Search

Background: Automatic segmentation of vertebrae in spinal x‐ray images is crucial for clinical diagnosis, case analysis, and surgical planning of spinal lesions. Purpose: However, due to the inherent characteristics of x‐ray images, including low contrast, high noise, and uneven grey scale, it remains a critical and challenging problem in computer‐aided spine image analysis and disease diagnosis applications. Methods: In this paper, a Multiscale Feature Enhancement Network (MFENet), is proposed for segmenting whole spinal x‐ray images, to aid doctors in diagnosing spinal‐related diseases. To enhance feature extraction, the network incorporates a Dual‐branch Feature Extraction Module (DFEM) and a Semantic Aggregation Module (SAM). The DFEM has a parallel dual‐branch structure. The upper branch utilizes multiscale convolutional kernels to extract features from images. Employing convolutional kernels of different sizes helps capture details and structural information at different scales. The lower branch incorporates attention mechanisms to further optimize feature representation. By modeling the feature maps spatially and across channels, the network becomes more focused on key feature regions and suppresses task‐irrelevant information. The SAM leverages contextual semantic information to compensate for details lost during pooling and convolution operations. It integrates high‐level feature information from different scales to reduce segmentation result discontinuity. In addition, a hybrid loss function is employed to enhance the network's feature extraction capability. Results: In this study, we conducted a multitude of experiments utilizing dataset provided by the Spine Surgery Department of Henan Provincial People's Hospital. The experimental results indicate that our proposed MFENet demonstrates superior segmentation performance in spinal segmentation on x‐ray images compared to other advanced methods, achieving 92.61 ± 0.431 for MIoU, 92.42 ± 0.329 for DSC, and 99.51 ± 0.037 for Global_accuracy. Conclusions: Our model is able to more effectively learn and extract global contextual semantic information, significantly improving spinal segmentation performance, further aiding doctors in analyzing patient conditions. [ABSTRACT FROM AUTHOR]

Detecting small objects in images poses significant challenges due to their limited pixel representation and the difficulty in extracting sufficient features, often leading to missed or false detections. To address these challenges and enhance detection accuracy, this paper presents an improved small object detection algorithm, CRL-YOLOv5. The proposed approach integrates the Convolutional Block Attention Module (CBAM) attention mechanism into the C3 module of the backbone network, which enhances the localization accuracy of small objects. Additionally, the Receptive Field Block (RFB) module is introduced to expand the model's receptive field, thereby fully leveraging contextual information. Furthermore, the network architecture is restructured to include an additional detection layer specifically for small objects, allowing for deeper feature extraction from shallow layers. When tested on the VisDrone2019 small object dataset, CRL-YOLOv5 achieved an mAP50 of 39.2%, representing a 5.4% improvement over the original YOLOv5, effectively boosting the detection precision for small objects in images. [ABSTRACT FROM AUTHOR]

Content-based image retrieval (CBIR) systems are vital for managing the large volumes of data produced by medical imaging technologies. They enable efficient retrieval of relevant medical images from extensive databases, supporting clinical diagnosis, treatment planning, and medical research.This study aims to enhance CBIR systems’ effectiveness in medical image analysis by introducing the VisualSift Ensembling Integration with Attention Mechanisms (VEIAM). VEIAM seeks to improve diagnostic accuracy and retrieval efficiency by integrating robust feature extraction with dynamic attention mechanisms.VEIAM combines Scale-Invariant Feature Transform (SIFT) with selective attention mechanisms to emphasize crucial regions within medical images dynamically. Implemented in Python, the model integrates seamlessly into existing medical image analysis workflows, providing a robust and accessible tool for clinicians and researchers.The proposed VEIAM model demonstrated an impressive accuracy of 97.34% in classifying and retrieving medical images. This performance indicates VEIAM’s capability to discern subtle patterns and textures critical for accurate diagnostics.By merging SIFT-based feature extraction with attention processes, VEIAM offers a discriminatively powerful approach to medical image analysis. Its high accuracy and efficiency in retrieving relevant medical images make it a promising tool for enhancing diagnostic processes and supporting medical research in CBIR systems. [ABSTRACT FROM AUTHOR]

Detecting wildfires presents significant challenges due to the presence of various potential targets in fire imagery, such as smoke, vehicles, and people. To address these challenges, we propose a novel multi-label classification model based on BiFormer's feature extraction method, which constructs sparse region-indexing relations and performs feature extraction only in key regions, thereby facilitating more effective capture of flame characteristics. Additionally, we introduce a feature screening method based on a domain-adversarial neural network (DANN) to minimize misclassification by accurately determining feature domains. Furthermore, a feature discrimination method utilizing a Graph Convolutional Network (GCN) is proposed, enabling the model to capture label correlations more effectively and improve performance by constructing a label correlation matrix. This model enhances cross-domain generalization capability and improves recognition performance in fire scenarios. In the experimental phase, we developed a comprehensive dataset by integrating multiple fire-related public datasets, and conducted detailed comparison and ablation experiments. Results from the tenfold cross-validation demonstrate that the proposed model significantly improves recognition of multi-labeled images in fire scenarios. Compared with the baseline model, the mAP increased by 4.426%, CP by 4.14% and CF1 by 7.04%. [ABSTRACT FROM AUTHOR]

In this paper, we propose a representation of the visual information about Multiplayer Online Battle Arena (MOBA) video games using an adapted unsupervised deep learning architecture called Convolutional Accordion Auto-Encoder (Conv_A2E). Our study includes a presentation of current representations of MOBA video game information and why our proposal offers a novel and useful solution to this task. This approach aims to achieve dimensional reduction and refined feature extraction of the visual data. To enhance the model's performance, we tested several attention mechanisms for computer vision, evaluating algorithms from the channel attention and spatial attention families, and their combination. Through experimentation, we found that the best reconstruction of the visual information with the Conv_A2E was achieved when using a spatial attention mechanism, deformable convolution, as its mean squared error (MSE) during testing was the lowest, reaching a value of 0.003893, which means that its dimensional reduction is the most generalist and representative for this case study. This paper presents one of the first approaches to applying attention mechanisms to the case study of MOBA video games, representing a new horizon of possibilities for research. [ABSTRACT FROM AUTHOR]

To address issues such as misdetection and omission due to low light, image defocus, and worker occlusion in coal-rock image recognition, a new method called YOLOv8-Coal, based on YOLOv8, is introduced to enhance recognition accuracy and processing speed. The Deformable Convolution Network version 3 enhances object feature extraction by adjusting sampling positions with offsets and aligning them closely with the object's shape. The Polarized Self-Attention module in the feature fusion network emphasizes crucial features and suppresses unnecessary information to minimize irrelevant factors. Additionally, the lightweight C2fGhost module combines the strengths of GhostNet and the C2f module, further decreasing model parameters and computational load. The empirical findings indicate that YOLOv8-Coal has achieved substantial enhancements in all metrics on the coal rock image dataset. More precisely, the values for AP50, AP50:95, and AR50:95 were improved to 77.7%, 62.8%, and 75.0% respectively. In addition, optimal localization recall precision (oLRP) were decreased to 45.6%. In addition, the model parameters were decreased to 2.59M and the FLOPs were reduced to 6.9G. Finally, the size of the model weight file is a mere 5.2 MB. The enhanced algorithm's advantage is further demonstrated when compared to other commonly used algorithms. [ABSTRACT FROM AUTHOR]

Facial expression recognition (FER) plays a crucial role in affective computing, enhancing human-computer interaction by enabling machines to understand and respond to human emotions. Despite advancements in deep learning, current FER systems often struggle with challenges such as occlusions, head pose variations, and motion blur in natural environments. These challenges highlight the need for more robust FER solutions. To address these issues, we propose the Attention-Enhanced Multi-Layer Transformer (AEMT) model, which integrates a dual-branch Convolutional Neural Network (CNN), an Attentional Selective Fusion (ASF) module, and a Multi-Layer Transformer Encoder (MTE) with transfer learning. The dual-branch CNN captures detailed texture and color information by processing RGB and Local Binary Pattern (LBP) features separately. The ASF module selectively enhances relevant features by applying global and local attention mechanisms to the extracted features. The MTE captures long-range dependencies and models the complex relationships between features, collectively improving feature representation and classification accuracy. Our model was evaluated on the RAF-DB and AffectNet datasets. Experimental results demonstrate that the AEMT model achieved an accuracy of 81.45% on RAF-DB and 71.23% on AffectNet, significantly outperforming existing state-of-the-art methods. These results indicate that our model effectively addresses the challenges of FER in natural environments, providing a more robust and accurate solution. The AEMT model significantly advances the field of FER by improving the robustness and accuracy of emotion recognition in complex real-world scenarios. This work not only enhances the capabilities of affective computing systems but also opens new avenues for future research in improving model efficiency and expanding multimodal data integration. [ABSTRACT FROM AUTHOR]

The western black-crested gibbon (Nomascus concolor) is a rare and endangered primate that inhabits southern China and northern Vietnam, and has become a key conservation target due to its distinctive call and highly endangered status, making its identification and monitoring particularly urgent. Identifying calls of the western black-crested gibbon using passive acoustic monitoring data is a crucial method for studying and analyzing these gibbons; however, traditional call recognition models often overlook the temporal information in audio features and fail to adapt to channel-feature weights. To address these issues, we propose an innovative deep learning model, VBSNet, designed to recognize and classify a variety of biological calls, including those of endangered western black-crested gibbons and certain bird species. The model incorporates the image feature extraction capability of the VGG16 convolutional network, the sequence modeling capability of bi-directional LSTM, and the feature selection capability of the SE attention module, realizing the multimodal fusion of image, sequence and attention information. In the constructed dataset, the VBSNet model achieved the best performance in the evaluation metrics of accuracy, precision, recall, and F1-score, realizing an accuracy of 98.35%, demonstrating high accuracy and generalization ability. This study provides an effective deep learning method in the field of automated bioacoustic monitoring, which is of great theoretical and practical significance for supporting wildlife conservation and maintaining biodiversity. [ABSTRACT FROM AUTHOR]

Sericulture has a pivotal position in the agricultural production of China. The number of silkworms is one of the key indicators in the process of silkworm breeding, which is crucial to the feeding and healthy growth of silkworms. Accurate detection of the number of silkworms can ensure the quality of sericulture products. However, manual counting has not fully met the needs of large-scale production in recent years. There is a high demand to combine AI technology for silkworm counting. In this study, silkworm detection and counting were studied from the perspective of deep learning. Several challenges were solved in the detection of silkworms. Since the silkworms belonged to the small targets, it was difficult to detect the fewer features; When the number of silkworms was high, serious masking occurred to cause interference in the detection of densely distributed silkworms; It was difficult to ensure that the clear images of silkworms were captured from the actual production. The blur images were easy to deteriorate the accuracy of detection. Two tasks were covered as follows. 1) The image dataset of silkworms was constructed; 2) MAM-YOLO, a silkworm detection and counting model was proposed using multi-dimensional perception. The pictures of silkworms were first taken from the Guangxi Sericulture Technology Promotion Station. These pictures of silkworms were preliminarily processed. The polygon annotation was then selected, according to the body and distribution of silkworms. The pictures of silkworms were annotated with the open-source tool (Labelme). Finally, the dataset was expanded using image transformation and noise addition. The silkworm picture dataset was obtained for the subsequent training, verification, and testing. An improved YOLOv8 (MAM-YOLO) was proposed to detect the silkworm, in order to realize the accurate detection and counting. A silkworm detection and counting model was proposed using space-to-depth transformation and image deblurring. The fine-grained detection performance of blurred images was improved in the MAMYOLO model. The space-to-depth transformed convolution was utilized instead of stride convolution for the pooling layer in the feature extraction. The fine-grained features were extracted from the images, in order to further improve the detection of the low-resolution and small targets. In addition, the impact of blurred images on the detection performance was reduced by introducing a deblurring module into the model. RCS-OSA module was added to the YOLOv8 model. The network feature extraction was improved to integrate the information of different sizes in the receptive fields. The detection head of the original network was improved into the multi-dimensional sensing head. The three-dimensional features of size, space, and channel were aggregated to enhance the detection of silkworm targets. The loss function was replaced with the EIOU loss function. The regression accuracy of the detection frame was improved to reduce the object occlusion. Experimental results show that the MAM-YOLO more accurately detected the silkworm targets. The detection speed of mAP@0.5:0.95 reached 74.8% and 22.2 frames per second, indicating the better functions of silkworm detection and counting with high precision and high rate. The detection requirements were fully met in the actual work of the Guangxi sericulture station. The results show that the silkworm detection and counting system of the technical indicators met the application requirements. The finding can provide the detection services for the actual production work of the silkworm breeding station. In the end, a generalization experiment was conducted on the Global Wheat Challenge 2021 dataset. It was found that the MAM-YOLO was also achieved in the better detection performance of wheat ears, particularly on the small targets similar to dense occlusion.Sericulture has a pivotal position in China's agricultural production. The number of silkworms is a key indicator in the process of silkworm breeding, which is crucial to the feeding and healthy growth of silkworms, and accurate detection of the number of silkworms can ensure the quality of sericulture products. The traditional way of counting silkworms manually is inefficient, and it needs to be combined with technology to intelligently complete this kind of heavy work. In this thesis, from the perspective of deep learning, a silkworm detection and counting method is studied, and several problems in silkworm detection are solved: silkworms belong to small targets, which have fewer features so that they are not easy to detect; silkworms are densely distributed, and when the number of silkworms is high, serious masking phenomenon occurs which causes interference in the detection; it is difficult to ensure that the images of silkworms obtained in the actual production are clear, and blur images will make the detection affected. The main research content of this thesis is as follows: This chapter completed two main tasks: 1) Completed the production of silkworm image data set; 2) MAM-YOLO, a silkworm detection and counting model based on multi-dimensional perception, was proposed.In this paper, pictures of silkworms were taken by workers from Guangxi Sericulture Technology Promotion Station. First, these pictures of silkworms were preliminically processed. Then, according to the characteristics of the body and distribution of silkworms, the polygon annotation method was selected, and the pictures of silkworms were annotated with the open-source tool Labelme. Finally, the data set is expanded by means of image transformation and noise addition, and finally the silkworm picture data set is obtained for subsequent training, verification and testing.This chapter proposes an improved silkworm detection method based on YOLOv8, MAM-YOLO, which realizes accurate detection and counting of silkworm. Secondly, aiming at the problems of loss of fine-grained information and poor detection performance of blurred images in the MAM-YOLO model, a silkworm detection and counting model based on spaceto-depth transformation and image deblurring is proposed. The model utilizes space-to-depth transformed convolution instead of stride convolution and pooling layer in the feature extraction process, so that fine-grained features in the image can be extracted, which further improves the detection capability for low-resolution and small targets. In addition, the impact of blurred images on the detection performance is reduced by introducing a deblurring module into the model. By adding RCSOSA module to YOLOv8 model, the capability of network feature extraction is improved, and the information of different size receptive fields is integrated. The detection head of the original network is improved into multi-dimensional sensing head, and the three-dimensional characteristics of size, space and channel are aggregated to enhance the detection ability of silkworm targets. By replacing the loss function with EIOU loss function, the regression accuracy of detection frame is improved and the problem caused by object occlusion is reduced. Experiments show that MAM-YOLO can more accurately detect silkworm targets, and the detection speed of mAP@0.5:0.95 reaches 74.8% and 22.2 frames per second, which realizes the functions of silkworm detection and counting with high precision and high rate, and can meet the detection requirements in the actual work of Guangxi sericulture station. results show that the silkworm detection and counting system of the technical indicators have met the application requirements, and can provide detection services for the actual production work of silkworm breeding station.In the end, a generalization experiment was conducted on the Global Wheat Challenge 2021 data set, and it was found that MAM-YOLO also showed a high improvement in the detection performance of wheat ears, proving that the proposed method can generalize to other small targets similar to dense occlusion. [ABSTRACT FROM AUTHOR]

In defect detection on metal surfaces, there are many small defects with subtle features that are difficult to distinguish from the background environment using mainstream object detection methods. To alleviate this issue, this study proposes an improved CenterNet model for enhancing the features of small defects on metal surfaces, namely MSDD. In this work, we utilize an attention mechanism to reconstruct the basic feature extraction module in the metal defect feature extraction network, aiming to enhance the focus on features related to small defects. Additionally, we redesign an efficient deconvolution module to extract multi‐scale defect feature information, capturing details of small defects at different scales. Finally, leveraging the idea of reparameterization to enhance feature representation capabilities, we optimize the output of the detection head, thereby strengthening the model's ability to capture features of small objects on the metal surface. On the NEU‐DET dataset, compared to the CenterNet baseline, the improved feature enhancement network shows a 4.9% improvement. Furthermore, with an accuracy of 80.2% compared to mainstream object detection methods, it outperforms other state‐of‐the‐art (SOTA) level object detection methods, significantly enhancing the detection accuracy and performance of the network model. [ABSTRACT FROM AUTHOR]

Video segmentation and object tracking are critical tasks in computer vision with applications spanning surveillance, autonomous driving, and interactive media. Traditional methods often struggle with the dynamic nature of video data, where object occlusions, variations in illumination, and complex motion patterns present significant challenges. Existing segmentation and tracking systems frequently suffer from inaccuracies in handling real-time video sequences, particularly in distinguishing and tracking multiple overlapping objects. The limitations of current models in addressing these issues necessitate the development of more advanced techniques that can effectively manage dynamic scenes and improve tracking accuracy. To address these challenges, we propose an advanced machine learning technique, AI-Enhanced TrackSegNet, which integrates deep learning with novel attention mechanisms for improved video segmentation and object tracking. Our method utilizes a combination of Convolutional Neural Networks (CNNs) for feature extraction and Long Short-Term Memory (LSTM) networks for temporal sequence modeling. We introduce an attention-based mechanism to dynamically focus on relevant features, enhancing the model's ability to handle occlusions and varying object appearances. The model was trained on a diverse dataset of video sequences, incorporating both synthetic and real-world footage. The AI-Enhanced TrackSegNet demonstrated significant improvements in performance compared to existing techniques. Our method achieved an average Intersection over Union (IoU) score of 86.7% for segmentation and a tracking precision rate of 91.3% on the MOT17 benchmark dataset. These results represent a 10.2% improvement in IoU and a 7.5% increase in tracking precision compared to state-of-the-art methods. The model also exhibited enhanced robustness in complex scenes, handling occlusions and motion variations with greater accuracy. [ABSTRACT FROM AUTHOR]

Lemon, as an important cash crop with rich nutritional value, holds significant cultivation importance and market demand worldwide. However, lemon diseases seriously impact the quality and yield of lemons, necessitating their early detection for effective control. This paper addresses this need by collecting a dataset of lemon diseases, consisting of 726 images captured under varying light levels, growth stages, shooting distances and disease conditions. Through cropping high-resolution images, the dataset is expanded to 2022 images, comprising 4441 healthy lemons and 718 diseased lemons, with approximately 1–6 targets per image. Then, we propose a novel model lemon surface disease YOLO (LSD-YOLO), which integrates Switchable Atrous Convolution (SAConv) and Convolutional Block Attention Module (CBAM), along with the design of C2f-SAC and the addition of a small-target detection layer to enhance the extraction of key features and the fusion of features at different scales. The experimental results demonstrate that the proposed LSD-YOLO achieves an accuracy of 90.62% on the collected datasets, with mAP@50–95 reaching 80.84%. Compared with the original YOLOv8n model, both mAP@50 and mAP@50–95 metrics are enhanced. Therefore, the LSD-YOLO model proposed in this study provides a more accurate recognition of healthy and diseased lemons, contributing effectively to solving the lemon disease detection problem. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Chongqing University of Technology (Natural Science) is the property of Chongqing University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Aiming at the problems of target missed detection and misdetection caused by diverse target scales, many similar targets and target aggregation in UAV aerial image target detection, DA-YOLO, an improved UAV aerial image target detection algorithm for YOLOv5, is proposed. Firstly, a multi-scale dynamic feature-weighted fusion network composed of feature map attention generator and dynamic weight learning module is proposed, the feature map attention generator is integrated to process the more important features of different scale targets, and the weight learning module adaptively adjusts the learning of target features of different scales, which can enhance the recognition of the network under the diversity of target scales and reduce target missed detection. Secondly, a parallel selective attention mechanism (PSAM) is designed to be added to the feature extraction network, which strengthens the expression of features to obtain a better feature map by dynamically fusing spatial information and channel information, and improves the network's ability to distinguish similar targets to reduce false detection. Finally, Soft-NMS is used instead of the non-maximum suppression (NMS) adopted in YOLOv5 to improve the missed detection and false detection in the target aggregation scenario. Experimental results show that the detection accuracy of the improved algorithm on the VisDrone dataset reaches 37.79%, which is 5.59 percentage points higher than that of the YOLOv5s algorithm, and the improved algorithm can be better applied to the target detection of UAV aerial images. [ABSTRACT FROM AUTHOR]

The objective of multimodal sentiment analysis is to extract and integrate feature information from text, image, and audio data accurately, in order to identify the emotional state of the speaker. While multimodal fusion schemes have made some progress in this research field, previous studies still lack adequate approaches for handling inter-modal information consistency and the fusion of different categorical features within a single modality. This study aims to effectively extract sentiment coherence information among video, audio, and text and consequently proposes a multimodal sentiment analysis method named joint chain interactive attention (VAE-JCIA, Video Audio Essay–Joint Chain Interactive Attention). In this approach, a 3D CNN is employed for extracting facial features from video, a Conformer is employed for extracting audio features, and a Funnel-Transformer is employed for extracting text features. Furthermore, the joint attention mechanism is utilized to identify key regions where sentiment information remains consistent across video, audio, and text. This process acquires reinforcing features that encapsulate information regarding consistency among the other two modalities. Inter-modal feature interactions are addressed through chained interactive attention, and multimodal feature fusion is employed to efficiently perform emotion classification. The method is experimentally validated on the CMU-MOSEI dataset and the IEMOCAP dataset. The experimental results demonstrate that the proposed method significantly enhances the performance of the multimodal sentiment analysis model. [ABSTRACT FROM AUTHOR]

In pursuit of cost-effective manufacturing, enterprises are increasingly adopting the practice of utilizing recycled semiconductor chips. To ensure consistent chip orientation during packaging, a circular marker on the front side is employed for pin alignment following successful functional testing. However, recycled chips often exhibit substantial surface wear, and the identification of the relatively small marker proves challenging. Moreover, the complexity of generic target detection algorithms hampers seamless deployment. Addressing these issues, this paper introduces a lightweight YOLOv8s-based network tailored for detecting markings on recycled chips, termed Van-YOLOv8. Initially, to alleviate the influence of diminutive, low-resolution markings on the precision of deep learning models, we utilize an upscaling approach for enhanced resolution. This technique relies on the Super-Resolution Generative Adversarial Network with Extended Training (SRGANext) network, facilitating the reconstruction of high-fidelity images that align with input specifications. Subsequently, we replace the original YOLOv8smodel's backbone feature extraction network with the lightweight VanillaNetwork (VanillaNet), simplifying the branch structure to reduce network parameters. Finally, a Hybrid Attention Mechanism (HAM) is implemented to capture essential details from input images, improving feature representation while concurrently expediting model inference speed. Experimental results demonstrate that the Van-YOLOv8 network outperforms the original YOLOv8s on a recycled chip dataset in various aspects. Significantly, it demonstrates superiority in parameter count, computational intricacy, precision in identifying targets, and speed when compared to certain prevalent algorithms in the current landscape. The proposed approach proves promising for real-time detection of recycled chips in practical factory settings. [ABSTRACT FROM AUTHOR]

Medium- to high-resolution imagery is indispensable for various applications. Combining images from Landsat 8 and Sentinel-2 can improve the accuracy of observing dynamic changes on the Earth's surface. Many researchers use Sentinel-2 10 m resolution data in conjunction with Landsat 8 30 m resolution data to generate 10 m resolution data series. However, current fusion techniques have some algorithmic weaknesses, such as simple processing of coarse or fine images, which fail to extract image features to the fullest extent, especially in rapidly changing land cover areas. Facing the aforementioned limitations, we proposed a multiscale and attention mechanism-based residual spatiotemporal fusion network (MARSTFN) that utilizes Sentinel-2 10 m resolution data and Landsat 8 15 m resolution data as auxiliary data to upgrade Landsat 8 30 m resolution data to 10 m resolution. In this network, we utilized multiscale and attention mechanisms to extract features from coarse and fine images separately. Subsequently, the features outputted from all input branches are combined and further feature information is extracted through residual networks and skip connections. Finally, the features obtained from the residual network are merged with the feature information of the coarsely processed images from the multiscale mechanism to generate accurate prediction images. To assess the efficacy of our model, we compared it with existing models on two datasets. Results demonstrated that our fusion model outperformed baseline methods across various evaluation indicators, highlighting its ability to integrate Sentinel-2 and Landsat 8 data to produce 10 m resolution data. [ABSTRACT FROM AUTHOR]

In spine imaging, efficient automatic segmentation is crucial for clinical decision‐making, yet current models increase accuracy at the expense of elevated parameter counts and computational complexity, complicating integration with contemporary medical devices. Addressing identified challenges, this research introduces LE‐NeXt, a spine segmentation framework utilizing multi‐dimensional spatial attention and multi‐scale feature extraction, optimizing the architecture via convolution and MLP. It integrates lightweight convolutions and attention mechanisms within an encoder‐decoder model, enhancing stage‐specific feature extraction while ensuring efficiency. Experimental analyses on VerSe and SpineWeb datasets demonstrate that LE‐NeXt outperforms the lightweight U‐NeXt, enhancing IoU accuracy from 87.7 to 89.8 on VerSe, and exceeds the performance of established networks such as U‐Net and its variants. Significantly, on SpineWeb, LE‐NeXt not only surpasses Trans U‐Net in accuracy but also achieves a considerable reduction in both parameter count and computational complexity. These results emphasize LE‐NeXt's effectiveness in improving segmentation precision efficiently, optimally balancing computational efficiency and accuracy. [ABSTRACT FROM AUTHOR]

Due to the limitations of traditional retinal blood vessel segmentation algorithms in feature extraction, vessel breakage often occurs at the end. To address this issue, a retinal vessel segmentation algorithm based on a modified U-shaped network is proposed in this paper. This algorithm can extract multi-scale vascular features and perform segmentation in an end-to-end manner. First, in order to improve the low contrast of the original image, pre-processing methods are employed. Second, a multi-scale residual convolution module is employed to extract image features of different granularities, while residual learning improves feature utilization efficiency and reduces information loss. In addition, a selective kernel unit is incorporated into the skip connections to obtain multi-scale features with varying receptive field sizes achieved through soft attention. Subsequently, to further extract vascular features and improve processing speed, a residual attention module is constructed at the decoder stage. Finally, a weighted joint loss function is implemented to address the imbalance between positive and negative samples. The experimental results on the DRIVE, STARE, and CHASE_DB1 datasets demonstrate that MU-Net exhibits better sensitivity and a higher Matthew's correlation coefficient (0.8197, 0.8051; STARE: 0.8264, 0.7987; CHASE_DB1: 0.8313, 0.7960) compared to several state-of-the-art methods. [ABSTRACT FROM AUTHOR]

With the advancement of machine vision technology, pig face recognition has garnered significant attention as a key component in the establishment of precision breeding models. In order to explore non-contact individual pig recognition, this study proposes a lightweight pig face feature learning method based on attention mechanism and two-stage transfer learning. Using a combined approach of online and offline data augmentation, both the self-collected dataset from Shanxi Agricultural University's grazing station and public datasets underwent enhancements in terms of quantity and quality. YOLOv8 was employed for feature extraction and fusion of pig face images. The Coordinate Attention (CA) module was integrated into the YOLOv8 model to enhance the extraction of critical pig face features. Fine-tuning of the feature network was conducted to establish a pig face feature learning model based on two-stage transfer learning. The YOLOv8 model achieved a mean average precision (mAP) of 97.73% for pig face feature learning, surpassing lightweight models such as EfficientDet, SDD, YOLOv5, YOLOv7-tiny, and swin_transformer by 0.32, 1.23, 1.56, 0.43 and 0.14 percentage points, respectively. The YOLOv8-CA model's mAP reached 98.03%, a 0.3 percentage point improvement from before its addition. Furthermore, the mAP of the two-stage transfer learning-based pig face feature learning model was 95.73%, exceeding the backbone network and pre-trained weight models by 10.92 and 3.13 percentage points, respectively. The lightweight pig face feature learning method, based on attention mechanism and two-stage transfer learning, effectively captures unique pig features. This approach serves as a valuable reference for achieving non-contact individual pig recognition in precision breeding. [ABSTRACT FROM AUTHOR]

To cope with the challenges of autonomous driving in complex road environments, the need for collaborative multi-tasking has been proposed. This research direction explores new solutions at the application level and has become a hot topic of great interest. In the field of natural language processing and recommendation algorithms, the use of multi-task learning networks has been proven to reduce time, computing power, and storage usage in various task coupling cases. Due to the characteristics of the multi-task learning network, it has also been applied to visual road feature extraction in recent years. This article proposes a multi-task road feature extraction network that combines group convolution with transformer and squeeze excitation attention mechanisms. The network can simultaneously perform drivable area segmentation, lane line segmentation, and traffic object detection tasks. The experimental results of the BDD-100K dataset show that the proposed method performs well for different tasks and has a higher accuracy than similar algorithms. The proposed method provides new ideas and methods for the autonomous road perception of vehicles and the generation of highly accurate maps in visual-based autonomous driving processes. [ABSTRACT FROM AUTHOR]

This study investigates the use of few-shot learning algorithms to improve classification performance in situations where traditional deep learning methods fail due to a lack of training data. Specifically, we propose a few-shot learning approach using the Almost No Inner Loop (ANIL) algorithm and attention modules to classify tomato diseases in the Plant Village dataset. The attended features obtained from the five separate attention modules are classified using a Multi Layer Perceptron (MLP) classifier, and the soft voting method is used to weigh the classification scores from each classifier. The results demonstrate that our proposed approach achieves state-of-the-art accuracy rates of 97.05% and 97.66% for 10-shot and 20-shot classification, respectively. Our approach demonstrates the potential for incorporating attention mechanisms in feature extraction processes and suggests new avenues for research in few-shot learning methods. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Jilin University (Science Edition) / Jilin Daxue Xuebao (Lixue Ban) is the property of Zhongguo Xue shu qi Kan (Guang Pan Ban) Dian zi Za zhi She and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Simple Summary: Research on the target area segmentation of cattle can improve the precision management and breeding level of pastures. When the cattle target is divided in the scene, we can further analyze animal habits by combining the scene information (drinking area, feeding area, rest area, etc.), which is of great significance for guiding farming production and management. Therefore, in this study, an improved deep learning semantic segmentation algorithm was proposed to realize the segmentation of cattle regions, and its performance was also verified in an actual breeding environment. The cattle regions obtained in this study provide a data-driven and technical basis for further analyses of cattle habits and body conditions. Obtaining animal regions and the relative position relationship of animals in the scene is conducive to further studying animal habits, which is of great significance for smart animal farming. However, the complex breeding environment still makes detection difficult. To address the problems of poor target segmentation effects and the weak generalization ability of existing semantic segmentation models in complex scenes, a semantic segmentation model based on an improved DeepLabV3+ network (Imp-DeepLabV3+) was proposed. Firstly, the backbone network of the DeepLabV3+ model was replaced by MobileNetV2 to enhance the feature extraction capability of the model. Then, the layer-by-layer feature fusion method was adopted in the Decoder stage to integrate high-level semantic feature information with low-level high-resolution feature information at multi-scale to achieve more precise up-sampling operation. Finally, the SENet module was further introduced into the network to enhance information interaction after feature fusion and improve the segmentation precision of the model under complex datasets. The experimental results demonstrate that the Imp-DeepLabV3+ model achieved a high pixel accuracy (PA) of 99.4%, a mean pixel accuracy (MPA) of 98.1%, and a mean intersection over union (MIoU) of 96.8%. Compared to the original DeepLabV3+ model, the segmentation performance of the improved model significantly improved. Moreover, the overall segmentation performance of the Imp-DeepLabV3+ model surpassed that of other commonly used semantic segmentation models, such as Fully Convolutional Networks (FCNs), Lite Reduced Atrous Spatial Pyramid Pooling (LR-ASPP), and U-Net. Therefore, this study can be applied to the field of scene segmentation and is conducive to further analyzing individual information and promoting the development of intelligent animal farming. [ABSTRACT FROM AUTHOR]

With the wide application of visual sensors and development of digital image processing technology, image copy-move forgery detection (CMFD) has become more and more prevalent. Copy-move forgery is copying one or several areas of an image and pasting them into another part of the same image, and CMFD is an efficient means to expose this. There are improper uses of forged images in industry, the military, and daily life. In this paper, we present an efficient end-to-end deep learning approach for CMFD, using a span-partial structure and attention mechanism (SPA-Net). The SPA-Net extracts feature roughly using a pre-processing module and finely extracts deep feature maps using the span-partial structure and attention mechanism as a SPA-net feature extractor module. The span-partial structure is designed to reduce the redundant feature information, while the attention mechanism in the span-partial structure has the advantage of focusing on the tamper region and suppressing the original semantic information. To explore the correlation between high-dimension feature points, a deep feature matching module assists SPA-Net to locate the copy-move areas by computing the similarity of the feature map. A feature upsampling module is employed to upsample the features to their original size and produce a copy-move mask. Furthermore, the training strategy of SPA-Net without pretrained weights has a balance between copy-move and semantic features, and then the module can capture more features of copy-move forgery areas and reduce the confusion from semantic objects. In the experiment, we do not use pretrained weights or models from existing networks such as VGG16, which would bring the limitation of the network paying more attention to objects other than copy-move areas.To deal with this problem, we generated a SPANet-CMFD dataset by applying various processes to the benchmark images from SUN and COCO datasets, and we used existing copy-move forgery datasets, CMH, MICC-F220, MICC-F600, GRIP, Coverage, and parts of USCISI-CMFD, together with our generated SPANet-CMFD dataset, as the training set to train our model. In addition, the SPANet-CMFD dataset could play a big part in forgery detection, such as deepfakes. We employed the CASIA and CoMoFoD datasets as testing datasets to verify the performance of our proposed method. The Precision, Recall, and F1 are calculated to evaluate the CMFD results. Comparison results showed that our model achieved a satisfactory performance on both testing datasets and performed better than the existing methods. [ABSTRACT FROM AUTHOR]

Rice lodging not only brings trouble to harvesting but also reduces yield. Therefore, the effective identification of rice lodging is of great significance. In this paper, we have designed a bilinear interpolation upsampling feature fusion module (BIFF) to decompose the quadruple upsampling of the connected part of encoder and decoder into two double upsampling processes and insert the intermediate feature layer in the backbone network for feature fusion in this process. The global attention mechanism(GAM) attention module is added to the feature extraction network, allowing the network to effectively focus on the lodging regions, thus effectively improving the segmentation effect. The average accuracy of the improved network is 93.55%, mrecall is 93.65%, and mIoU is 88.10%, and the feasibility of the improvement is demonstrated by ablation experiments and comparison with other algorithms. In addition, the angle calculation method is designed by combining the detection algorithm, adding a detection head branch to the output results for reading the distance information from the depth camera, and combining the distance information with mechanical analysis to determine the relationship between the angle of the stalk and the vertical direction when the rice is upright, tilted and fallen. A comparison of the calculated angle with the actual measured angle gives a final average error of approximately 5.364%, indicating that the harvest boundary extraction algorithm in this paper is highly accurate and has value for application in real-time harvesting scenarios. [ABSTRACT FROM AUTHOR]

Road information is an important geographic information. Road information extracted from remote sensing images has been widely used in map, traffic, navigation and many other fields. However, the autonomous extraction of road information from high resolution remote sensing images has some problems such as incoherence, incompleteness and poor connectivity, therefore, a semantic segmentation model for roads in high resolution remote sensing images, called AGD-Linknet, is proposed, which integrates attention mechanisms, gated decoder block, and dilated convolution. This model mainly consists of three parts. Firstly, the stem block is used as the initial convolution layer of the model to reduce the information loss in the convolution stage; Secondly, the series-parallel combined dilated convolution and coordinate attention block into the center of the network, which enlarges the receptive field of the network and improves the feature extraction ability of spatial domain and channel domain information; Finally, in the decoder part, gated convolution is introduced to improve the extraction of road edge. Compared with U-Net, Linknet and D-Linknet on the DeepGlobe dataset, the proposed AGD-Linknet has improved the pixel accuracy, mean intersection over union and F1-Score index of road recognition by 1.41%-11.52%, 0.0077-0.1473, 0.0057-0.1292, and has certain effectiveness and feasibility in many scenarios in rural areas, urban, and suburbs. And can be apply to the tasks of road recognition and extraction in high-resolution remote sensing.

Industrial inspection is crucial for maintaining quality and safety in industrial processes. Deep learning models have recently demonstrated promising results in such tasks. This paper proposes YOLOX-Ray, an efficient new deep learning architecture tailored for industrial inspection. YOLOX-Ray is based on the You Only Look Once (YOLO) object detection algorithms and integrates the SimAM attention mechanism for improved feature extraction in the Feature Pyramid Network (FPN) and Path Aggregation Network (PAN). Moreover, it also employs the Alpha-IoU cost function for enhanced small-scale object detection. YOLOX-Ray's performance was assessed in three case studies: hotspot detection, infrastructure crack detection and corrosion detection. The architecture outperforms all other configurations, achieving m A P 50 values of 89%, 99.6% and 87.7%, respectively. For the most challenging metric, m A P 50 : 95 , the achieved values were 44.7%, 66.1% and 51.8%, respectively. A comparative analysis demonstrated the importance of combining the SimAM attention mechanism with Alpha-IoU loss function for optimal performance. In conclusion, YOLOX-Ray's ability to detect and to locate multi-scale objects in industrial environments presents new opportunities for effective, efficient and sustainable inspection processes across various industries, revolutionizing the field of industrial inspections. [ABSTRACT FROM AUTHOR]

Small objects detection is a challenging task in computer vision due to the limited semantic information that can be extracted and the susceptibility to background interference. In this paper, we propose a decoupled multi-scale small object detection algorithm named DMS-YOLOv5. The algorithm incorporates a receptive field module into the feature extraction network for better focus on low-resolution small objects. The coordinate attention mechanism, which combines spatial and channel attention information, is introduced to reduce interference from background information and enhance the network's attention to object information. A detection layer tailored to small-sized objects is added to compensate for the loss of small object information in multiple downsampling operations, greatly improving the detection capability of small objects. Next, The decoupled network is introduced into the detection head network for branch processing of classification and bounding box regression tasks. Finally, the bounding box loss function is improved to alleviate missed detection problems caused by the concentration of small objects and mutual occlusion between objects. The improved method achieved a mean average precision improvement of 12.1% on VisDrone2019-DET dataset compared to the original method. In comparison experiments with similar methods, our proposed method also demonstrated good performance, validating its effectiveness. [ABSTRACT FROM AUTHOR]

Predicting accurate normal maps of objects from two-dimensional images in regions of complex structure and spatial material variations is challenging using photometric stereo methods due to the influence of surface reflection properties caused by variations in object geometry and surface materials. To address this issue, we propose a photometric stereo network called a RMAFF-PSN that uses residual multiscale attentional feature fusion to handle the "difficult" regions of the object. Unlike previous approaches that only use stacked convolutional layers to extract deep features from the input image, our method integrates feature information from different resolution stages and scales of the image. This approach preserves more physical information, such as texture and geometry of the object in complex regions, through shallow-deep stage feature extraction, double branching enhancement, and attention optimization. To test the network structure under real-world conditions, we propose a new real dataset called Simple PS data, which contains multiple objects with varying structures and materials. Experimental results on a publicly available benchmark dataset demonstrate that our method outperforms most existing calibrated photometric stereo methods for the same number of input images, especially in the case of highly non-convex object structures. Our method also obtains good results under sparse lighting conditions. [ABSTRACT FROM AUTHOR]

A real-time tuna detection network on mobile devices is a common tool for accurate tuna catch statistics. However, most object detection models have multiple parameters, and normal mobile devices have difficulties in satisfying real-time detection. Based on YOLOv3, this paper proposes a Tuna-YOLO, which is a lightweight object detection network for mobile devices. Firstly, following a comparison of the performance of various lightweight backbone networks, the MobileNet v3 was used as a backbone structure to reduce the number of parameters and calculations. Secondly, the SENET module was replaced with a CBAM attention module to further improve the feature extraction ability of tuna. Then, the knowledge distillation was used to make the Tuna-YOLO detect more accurate. We created a small dataset by deframing electronic surveillance video of fishing boats and labeled the data. After data annotation on the dataset, the K-means algorithm was used to get nine better anchor boxes on the basis of label information, which was used to improve the detection precision. In addition, we compared the detection performance of the Tuna-YOLO and three versions of YOLO v5-6.1 s/m/l after image enhancement. The results show that the Tuna-YOLO reduces the parameters of YOLOv3 from 234.74 MB to 88.45 MB, increases detection precision from 93.33% to 95.83%, and increases the calculation speed from 10.12 fps to 15.23 fps. The performance of the Tuna-YOLO is better than three versions of YOLO v5-6.1 s/m/l. Tuna-YOLO provides a basis for subsequent deployment of algorithms to mobile devices and real-time catch statistics. [ABSTRACT FROM AUTHOR]

As the monitor probes are used more and more widely these days, the task of detecting abnormal behaviors in surveillance videos has gained widespread attention. The generalization ability and parameter overhead of the model affect how accurate the detection result is. To deal with the poor generalization ability and high parameter overhead of the model in existing anomaly detection methods, we propose a three-dimensional multi-branch convolutional fusion network, named "Branch-Fusion Net". The network is designed with a multi-branch structure not only to significantly reduce parameter overhead but also to improve the generalization ability by understanding the input feature map from different perspectives. To ignore useless features during the model training, we propose a simple yet effective Channel Spatial Attention Module (CSAM), which sequentially focuses attention on key channels and spatial feature regions to suppress useless features and enhance important features. We combine the Branch-Fusion Net and the CSAM as a local feature extraction network and use the Bi-Directional Gated Recurrent Unit (Bi-GRU) to extract global feature information. The experiments are validated on a self-built Crimes-mini dataset, and the accuracy of anomaly detection in surveillance videos reaches 93.55% on the test set. The result shows that the model proposed in the paper significantly improves the accuracy of anomaly detection in surveillance videos with low parameter overhead. [ABSTRACT FROM AUTHOR]

With the global aging problem becoming more and more serious, the initial screening for Alzheimer's disease (AD) will become increasingly important. We understand that facial expressions are related to the severity of dementia, but there is no face‐related data in the existing Alzheimer's dataset. This article attempts to establish a facial video‐based AD recognition dataset through a human‐computer interaction method. This interactive task was designed for AD in attention, execution, visual space ability, facial apraxia, and facial changes in task success and failure. Using this task as the collection method, the final dataset includes 102 faces video data, specific task scores, and emotional self‐evaluation. For baseline evaluation, the improved local binary pattern on three orthogonal planes and RF were employed respectively for feature extraction and classification with the 5‐fold cross‐validation method. The best performance was 76.00% for 3‐class classification. In addition, a frame attention network based on fine‐grained local region localization was proposed, which improved the accuracy of cognitive classification to 84.45%. Finally, the analysis was conducted for the association of expressions with cognition and emotion in the AD dataset. This study aims to solve the current lack of standards for AD in the field of facial recognition and contribute to future research and clinical applications. [ABSTRACT FROM AUTHOR]

The automatic and accurate analysis of medical images (e.g., segmentation,detection, classification) are prerequisites for modern disease diagnosis and prognosis. Computer-aided diagnosis (CAD) systems empower accurate and effective detection of various diseases and timely treatment decisions. The past decade witnessed a spur in deep learning (DL)-based CADs showing outstanding performance across many health care applications. Medical imaging is hindered by multiple sources of uncertainty ranging fromnteasurement (aleatoric) errors, physiological variability, and limited medical knowledge (epistemic errors). However, uncertainty quantification (UQ) in most existing DL methods is insufficiently investigated, particularly in medical image analysis. Therefore, to address this gap, in this article, we propose a simple yet novel hierarchical attentive multilevel feature fusion model with an uncertainty-aware module for medical image classification coined Hercules. This approach is tested on several real medical image classification challenges. The proposed Hercules model consists of two main feature fusion blocks, where the former concentrates on attention-based fusion with uncertainty quantification module and the latter uses the raw features. Hercules was evaluated across three medical imaging datasets, i.e., retinal OCT, lung CT, and chest X-ray. Hercules produced the best classification accuracy in retinal OCT (94.21%), lung CT (99.59%), and chest X-ray (96.50%) datasets, respectively, against other state-of-the-art medical image classification methods. [ABSTRACT FROM AUTHOR]

During the process of producing hot‐rolled strips in the metallurgical industry, various defects inevitably appear on its surface due to harsh environments and complex manufacturing, consequently bringing about quality problems and economic loss. However, the existing detection methods are difficult to meet the actual requirements of commercial production due to their problems, such as low efficiency and low accuracy. Herein, an improved You only look once X (YOLOX) model for detecting strip surface defects is proposed. Based on the existing YOLOX model, herein, the MobileViT block is introduced to enhance the capability of feature extraction of the backbone network output. The feature pyramid networks through efficient channel attention (ECA) module to strengthen important channel weights are improved, and finally, the original positioning loss function by efficient intersection over union (EIOU) to increase the locating accuracy is replaced. The experimental results show that the improved YOLOX model can obtain 80.67 mAP and 75.69 mAP detection effects on the Northeast University dataset and Xsteel surface defect dataset, respectively. Compared with the original YOLOX, the model increases by 3.95 mAP and 4.02 mAP, respectively. The data fully show that the improved YOLOX model proposed herein is more effective for strip surface defect detection. [ABSTRACT FROM AUTHOR]

Accurate and robust localization is a fundamental need for mobile agents. Visual–inertial odometry (VIO) algorithms exploit the information from the camera and inertial sensors to estimate position and translation. Recent deep-learning-based VIO models attract attention as they provide pose information in a data-driven way, without the need of designing hand-crafted algorithms. Existing learning-based VIO models rely on recurrent models to fuse multimodal data and process sensor signals, which are hard to train and not efficient enough. We propose a novel learning-based VIO framework with external memory attention that effectively and efficiently combines visual and inertial features for state estimation. Our proposed model is able to estimate pose accurately and robustly, even in challenging scenarios, for example, on overcast days and water-filled ground, which are difficult for traditional VIO algorithms to extract visual features. Experiments validate that it outperforms both traditional and learning-based VIO baselines in different scenes. [ABSTRACT FROM AUTHOR]

The spatial distribution of remote-sensing scene images is highly complex in character, so how to extract local key semantic information and discriminative features is the key to making it possible to classify accurately. However, most of the existing convolutional neural network (CNN) models tend to have global feature representations and lose the shallow features. In addition, when the network is too deep, gradient disappearance and overfitting tend to occur. To solve these problems, a lightweight, multi-instance CNN model for remote sensing scene classification is proposed in this paper: MILRDA. In the instance extraction and classifier part, more discriminative features are extracted by the constructed residual dense attention block (RDAB) while retaining shallow features. Then, the extracted features are transformed into instance-level vectors and the local information associated with bag-level labels is highlighted by the proposed channel-attention-based multi-instance pooling, while suppressing the weights of useless objects or backgrounds. Finally, the network is constrained by the cross-entropy loss function to output the final prediction results. The experimental results on four public datasets show that our proposed method can achieve comparable results to other state-of-the-art methods. Moreover, the visualization of feature maps shows that MILRDA can find more effective features. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

With the wide application of visual sensors and development of digital image processing technology, image copy-move forgery detection (CMFD) has become more and more prevalent. Copy-move forgery is copying one or several areas of an image and pasting them into another part of the same image, and CMFD is an efficient means to expose this. There are improper uses of forged images in industry, the military, and daily life. In this paper, we present an efficient end-to-end deep learning approach for CMFD, using a span-partial structure and attention mechanism (SPA-Net). The SPA-Net extracts feature roughly using a pre-processing module and finely extracts deep feature maps using the span-partial structure and attention mechanism as a SPA-net feature extractor module. The span-partial structure is designed to reduce the redundant feature information, while the attention mechanism in the span-partial structure has the advantage of focusing on the tamper region and suppressing the original semantic information. To explore the correlation between high-dimension feature points, a deep feature matching module assists SPA-Net to locate the copy-move areas by computing the similarity of the feature map. A feature upsampling module is employed to upsample the features to their original size and produce a copy-move mask. Furthermore, the training strategy of SPA-Net without pretrained weights has a balance between copy-move and semantic features, and then the module can capture more features of copy-move forgery areas and reduce the confusion from semantic objects. In the experiment, we do not use pretrained weights or models from existing networks such as VGG16, which would bring the limitation of the network paying more attention to objects other than copy-move areas.To deal with this problem, we generated a SPANet-CMFD dataset by applying various processes to the benchmark images from SUN and COCO datasets, and we used existing copy-move forgery datasets, CMH, MICC-F220, MICC-F600, GRIP, Coverage, and parts of USCISI-CMFD, together with our generated SPANet-CMFD dataset, as the training set to train our model. In addition, the SPANet-CMFD dataset could play a big part in forgery detection, such as deepfakes. We employed the CASIA and CoMoFoD datasets as testing datasets to verify the performance of our proposed method. The Precision, Recall, and F1 are calculated to evaluate the CMFD results. Comparison results showed that our model achieved a satisfactory performance on both testing datasets and performed better than the existing methods.

When performing robotic automatic sorting and assembly operations of multi-category hardware, there are some problems with the existing convolutional neural network visual recognition algorithms, such as large computing power consumption, low recognition efficiency, and a high rate of missed detection and false detection. A novel efficient convolutional neural algorithm for multi-category aliasing hardware recognition is proposed in this paper. On the basis of SSD, the novel algorithm uses Resnet-50 instead of VGG16 as the backbone feature extraction network, and it integrates ECA-Net and Improved Spatial Attention Block (ISAB): two attention mechanisms to improve the ability of learning and extract target features. Then, we pass the weighted features to extra feature layers to build an improved SSD algorithm. At last, in order to compare the performance difference between the novel algorithm and the existing algorithms, three kinds of hardware with different sizes are chosen to constitute an aliasing scene that can simulate an industrial site, and some comparative experiments have been completed finally. The experimental results show that the novel algorithm has an mAP of 98.20% and FPS of 78, which are better than Faster R-CNN, YOLOv4, YOLOXs, EfficientDet-D1, and original SSD in terms of comprehensive performance. The novel algorithm proposed in this paper can improve the efficiency of robotic sorting and assembly of multi-category hardware. [ABSTRACT FROM AUTHOR]

Deep learning has substantially pushed forward remote sensing image change detection through extracting discriminative hierarchical features. However, as the increasingly high-resolution remote sensing images have abundant spatial details but limited spectral information, the use of conventional backbone networks would give rise to blurry boundaries between different semantics among hierarchical features. This explains why most false alarms in the final predictions distribute around change boundaries. To alleviate the problem, we pay attention to feature refinement and propose deep learning networks that deliver improved separability (ISNet). Our ISNet reaps the advantages from two strategies applied to refining bitemporal feature hierarchies: 1) margin maximization that clarifies the gap between changed and unchanged semantics and 2) targeted arrangement of attention mechanisms that direct the use of channel attention (CA) and spatial attention (SA) for highlighting semantic and positional information, respectively. Specifically, we insert CA modules into share-weighted backbone networks to facilitate semantic-specific feature extraction. The semantic boundaries in the extracted bitemporal hierarchical features are then clarified by margin maximization modules, followed by SA modules to enhance positional change responses. A top–down fusion pathway makes the final refined features cover multiscale representations and have strong separability for remote sensing image change detection. Extensive experimental evaluations demonstrate that our ISNet achieves state-of-the-art performance on the LEVIR-CD, SYSU-CD, and Season-Varying datasets in terms of overall accuracy (OA), Intersection-of-Union (IoU), and F1 score. Code is available at https://github.com/xingronaldo/ISNet. [ABSTRACT FROM AUTHOR]

Visual sound separation (VSS) aims to obtain each sound component from the mixed audio signals with the guidance of visual information. Existing works mainly capture the global-level audio-visual correspondence and exploit various visual features to enhance the appearance and motion features of visual modality. However, they commonly neglect the intrinsic properties of the audio modality, resulting in less effective audio feature extraction and unbalanced audio-visual correspondence. To tackle this problem, we propose a novel end-to-end framework termed Modeling Two-Stream Correspondence (MTSC) for VSS by explicitly extracting the timbre and content features in audio modality. The proposed MTSC method employs a two-stream architecture to enhance audio-visual correspondence for both the appearance-timbre and motion-content features. Moreover, with the advanced two-stream pipeline, more lightweight appearance and motion features for visual modality are exploited. Extensive experiments conducted on two benchmark musical instrument datasets demonstrate that with the above properties, our MTSC method remarkably outperforms seven state-of-the-art VSS approaches. The implementation code and extensive experimental results of the proposed MTSC method are provided at https://github.com/CFM-MSG/MTSC-VSS. [ABSTRACT FROM AUTHOR]

The scattering features of objects in synthetic aperture radar (SAR) imagery are highly sensitive to different azimuth angles, and detecting azimuth-sensitive objects in complex scenes becomes a challenging task. To address this issue, we propose a novel framework called the spatial orientation attention enhancement network (SOAEN) by using aircraft detection in complex scenes of SAR imagery as a case study. Taking YOLOX as the basic framework, this framework introduces the inverted pyramid ConvMixer network (IPCN), the spatial-orientation-enhanced path aggregation feature pyramid network (SOEPAFPN), and the anchor-free decoupled head (AFDH) to achieve performance improvement. A spatial orientation attention module is proposed and introduced into the path aggregation feature pyramid network to form a new structure, the SOEPAFPN, for capturing feature transformations in different directions, highlighting object features and suppressing background effects; the IPCN is adapted to replace the backbone network of YOLOX for enhancing the multiscale feature extraction capability and reducing the computational complexity, while the AFDH is used to decouple object localization and classification to improve the efficiency and accuracy of object localization and classification. The experimental results of the multiple real complex scenes on Gaofen-3 1 m images show that the proposed method achieves the highest detection accuracy, with an average detection rate of 91.22% compared with the YOLO series networks. [ABSTRACT FROM AUTHOR]

As the basic spatial unit of urban planning and management, it is necessary to know the distribution status of urban functional areas in time. Due to the complexity of urban land use, it is difficult to identify the urban functional areas using only remote sensing images. Social perception data can provide additional information for the identification of urban functional areas. However, the sources of remote sensing data and social perception data differ, with some differences in data forms. Existing methods cannot comprehensively consider the characteristics of these data for functional area identification. Therefore, in this study, we propose a multimodal deep learning method with an attention mechanism to fully utilize the data features of these two modalities and apply it to the recognition of urban functional areas. First, the pre-processed remote sensing images, points of interest, and building footprint data are divided into block-based target units of features by the road network. Next, the remote sensing image features and social perception data features of the target unit are extracted separately using a two-branch convolutional network. Finally, features are extracted sequentially along two separate dimensions, being channel and spatial, to generate an attention weight map for the identification and classification mapping of urban functional areas. The model framework was finally applied to the Ningbo dataset for testing, and the recognition accuracy was above 93%. The experimental results deduce, as a whole, that the prediction performance of the deep multimodal fusion model framework with an attention mechanism is comparatively superior to other traditional methods. It can provide a reference for the classification of urban land use and provide data support for urban planning and management. [ABSTRACT FROM AUTHOR]

Natural rubber is an essential raw material for industrial products and plays an important role in social development. A variety of diseases can affect the growth of rubber trees, reducing the production and quality of natural rubber. Therefore, it is of great significance to automatically identify rubber leaf disease. However, in practice, different diseases have complex morphological characteristics of spots and symptoms at different stages and scales, and there are subtle interclass differences and large intraclass variation between the symptoms of diseases. To tackle these challenges, a group multi-scale attention network (GMA-Net) was proposed for rubber leaf disease image recognition. The key idea of our method is to develop a group multi-scale dilated convolution (GMDC) module for multi-scale feature extraction as well as a cross-scale attention feature fusion (CAFF) module for multi-scale attention feature fusion. Specifically, the model uses a group convolution structure to reduce model parameters and provide multiple branches and then embeds multiple dilated convolutions to improve the model's adaptability to the scale variability of disease spots. Furthermore, the CAFF module is further designed to drive the network to learn the attentional features of multi-scale diseases and strengthen the disease features fusion at different scales. In this article, a dataset of rubber leaf diseases was constructed, including 2,788 images of four rubber leaf diseases and healthy leaves. Experimental results show that the accuracy of the model is 98.06%, which was better than other state-of-the-art approaches. Moreover, the model parameters of GMA-Net are only 0.65 M, and the model size is only 5.62 MB. Compared with MobileNetV1, V2, and ShuffleNetV1, V2 lightweight models, the model parameters and size are reduced by more than half, but the recognition accuracy is also improved by 3.86–6.1%. In addition, to verify the robustness of this model, we have also verified it on the PlantVillage public dataset. The experimental results show that the recognition accuracy of our proposed model is 99.43% on the PlantVillage dataset, which is also better than other state-of-the-art approaches. The effectiveness of the proposed method is verified, and it can be used for plant disease recognition. [ABSTRACT FROM AUTHOR]

Hyperspectral image (HSI) data have large numbers of bands that probably not all bands are equally informative and predictive for an effective HSI classification. Effective algorithms are highly desired in many real-world HSI applications, especially in cases requiring rapid learning with limited computing power. To address the abovementioned case, we present in this article a novel plug-and-play compact band weighting (CBW) module based on the attention-driven mechanism that evaluates different spectral bands according to their contributions to a given classification task. Compared to existing band weighting (BW) modules with tens of thousands of network parameters by deep learning, the proposed CBW is a lightweight module with only 20 parameters. Both model complexity and time cost are significantly reduced. The CBW module implements BW by making full use of the correlation among the adjacent spectral bands and spectral statistic information and, thereby, leads to the effect of recalibrated HSI. The experimental study has been conducted on three widely used HSI data sets, and results show the superiority of the proposed algorithm over current state-of-the-art methods of BW. The source code is available at https://github.com/JarvenYi/CBW. [ABSTRACT FROM AUTHOR]

• This is a polyp segmentation framework with advanced performance on five datasets. • A Multi-scale attention module is designed to extract detail information. • It fuses high-level features and mitigates semantic conflicts. • We construct a module to fuse same-level features and improve decoder. Clinically, automated polyp segmentation techniques have the potential to significantly improve the efficiency and accuracy of medical diagnosis, thereby reducing the risk of colorectal cancer in patients. Unfortunately, existing methods suffer from two significant weaknesses that can impact the accuracy of segmentation. Firstly, features extracted by encoders are not adequately filtered and utilized. Secondly, semantic conflicts and information redundancy caused by feature fusion are not attended to. To overcome these limitations, we propose a novel approach for polyp segmentation, named MLFF-Net, which leverages multi-level feature fusion and attention mechanisms. Specifically, MLFF-Net comprises three modules: Multi-scale Attention Module (MAM), High-level Feature Enhancement Module (HFEM), and Global Attention Module (GAM). Among these, MAM is used to extract multi-scale information and polyp details from the shallow output of the encoder. In HFEM, the deep features of the encoders complement each other by aggregation. Meanwhile, the attention mechanism redistributes the weight of the aggregated features, weakening the conflicting redundant parts and highlighting the information useful to the task. GAM combines features from the encoder and decoder features, as well as computes global dependencies to prevent receptive field locality. Experimental results on five public datasets show that the proposed method not only can segment multiple types of polyps but also has advantages over current state-of-the-art methods in both accuracy and generalization ability. [ABSTRACT FROM AUTHOR]

Recently, deep learning based video super-resolution (SR) methods combine the convolutional neural networks (CNN) with motion compensation to estimate a high-resolution (HR) video from its low-resolution (LR) counterpart. However, most previous methods conduct downscaling motion estimation to handle large motions, which can lead to detrimental effects on the accuracy of motion estimation due to the reduction of spatial resolution. Besides, these methods usually treat different types of intermediate features equally, which lack flexibility to emphasize meaningful information for revealing the high-frequency details. In this paper, to solve above issues, we propose a deep dual attention network (DDAN), including a motion compensation network (MCNet) and a SR reconstruction network (ReconNet), to fully exploit the spatio-temporal informative features for accurate video SR. The MCNet progressively learns the optical flow representations to synthesize the motion information across adjacent frames in a pyramid fashion. To decrease the mis-registration errors caused by the optical flow based motion compensation, we extract the detail components of original LR neighboring frames as complementary information for accurate feature extraction. In the ReconNet, we implement dual attention mechanisms on a residual unit and form a residual attention unit to focus on the intermediate informative features for high-frequency details recovery. Extensive experimental results on numerous datasets demonstrate the proposed method can effectively achieve superior performance in terms of quantitative and qualitative assessments compared with state-of-the-art methods. [ABSTRACT FROM AUTHOR]