Enhancement Characteristics Research Articles

Motion-guided small MAV detection in complex and non-planar scenes

In recent years, there has been a growing interest in the visual detection of micro aerial vehicles (MAVs) due to its importance in numerous applications. However, the existing methods based on either appearance or motion features encounter difficulties when the background is complex or the MAV is too small. In this paper, we propose a novel motion-guided MAV detector that can accurately identify small MAVs in complex and non-planar scenes. This detector first exploits a motion feature enhancement module to capture the motion features of small MAVs. Then it uses multi-object tracking and trajectory filtering to eliminate false positives caused by motion parallax. Finally, an appearance-based classifier and an appearance-based detector that operates on the cropped regions are used to achieve precise detection results. Our proposed method can effectively and efficiently detect extremely small MAVs from dynamic and complex backgrounds because it aggregates pixel-level motion features and eliminates false positives based on the motion and appearance features of MAVs. Experiments on the ARD-MAV dataset demonstrate that the proposed method could achieve high performance in small MAV detection under challenging conditions and outperform other state-of-the-art methods across various metrics.

MLFEU-NET: A Multi-scale Low-level Feature Enhancement Unet for breast lesions segmentation in ultrasound images

Breast lesions constantly threaten the health of females. Segmentation methods of breast lesions are important for clinical diagnosis, and neural networks have been widely used and played a significant role in this field. However, false detections and miss detections still exist due to the variability in the shape and location of breast lesions, artifacts and noise in breast ultrasound (BUS) images, and structural defects in conventional segmentation networks. In this paper, a multi-scale low-level feature enhancement Unet (MLFEU-net) structure is presented, consisting of the U-net structure, low-level feature enhancement block (LFEB), and a parallel multiscale feature fusion (PMFF) module. Specifically, LFEB enhances the detail information during shallow downsampling and further feature selection. Meanwhile, in the neck of Unet, PMFF module uses different scales of dilation convolution to provide different sizes of sensory fields, and to efficiently merge shallow and deep features filtered, leads to more accurate segmentation results. To evaluate the MLFEU-net’s segmentation ability, five quantitative metrics were used on two breast ultrasound datasets, and it was compared with several advanced segmentation techniques. The results illustrate that our approach surpasses other methods in performance. Moreover, robustness experiments validate the effectiveness of our approach in achieving robust segmentation of breast lesions.

Spectral domain strategies for hyperspectral super-resolution: Transfer learning and channel enhance network

As the network structures continue to innovate and evolve, significant achievements have been achieved in hyperspectral image super-resolution tasks. However, how to further explore the spectral domain potential from prior knowledge and channel-enhanced structures to achieve better performance has inspired the following two works: Firstly, to systematically compare prior knowledge of spectral with spatial domain for HSI-SR tasks, four transfer learning strategies are proposed. The superior performance of the Relevant Channel/Random Space (RCRS) strategy reveals the importance of spectral feature reconstruction in HSI-SR tasks. Meanwhile, an interesting phenomenon has been observed that even without training on real datasets, the model can already exhibit a core or even decent super-resolution capability based solely on prior knowledge of above four strategies. Secondly, a dual-branch channel network with complementary channel feature extraction (CCFE) and adjacent channel feature extraction (ACFE) module is designed for spectral feature enhancement, which demonstrate superior performance compared to state-of-the-art methods on six datasets. To conclude, the effectiveness of RCRS strategy with pseudo prior channel knowledge on seven dual-input and eight single-input networks, as well as superiority of the proposed channel-enhanced network indicate the importance of spectral properties for HSI-SR tasks.

Fault feature enhancement in seismic data based on steerable pyramid tensor voting

Although faulting is usually discontinuous for some geologic environments, many faults or parts of the faults are weak discontinuities or even invisible in seismic images due to seismic resolution and noises. Traditional fault detection methods often lead to blurred, low-integrity fault attributes. These blurred attributes may hinder insights into geologic interpretation. We develop a multiscale and multidirectional fault enhancement method called steerable pyramid tensor voting (SPTV) to overcome this difficulty. The proposed method consists of two cascaded steps. The steerable pyramid step generates multiscale and multidirectional seismic attributes. These attributes at different scales are enhanced by optimal directional filtering and then reconstructed to improve the fault resolution. The tensor voting step discovers the hidden fault features by voting from the adjacent faults. This step enhances the fault integrity and linearity and is able to extract the fault skeletons. We test our method using a benchmark model and confirm its effectiveness in identifying faults. We apply this method to identify the faults in a tight sand reservoir of the Sichuan Basin, China. The results show that the method can effectively enhance the fault and subfault features and enable the clear identification of faults compared with traditional methods. Furthermore, the results exhibit a reasonable congruency with the geologic and well-production data.

Signal processing for miniature mass spectrometer based on LSTM-EEMD feature digging

Miniature mass spectrometers exhibit immense application potential in on-site detection due to their small size and low cost. However, their detection accuracy is severely affected by factors such as sample pre-processing and environmental conditions. In this study, we propose a data processing method based on long short-term memory-ensemble empirical mode decomposition (LSTM-EEMD) to improve the quality of on-site detection data from miniature mass spectrometers. The EEMD method can clearly decompose the different physical feature components in the small-scale spectrometer signals, while the LSTM method can adaptively learn the internal feature relationships of the signals. Thus, by combining the two, the parameters for the EEMD signal reconstruction can be optimized in an adaptive manner, obtaining the optimized coefficients. Compared to the previous EEMD feature enhancement approach, the LSTM-EEMD method not only significantly improves the coefficient of determination (R2) and relative standard deviation (RSD) of the data, enhancing the linear range, but also achieves fully adaptive processing throughout the workflow, greatly boosting the efficiency. By leveraging a miniature mass spectrometer, data for N-acetyl-l-aspartic acid (NAA), 2-Hydroxyglutarate (2-HG), and γ-Aminobutyric acid (GABA) in actual blood samples have been obtained. The experimental results demonstrate that the LSTM-EEMD method can markedly enhance the accuracy and usability of the biological sample data in practical testing, providing new perspectives and possibilities for research and applications in the relevant domain.

Deep spatial and discriminative feature enhancement network for stereo matching

Deep spatial and discriminative feature enhancement network for stereo matching

Relevance Pooling Guidance and Class-Balanced Feature Enhancement for Fine-Grained Oriented Object Detection in Remote Sensing Images

Fine-grained object detection in remote sensing images is highly challenging due to class imbalance and high inter-class indistinguishability. The strategies employed by most existing methods to resolve these two challenges are relatively rudimentary, resulting in suboptimal model performance. To address these issues, we propose a fine-grained object-oriented detection method based on relevance pooling guidance and class balance feature enhancement. Firstly, we propose a global attention mechanism that dynamically retains spatial features pertinent to objects through relevance pooling during down-sampling, thereby enabling the model to acquire more discriminative features. Next, a class balance correction module is proposed to alleviate the class imbalance problem. This module employs feature translation and a learnable reinforcement coefficient to highlight the boundaries of tail class features while maintaining their distinctiveness. Furthermore, we present an enhanced contrastive learning strategy. By dynamically adjusting the contribution of inter-class samples and intra-class similarity measures, this strategy not only constrains inter-class feature distances but also facilitates tighter intra-class clustering, making it more suitable for imbalanced datasets. Evaluation on the FAIR1M and MAR20 datasets demonstrates that our method is superior compared to other methods in object detection and achieves 46.44 and 85.05% mean average precision, respectively.

Generating generalized zero-shot learning based on dual-path feature enhancement

Generating generalized zero-shot learning based on dual-path feature enhancement

A photovoltaic cell defect detection model capable of topological knowledge extraction

As the global transition towards clean energy accelerates, the demand for the widespread adoption of solar energy continues to rise. However, traditional object detection models prove inadequate for handling photovoltaic cell electroluminescence (EL) images, which are characterized by high levels of noise. To address this challenge, we developed an advanced defect detection model specifically designed for photovoltaic cells, which integrates topological knowledge extraction. Our approach begins with the introduction of a multi-scale dynamic context-based feature extraction method, capable of generating static context by thoroughly capturing the local texture and structural information of multi-scale defects. This static context is then combined with dynamic context to produce fine-grained local features. Subsequently, we developed a centralized feature pyramid structure, enhanced by spatial semantics, which models the explicit visual center. This structure effectively elucidates the relationship between local and global features in defect images, thereby improving the representation of defect characteristics. Finally, we implemented a feature enhancement strategy grounded in spatial semantic knowledge extraction. This strategy uncovers potential correlations among defect targets by constructing a spatial semantic topology of features, mapping these features to a higher-order representation, and ultimately delivering precise defect detection results.

Evaluation of microstructure and mechanical properties of Al1050/Al2O3/Gr composite processed by forming operation ECAP

Abstract Enhancement of microstructure and mechanical features of the hybrid aluminum matrix composite (HAMC) prepared by the stir-casting process (SCP) is quite significant for failure prevention during the service. In this work, the circular rod of hybrid aluminum matrix composite reinforced with particles of alumina (Al2O3) 50 μm, and graphite (Gr) 40–100 µm fabricated by SCP was adopted. This Al1050/Al2O3/Gr composite was subjected to severe plastic deformation using equal channel angular pressing (ECAP) at room temperature to show the impact of this process on the microstructure and mechanical features of the fabricated composite. Grain refinement, strength, and hardness were evaluated at different forming passes (1P, 3P, and 5P) with two channel angles (die angles) of 120° and 135°. The results revealed that the number of ECAP cycles has a significant effect on the refinement of the grain size. The fifth pass of ECAP (5P ECAP) with a die angle of 120° gave more refinement of the grains in the range of 36–75 nm compared to other passes. On the other hand, the strength and hardness relatively augment until 5P ECAP with increasing the cycle number at two die angles of 120° and 135°. The ultrafine grain can reduce the voids of the aluminum matrix then the hardness is enhanced.

Hybrid multiple instance learning network for weakly supervised medical image classification and localization

Weakly supervised medical image analysis is of great significance for computer-aided diagnosis due to the difficulty in obtaining accurately labeled medical data. In this paper, we proposed a new Multi-instance Learning (MIL) framework called HybridMIL integrating CNN Convolutional Neural Networks (CNN) and Broad Learning Systems (BLS). Our HybridMIL can overcome several challenging issues over existing MIL methods based on either CNN or BLS alone: (i) Multiple levels (i.e., different resolutions) of feature information can be simultaneously extracted through a newly proposed instance-level feature enhancement (IFE) module; (ii) Global-level semantic information contained in the deep layers can be better represented under the global-level semantic enhancement (GSE) module; (iii) Hybrid feature fusion (HFF) module is newly designed to effectively fuse and align the multi-level outputs of IFE and global-level semantic information of GSE for subsequent classification and localization tasks. The proposed HybridMIL is evaluated on various public medical and MIL benchmark datasets. The results indicate that HybridMIL surpasses other recent MIL models in terms of classification and localization performance by up to 8.5% and 9.0%, respectively. Lastly, we demonstrate the highly competitive performance of HybridMIL in general MIL problems, going beyond weakly supervised medical image analysis.

Principal component fusion based unexposed biological feature enhancement of fundus images

Principal component fusion based unexposed biological feature enhancement of fundus images

Lycopene detection in cherry tomatoes with feature enhancement and data fusion

Lycopene, a biologically active phytochemical with health benefits, is a key quality indicator for cherry tomatoes. While ultraviolet/visible/near-infrared (UV/Vis/NIR) spectroscopy holds promise for large-scale online lycopene detection, capturing its characteristic signals is challenging due to the low lycopene concentration in cherry tomatoes. This study improved the prediction accuracy of lycopene by supplementing spectral data with image information through spectral feature enhancement and spectra-image fusion. The feasibility of using UV/Vis/NIR spectra and image features to predict lycopene content was validated. By enhancing spectral bands corresponding to colors correlated with lycopene, the performance of the spectral model was improved. Additionally, direct spectra-image fusion further enhanced the prediction accuracy, achieving RP2, RMSEP, and RPD as 0.95, 8.96 mg/kg, and 4.25, respectively. Overall, this research offers valuable insights into supplementing spectral data with image information to improve the accuracy of non-destructive lycopene detection, providing practical implications for online fruit quality prediction.

RTIA-Mono: Real-time lightweight self-supervised monocular depth estimation with global-local information aggregation

Self-supervised monocular depth estimation has attracted significant attention in computer vision, especially for applications like autonomous driving and robotics. Recently, CNNs and Transformers have achieved tremendous success in this task. However, existing research primarily focuses on improving estimation accuracy, increasing model complexity poses challenges for deployment on edge computing devices. Shallow CNNs aid lightweight network construction but suffer limited receptive fields, hindering fusion of local geometric features and global semantic information. To address these issues, we propose an efficient real-time lightweight self-supervised architecture, RTIA-Mono, for monocular depth estimation. Firstly, we design a cross-stage feature fusion structure promoting feature aggregation and fusion across stages. Secondly, in each stage, we propose a Global Local Information Aggregation (GLIA) module integrating advantages of CNNs and Transformers to aggregate local and global features. Additionally, we introduce a Directional Feature Enhancement (DFE) module supplementing spatial structure information to mitigate spatial information loss from downsampling. Through sophisticated design, the proposed approach outperforms state-of-the-art methods on KITTI benchmark with the least parameters, and achieves a good balance between accuracy, complexity and inference speed. Furthermore, RTIA-Mono demonstrates excellent generalization on other datasets.

Multimodal Recommender Systems: A Survey

The recommender system (RS) has been an integral toolkit of online services. They are equipped with various deep learning techniques to model user preference based on identifier and attribute information. With the emergence of multimedia services, such as short videos, news and etc. , understanding these contents while recommending becomes critical. Besides, multimodal features are also helpful in alleviating the problem of data sparsity in RS. Thus, M ultimodal R ecommender S ystem (MRS) has attracted much attention from both academia and industry recently. In this paper, we will give a comprehensive survey of the MRS models, mainly from technical views. First, we conclude the general procedures and major challenges for MRS. Then, we introduce the existing MRS models according to four categories, i.e., Modality Encoder , Feature Interaction , Feature Enhancement and Model Optimization . Besides, to make it convenient for those who want to research this field, we also summarize the dataset and code resources. Finally, we discuss some promising future directions of MRS and conclude this paper. To access more details of the surveyed papers, such as implementation code, we open source a repository.

Analysis of the current status and hot topics in spinal schwannoma imaging research based on bibliometrics.

This study aims to explore the current hot topics and future research trends in spinal schwannoma imaging research, providing a reference for related studies and promoting the development of spinal schwannoma imaging. We conducted a literature search in the Web of Science database using the search terms (((TS = (Spinal schwannoma)) AND TS = (Imaging)) OR TS = (Spinal schwannoma)) AND TS = (image) to retrieve relevant articles. The collected data, including authors, keywords, journals, countries, institutions, and references, were subjected to visual analysis using the visualization software CiteSpace 6.4.2R and VOSviewer 1.6.19. A total of 310 relevant articles were identified. After further screening based on time limits, inclusion, and exclusion criteria, 179 articles were included in the study, consisting of 132 original articles and 42 reviews. These articles were authored by 1,034 authors from 35 countries and 324 institutions and were published in 82 different journals. The included articles cited a total of 6,583 references from 1,314 journals. Although the field of spinal schwannoma imaging research is not a popular research area in the medical community, there has been an increasing international interest in this field in recent years. While China ranks high in terms of the number of published articles, there is still a gap in terms of the quality and research level compared to developed countries in Europe and America. MRI, as the gold standard for diagnosing spinal schwannomas, is expected to be a research hotspot in terms of feature analysis, enhancement characteristics, and quantitative analysis. It is also hoped that China can increase its investment in research and contribute to the field by publishing high-quality articles in the future.

Hierarchical Progressive Image Forgery Detection and Localization Method Based on UNet

The rapid development of generative technologies has made the production of forged products easier, and AI-generated forged images are increasingly difficult to accurately detect, posing serious privacy risks and cognitive obstacles to individuals and society. Therefore, constructing an effective method that can accurately detect and locate forged regions has become an important task. This paper proposes a hierarchical and progressive forged image detection and localization method called HPUNet. This method assigns more reasonable hierarchical multi-level labels to the dataset as supervisory information at different levels, following cognitive laws. Secondly, multiple types of features are extracted from AI-generated images for detection and localization, and the detection and localization results are combined to enhance the task-relevant features. Subsequently, HPUNet expands the obtained image features into four different resolutions and performs detection and localization at different levels in a coarse-to-fine cognitive order. To address the limited feature field of view caused by inconsistent forgery sizes, we employ three sets of densely cross-connected hierarchical networks for sufficient interaction between feature images at different resolutions. Finally, a UNet network with a soft-threshold-constrained feature enhancement module is used to achieve detection and localization at different scales, and the reliance on a progressive mechanism establishes relationships between different branches. We use ACC and F1 as evaluation metrics, and extensive experiments on our method and the baseline methods demonstrate the effectiveness of our approach.

SGNN-T: Space graph neural network coupled transformer for molecular property prediction

Molecular properties play a crucial role in material discovery, protein interaction and drug development. The appearance of Graph Neural Network (GNN) significantly improved the performance of molecular property prediction. However, nodes in GNN only update the features of neighbor nodes, resulting in insufficient ability to encode global feature information. The self- attention mechanism in transformer can encode the global information except for local information of molecules, while its spatial information is insufficient. Since molecules are three-dimensional spatial structures, spatial geometry information is an important attribute for molecules properties. To consider these factors, a network model of Space Graph Neural Network coupled Transformer (SGNN-T) is proposed in this paper which can combine global and local molecule information with three-dimensional spatial structures for molecular properties prediction. In this model, Graph neural network Geometric Feature Fusion Module (GGFF) and Transformer Spatial Geometric Feature Enhancement Module (TSGFE) are included to enhance the spatial geometry learning ability of the network. The GGFF module constructs a parallel graph neural network by thinking over atoms, bonds and bond angles at the same time which effectively complements the spatial information of the network by leading into bond angles than normal GNN. The TSGFE module introduces the coordinates and centrality degree features coupled with the features by GGFF into transformer to further enhance the geometric expression ability of the module. Through these two parts, SGNN-T model can encode local and global information of molecules at the same time. Property prediction experiments are executed on the QM9, OMDB and MEGNet dataset. The results of MAE show the proposed model has the best performance than the popular models.

A Multi-Scale Liver Tumor Segmentation Method Based on Residual and Hybrid Attention Enhanced Network with Contextual Integration.

Liver cancer is one of the malignancies with high mortality rates worldwide, and its timely detection and accurate diagnosis are crucial for improving patient prognosis. To address the limitations of traditional image segmentation techniques and the U-Net network in capturing fine image features, this study proposes an improved model based on the U-Net architecture, named RHEU-Net. By replacing traditional convolution modules in the encoder and decoder with improved residual modules, the network's feature extraction capabilities and gradient stability are enhanced. A Hybrid Gated Attention (HGA) module is integrated before the skip connections, enabling the parallel processing of channel and spatial attentions, optimizing the feature fusion strategy, and effectively replenishing image details. A Multi-Scale Feature Enhancement (MSFE) layer is introduced at the bottleneck, utilizing multi-scale feature extraction technology to further enhance the expression of receptive fields and contextual information, improving the overall feature representation effect. Testing on the LiTS2017 dataset demonstrated that RHEU-Net achieved Dice scores of 95.72% for liver segmentation and 70.19% for tumor segmentation. These results validate the effectiveness of RHEU-Net and underscore its potential for clinical application.

THDet: A Lightweight and Efficient Traffic Helmet Object Detector based on YOLOv8

Traffic helmet object detection is playing an increasing important role in the smart traffic fields. However, object size variation and small-shaped helmet detection has still been a challenging problem by reason of their poor visual appearance in the image. In this work, we present an efficient traffic helmet detector through feature enhancement and lightweight design based on YOLOv8n called THDet. Specifically, we employ the coordinate attention into C2f blocks combined with softmax activate function to achieve feature channel aggregation and strong non-linear expression of the backbone for further effective feature extraction; Next, Focal_CIoU loss function embedded with Focal Loss method is utilized for the more precise measure of various objects bounding box regression and balance of positive and negative examples during training; Then, a new lightweight detection head style is designed only with two proper position heads (P3 & P4) to perform final classification and localization, through this scheme saving the 33.7% parameters than baseline method. Finally, Attention Refined Features Module (ARFM) is built to calibrate the multi-scale fused features by introducing 3-D weights generated from SimAttention to boost the final detection accuracy. Extensive experiments have demonstrated that our proposed method realizes noticeable performance in terms of detection accuracy and inference speed compared with baseline YOLOv8n and many end-to-end detectors of similar model size. Concretely, THDet achieves 0.447 at the overall evaluation metric of mAP0.5−0.95, accomplishing 3.2% detection accuracy improvement than YOLOv8n. Besides, THDet only holds 2.2M parameters with 295 FPS inference speed, reducing 33.4% parameters compared with YOLOv8n. The experimental results validate the effectiveness of our proposed method, showcasing that THDet outperforms the mainstream real-time detection algorithms in the terms of accuracy, inference speed and lightweight model design for traffic helmet object detection.