce a purposefully crafted dataset to complement the model’s training, and our experiments demonstrate superior performance compared to the baseline models. Our model achieves a precision of 72.5%, recall of 68.0%, and mAP@0.5 of 73.9, showcasing the effectiveness of our approach in enhancing detection accuracy. Leveraging a unique blend of feature extraction mechanisms, our model achieves remarkable accuracy in real-time soybean detection, outperforming established models like RT-DETR (Real-Time DEtection TransfoRmer) and YOLOv10. A detailed ablation study and comparative analysis with different YOLO versions and the transformer-based RT-DETR showcase the effectiveness of the proposed enhancements. Our work signifies a significant step toward advancing the field of precision agriculture, offering a model that is not only adaptive but also robust in identifying and localizing weeds in soybean fields.

Transfer-learning enabled micro-expression recognition using dense connections and mixed attention

Micro-expression recognition (MER) is a challenging computer vision problem, where the limited amount of available training data and insufficient intensity of the facial expressions are among the main issues adversely affecting the performance of existing recognition models. To address these challenges, this paper explores a transfer–learning enabled MER model using a densely connected feature extraction module with mixed attention. Unlike previous works that utilize transfer learning to facilitate MER and extract local facial-expression information, our model relies on pretraining with three diverse macro-expression datasets and, as a result, can: (i) overcome the problem of insufficient sample size and limited training data availability, (ii) leverage (related) domain-specific information from multiple datasets with diverse characteristics, and (iii) improve the model adaptability to complex scenes. Furthermore, to enhance the intensity of the micro-expressions and improve the discriminability of the extracted features, the Euler video magnification (EVM) method is adopted in the preprocessing stage and then used jointly with a densely connected feature extraction module and a mixed attention mechanism to derive expressive feature representations for the classification procedure. The proposed feature extraction mechanism not only guarantees the integrity of the extracted features but also efficiently captures local texture cues by aggregating the most salient information from the generated feature maps, which is key for the MER task. The experimental results on multiple datasets demonstrate the robustness and effectiveness of our model compared to the state-of-the-art.

Monitoring of floating structures using segmentation and object classification of remotely-sensed images

Abstract With the rapid development of Marine aquaculture, remote sensing monitoring of high space separation Marine floating raft aquaculture has become a key technology to ensure aquaculture efficiency and environmental sustainability. A novel multi-scale layered cascade capsule network (MSHCN) model is proposed to improve the accuracy and efficiency of remote sensing image analysis for high space separation Marine floating raft culture. Firstly, based on the basic principle of the capsule network, the traditional convolutional neural network is improved, and a multi-scale feature extraction mechanism is introduced to adapt to the diversity and complexity of images in the culture area. Secondly, by building a hierarchical cascade structure, the model can further improve its ability to capture small and large scale features, so as to better understand the spatial distribution characteristics of the farming environment. In addition, in the experimental part, we verified with real remote sensing data sets, and the results showed that the MSHCN model outperformed the current mainstream methods on several key performance indicators, including classification accuracy, detection speed and model generalization ability. Finally, the potential value of this model in practical application and future optimization direction is discussed, which provides a new idea and tool for the development of remote sensing monitoring technology for high space separation floating raft culture.

Traffic sign recognition using optimized YOLO network based on multi-scale feature extraction and attention mechanism

Traffic sign recognition using optimized YOLO network based on multi-scale feature extraction and attention mechanism

Intelligent accounting optimization method based on meta-heuristic algorithm and CNN.

The evolution of social intelligence has led to the adoption of intelligent accounting practices in enterprises. To enhance the efficiency of enterprise accounting operations and improve the capabilities of accountants, we propose an intelligent accounting optimization approach that integrates meta-heuristic algorithms with convolutional neural networks (CNN). First, we enhance the CNN framework by incorporating document and voucher information into accounting audits, creating a multi-modal feature extraction mechanism. Utilizing these multi-modal accounting features, we then introduce a method for assessing accounting quality, which objectively evaluates financial performance. Finally, we propose an optimization technique based on meta-heuristic principles, combining genetic algorithms with annealing models to improve the accounting system. Experimental results validate our approach, demonstrating an accuracy of 0.943 and a mean average precision (mAP) score of 0.812. This method provides technological support for refining accounting audit mechanisms.

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.

ESMformer: Error-aware self-supervised transformer for multi-view 3D human pose estimation

Multi-view 3D human pose estimation (HPE) currently faces several key challenges. Information from different viewpoints exhibits high variability due to complex environmental factors, posing difficulties in cross-view feature extraction and fusion. Additionally, multi-view 3D labeled pose data is rather scarce, and the impact of input 2D poses on 3D HPE accuracy has received little attention. To address these issues, we propose an Error-aware Self-supervised transformer framework for Multi-view 3D HPE (ESMformer). Firstly, we introduce a single-view multi-level feature extraction module to enhance pose features in individual viewpoints, which incorporates a novel relative attention mechanism for representative feature extraction at different levels. Subsequently, we develop multi-view intra-level and cross-level fusion modules to exploit spatio-temporal feature dependencies among human joints, and progressively fuse pose information from all views and levels. Furthermore, we explore an error-aware self-supervised learning strategy to reduce the model’s reliance on 3D pose annotations and mitigate the impact of incorrect 2D poses. This strategy adaptively selects reliable input 2D poses based on 3D pose prediction errors. Experiments on three popular benchmarks show that ESMformer achieves state-of-the-art results and maintains cost-effective computational complexity. Notably, ESMformer does not rely on any 3D pose annotations or prior human body knowledge, making it highly versatile and adaptable in practical applications.11The code and models are available at https://github.com/z0911k/ESMformer.

Optimization and Application of Improved YOLOv9s-UI for Underwater Object Detection

The You Only Look Once (YOLO) series of object detection models is widely recognized for its efficiency and real-time performance, particularly under the challenging conditions of underwater environments, characterized by insufficient lighting and visual disturbances. By modifying the YOLOv9s model, this study aims to improve the accuracy and real-time capabilities of underwater object detection, resulting in the introduction of the YOLOv9s-UI detection model. The proposed model incorporates the Dual Dynamic Token Mixer (D-Mixer) module from TransXNet to improve feature extraction capabilities. Additionally, it integrates a feature fusion network design from the LocalMamba network, employing channel and spatial attention mechanisms. These attention modules effectively guide the feature fusion process, significantly enhancing detection accuracy while maintaining the model’s compact size of only 9.3 M. Experimental evaluation on the UCPR2019 underwater object dataset shows that the YOLOv9s-UI model has higher accuracy and recall than the existing YOLOv9s model, as well as excellent real-time performance. This model significantly improves the ability of underwater target detection by introducing advanced feature extraction and attention mechanisms. The model meets portability requirements and provides a more efficient solution for underwater detection.

SPCN: An Innovative Soybean Pod Counting Network Based on HDC Strategy and Attention Mechanism

Soybean pod count is a crucial aspect of soybean plant phenotyping, offering valuable reference information for breeding and planting management. Traditional manual counting methods are not only costly but also prone to errors. Existing detection-based soybean pod counting methods face challenges due to the crowded and uneven distribution of soybean pods on the plants. To tackle this issue, we propose a Soybean Pod Counting Network (SPCN) for accurate soybean pod counting. SPCN is a density map-based architecture based on Hybrid Dilated Convolution (HDC) strategy and attention mechanism for feature extraction, using the Unbalanced Optimal Transport (UOT) loss function for supervising density map generation. Additionally, we introduce a new diverse dataset, BeanCount-1500, comprising of 24,684 images of 316 soybean varieties with various backgrounds and lighting conditions. Extensive experiments on BeanCount-1500 demonstrate the advantages of SPCN in soybean pod counting with an Mean Absolute Error(MAE) and an Mean Squared Error(MSE) of 4.37 and 6.45, respectively, significantly outperforming the current competing method by a substantial margin. Its excellent performance on the Renshou2021 dataset further confirms its outstanding generalization potential. Overall, the proposed method can provide technical support for intelligent breeding and planting management of soybean, promoting the digital and precise management of agriculture in general.

Multifunctional Metasurface Polarizers Synthesis Using Effective Data Generation with Adaptive Attention and Space‐To‐Depth Enhanced Network

AbstractConventional approaches for designing metasurfaces with the desired scattering response are often a time‐consuming and onerous process that heavily relies on the design experience and empirical methods. In this paper, a time and resource‐efficient design method is proposed for the synthesis of multiple metasurface polarizers with an efficient automated data generation process by utilizing the structure validation surrogate model (SVSM). Moreover, a novel attention and space‐to‐depth enhanced adaptive tandem network (ASE‐ATN) is proposed which integrates symmetry‐aware space‐to‐depth (SA‐SPD) along with the attention mechanism for efficient feature extraction. Several design examples of metasurfaces with linear polarization to orthogonal linear polarization (LP‐OLP) and/or linear polarization to circular polarization (LP‐CP) conversion capabilities are provided to verify the performance of the proposed method. The feasibility of the proposed method is demonstrated through the fabrication and measurement of a dual‐band dual‐polarization converter.

Weather forecast image classification based on KNN, CNN, and VGG19

Image classification is an important technology in the field of computer vision, with the main objective of categorizing input images into different classes. It is used in many areas, such as weather forecasting. The significance of weather forecast image classification spans several areas, including enhancing weather forecast accuracy, aiding agriculture, optimizing the energy sector, ensuring transportation safety, and informing urban planning and construction. In this study, the author investigates the precision and efficiency of three machine learning algorithmsConvolutional Neural Network (CNN), K-Nearest Neighbor (KNN), and Visual Geometry Group 19 (VGG19)in classifying weather forecast images. The research meticulously elucidates the methodologies underlying each algorithm, outlining their distinct architectures, training processes, and feature extraction mechanisms. The research highlights that the self-built CNN model excels in accuracy and efficiency, while VGG19 achieves higher accuracy (93%) with longer training time. In contrast, the KNN model shows shorter training time (32.50 seconds) with slightly lower accuracy (80%).

A multi-feature spatial–temporal fusion network for traffic flow prediction

The traffic flow prediction is the key to alleviate traffic congestion, yet very challenging due to the complex influence factors. Currently, the most of deep learning models are designed to dig out the intricate dependency in continuous standardized sequences, which are dependent to high requirements for data continuity and regularized distribution. However, the data discontinuity and irregular distribution are inevitable in the real-world practical application, then we need find a way to utilize the powerful effect of the multi-feature fusion rather than continuous relation in standardized sequences. To this end, we conduct the prediction based on the multiple traffic features reflecting the complex influence factors. Firstly, we propose the ATFEM, an adaptive traffic features extraction mechanism, which can select important influence factors to construct joint temporal features matrix and global spatial features matrix according to the traffic condition. In this way, the feature’s representation ability can be improved. Secondly, we propose the MFSTN, a multi-feature spatial–temporal fusion network, which include the temporal transformer encoder and graph attention network to obtain the latent representation of spatial–temporal features. Especially, we design the scaled spatial–temporal fusion module, which can automatically learn optimal fusion weights, further adapt to inconsistent spatial–temporal dimensions. Finally, the multi-layer perceptron gets the mapping function between these comprehensive features and traffic flow. This method helps to improve the interpretability of the prediction. Experimental results show that the proposed model outperforms a variety of baselines, and it can accurately predict the traffic flow when the data missing rate is high.

Enhanced multi view 3D reconstruction with improved MVSNet

Although 3D reconstruction has been widely used in many fields as a key component of environment perception, existing technologies still have the potential for further improvement in 3D scene reconstruction. We propose an improved reconstruction algorithm based on the MVSNet network architecture. To glean richer pixel details from images, we suggest deploying a DE module integrated with a residual framework, which supplants the prevailing feature extraction mechanism. The DE module uses ECA-Net and dilated convolution to expand the receptive field range, performing feature splicing and fusion through the residual structure to retain the global information of the original image. Moreover, harnessing attention mechanisms refines the 3D cost volume's regularization process, bolstering the integration of information across multi-scale feature volumes, consequently enhancing depth estimation precision. When assessed our model using the DTU dataset, findings highlight the network's 3D reconstruction scoring a completeness (comp) of 0.411 mm and an overall quality of 0.418 mm. This performance is higher than that of traditional methods and other deep learning-based methods. Additionally, the visual representation of the point cloud model exhibits marked advancements. Trials on the Blended MVS dataset signify that our network exhibits commendable generalization prowess.