Search

Background: Emergence of antibiotic resistance in bacteria is an important threat to global health. Antibiotic resistance genes (ARGs) are some of the key components to define bacterial resistance and their spread in different environments. Identification of ARGs, particularly from high-throughput sequencing data of the specimens, is the state-of-the-art method for comprehensively monitoring their spread and evolution. Current computational methods to identify ARGs mainly rely on alignment-based sequence similarities with known ARGs. Such approaches are limited by choice of reference databases and may potentially miss novel ARGs. The similarity thresholds are usually simple and could not accommodate variations across different gene families and regions. It is also difficult to scale up when sequence data are increasing. Results: In this study, we developed ARGNet, a deep neural network that incorporates an unsupervised learning autoencoder model to identify ARGs and a multiclass classification convolutional neural network to classify ARGs that do not depend on sequence alignment. This approach enables a more efficient discovery of both known and novel ARGs. ARGNet accepts both amino acid and nucleotide sequences of variable lengths, from partial (30–50 aa; 100–150 nt) sequences to full-length protein or genes, allowing its application in both target sequencing and metagenomic sequencing. Our performance evaluation showed that ARGNet outperformed other deep learning models including DeepARG and HMD-ARG in most of the application scenarios especially quasi-negative test and the analysis of prediction consistency with phylogenetic tree. ARGNet has a reduced inference runtime by up to 57% relative to DeepARG. Conclusions: ARGNet is flexible, efficient, and accurate at predicting a broad range of ARGs from the sequencing data. ARGNet is freely available at https://github.com/id-bioinfo/ARGNet, with an online service provided at https://ARGNet.hku.hk. 7nR91jX7ySRMRPZrE5tsTy Video Abstract [ABSTRACT FROM AUTHOR]

EEG emotion recognition plays a significant role in various mental health services. Deep learning-based methods perform excellently, but still suffer from interpretability. Although methods such as Gradient-weighted Class Activation Mapping(Grad-CAM) can cope with the above problem, their coarse granularity cannot accurately reveal the mechanism to promote emotional intelligence. In this paper, fine-grained interpretability is proposed, called Concat-aided Grad-CAM. Specifically, the multi-level feature mapping before the fully connected layer is concatenated to obtain the gradients of the target concept so that the discriminant information can be directly located in the high-precision area. Unlike coarse-grained interpretability methods applied in EEG emotion recognition, it can accurately highlight the EEG channels related to emotion rather than an obscure area. In addition, a systematic brain functional network is proposed to reveal the relationship between those channels and to further improve emotion recognition performance. The channels with greater contributions are connected, and those connections are learned by dynamic graph convolutional networks, while the others are independent to eliminate interference. Experiments on four EEG emotion recognition datasets manifest that Concat-aided Grad-CAM can be interpreted by the fine-grained. In addition, it has been shown that the learned brain functional network can improve the performance of the baselines. Significantly, the experiment results achieve state-of-the-art performance in multiple experiments. [ABSTRACT FROM AUTHOR]

The moisture content of cigar leaves during drying is an important indicator for controlling the management of drying rooms. At present, the determination of cigar leaf moisture content is mainly dependent on traditional destructive detection methods, which are inefficient and damaging to plants. In this study, a Convolution Neural Network method consisting of digital images for monitoring the moisture content of cigar leaves during the drying process was proposed. In this study, the Convolution Neural Network model was trained to learn the relationship between the images and the corresponding moisture content using the extracted colour, shape, and texture features as input factors. In order to compare the Convolution Neural Network estimation results, a widely used traditional machine learning algorithm was applied. The results demonstrated that the estimated value of Convolution Neural Network agreed with the predicted value; the R² was 0.9044, and the average accuracy was 87.34%. These results were better than those produced by traditional machine learning methods. The generalization test of the proposed method was conducted using varieties of cigar leaves in other drying rooms. The results showed that Convolution Neural Network is a viable method for an accurate estimation of the moisture content, the R² was 0.8673 and the average accuracy was 86.81%. The Convolution Neural Network established by the features extracted from digital images could accurately estimate the moisture content of cigar leaves during drying and was therefore shown to be an effective monitoring tool. [ABSTRACT FROM AUTHOR]

In the process of stereo super-resolution reconstruction, in addition to the richness of the extracted feature information directly affecting the texture details of the reconstructed image, the texture details of the corresponding pixels between stereo image pairs also have an important impact on the reconstruction accuracy in the process of network learning. Therefore, aiming at the information interaction and stereo consistency of stereo image pairs, a cross-view attention interaction fusion stereo super-resolution algorithm is proposed. Firstly, based on parallax attention mechanism and triple attention mechanism, an attention stereo fusion module is constructed. The attention stereo fusion module is inserted between different levels of two single image super-resolution network branches, and the attention weight is calculated through the cross dimensional interaction of the three branches. It makes full use of the ability of single image super-resolution network to extract single view information and further maintaining the stereo consistency between stereo image pairs. Then, an enhanced cross-view interaction strategy including three fusion methods is proposed. Specifically, the vertical sparse fusion method is used to integrate the interior view information of different levels in the two single image super-resolution sub branches, the horizontal dense fusion method is used to connect the adjacent attention stereo fusion modules and the constraint between stereo image consistency is further strengthened in combination with the feature fusion method. Finally, the experimental results on Flickr 1024, Middlebury and KITTI benchmark datasets show that the proposed algorithm is superior to the existing stereo image super-resolution methods in quantitative measurement and qualitative visual quality while maintaining the stereo consistency of image pairs. [ABSTRACT FROM AUTHOR]

Remote sensing image classification (RSIC) is a classical and fundamental task in the intelligent interpretation of remote sensing imagery, which can provide unique labeling information for each acquired remote sensing image. Thanks to the potent global context information extraction ability of the multi-head self-attention (MSA) mechanism, visual transformer (ViT)-based architectures have shown excellent capability in natural scene image classification. However, in order to achieve powerful RSIC performance, it is insufficient to capture global spatial information alone. Specifically, for fine-grained target recognition tasks with high inter-class similarity, discriminative and effective local feature representations are key to correct classification. In addition, due to the lack of inductive biases, the powerful global spatial context representation capability of ViT requires lengthy training procedures and large-scale pre-training data volume. To solve the above problems, a hybrid architecture of convolution neural network (CNN) and ViT is proposed to improve the RSIC ability, called P 2 FEViT, which integrates plug-and-play CNN features with ViT. In this paper, the feature representation capabilities of CNN and ViT applying for RSIC are first analyzed. Second, aiming to integrate the advantages of CNN and ViT, a novel approach embedding CNN features into the ViT architecture is proposed, which can make the model synchronously capture and fuse global context and local multimodal information to further improve the classification capability of ViT. Third, based on the hybrid structure, only a simple cross-entropy loss is employed for model training. The model can also have rapid and comfortable convergence with relatively less training data than the original ViT. Finally, extensive experiments are conducted on the public and challenging remote sensing scene classification dataset of NWPU-RESISC45 (NWPU-R45) and the self-built fine-grained target classification dataset called BIT-AFGR50. The experimental results demonstrate that the proposed P 2 FEViT can effectively improve the feature description capability and obtain outstanding image classification performance, while significantly reducing the high dependence of ViT on large-scale pre-training data volume and accelerating the convergence speed. The code and self-built dataset will be released at our webpages. [ABSTRACT FROM AUTHOR]

The automatic identification of construction projects, which can be considered as complex scenes, is a technical challenge for the supervision of soil and water conservation in urban areas. Construction projects in high-resolution remote sensing images have no unified semantic definition, thereby exhibiting significant differences in image features. This paper proposes an identification method for construction projects based on the detection of detailed ground objects, which construction projects comprise, including movable slab houses, buildings under construction, dust screens, and bare soil (rock). To create the training data set, we select highly informative detailed ground objects from high-resolution remote sensing images. Then, the Faster RCNN (region-based convolutional neural network) algorithm is used to detect construction projects and the highly informative detailed ground objects separately. The merging of detection boxes and the correction of detailed ground object combinations are used to jointly improve the confidence of construction project detection results. The empirical experiments show that the accuracy evaluation indicators of this method on a data set of Wuhan construction projects outperform other comparative methods, and its AP value and F1 score reached 0.773 and 0.417, respectively. The proposed method can achieve satisfactory identification results for construction projects with complex scenes, and can be applied to the comprehensive supervision of soil and water conservation in construction projects. [ABSTRACT FROM AUTHOR]

In recent years, some research results have been achieved in the field of image steganalysis. However, there are still problems of difficulty in extracting steganographic features from images with low embedding rates and unsatisfactory detection performance of steganalysis. In this paper, we propose an image steganalysis method based on the attention mechanism and transfer learning. The method constructs a network model based on a convolutional neural network, including a preprocessing layer, a transposed convolutional layer, an ordinary convolutional layer, and a fully connected layer. We introduce the efficient channel attention module after the ordinary convolutional layer to focus on the steganographic region of the image, capture the local cross-channel interaction information, realize the adaptive adjustment of feature weights, and enhance the ability to extract steganographic features. Meanwhile, we apply the transfer learning method to use the training model parameters of high embedding rate images as the initialization parameters of the training model of the low embedding rate to achieve feature migration and further improve the steganalysis performance of the low embedding rate. The experimental results show that compared to the typical Xu-Net and Yedroudj-Net models, the detection accuracy of the proposed method is improved by 16.36% to 30.66% and by 35.59 to 37.83% for the embedding rates of 0.05 bpp, 0.1 bpp, and 0.2 bpp, respectively. Compared to the state-of-the-art Shen-Net model with low embedding rates, the detection accuracy is improved by 3.43% to 6.41%. This demonstrates the higher detection performance of the proposed method for steganalysis of low embedding rate images. [ABSTRACT FROM AUTHOR]

Graph convolutional neural network (GCN) has effectively boosted the multi-label image recognition task by modeling correlation among labels. In previous methods, label correlation is computed based on statistical information through label diffusion, and therefore the same for all samples. This, however, makes graph inference on labels insufficient to handle huge variations among numerous image instances. In this paper, we propose an instance-aware graph convolutional neural network (IA_GCN) framework for the multi-label classification. As a whole, two fused branches of sub-networks are involved in the framework: a global branch modeling the whole image and a local branch exploring dependencies among regions of interests (ROIs). For both the branches, an image-dependent label correlation matrix (ID_LCM), fusing both the statistical label correlation matrix (LCM) and an individual one of each image instance, is constructed to inject adaptive information of label-awareness into the learned features of the model through graph convolution. Specifically, the individual LCM of each image is obtained by mining the label dependencies based on the predicted label scores of those detected ROIs. In this process, considering the contribution differences of ROIs to multi-label classification, variational inference is introduced to learn adaptive scaling factors for those ROIs by considering their complex distribution. Finally, extensive experiments on MS-COCO and VOC datasets show that our proposed approach outperforms existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Deep learning, especially 3D convolutional neural networks (CNNs), has been proved to be an excellent feature extractor in the hyperspectral image (HSI) classification. However, simply accumulating conventional 3D convolution units and blindly increasing the depth of the network does not improve the model performance effectively. Besides, most deep learning models tend to struggle due to the serious overfitting problem under the condition of small sample, this seriously restricts the accuracy of model classification. To solve the abovementioned problems, we proposed a multiscale 3D convolution with context attention network for HSI classification. Specifically, we introduce a multiscale 3D convolution composed of convolution kernels of different sizes to replace the conventional 3D convolution to enlarge the receptive field and adaptively detect the HSI features in different scales. Then, based on multiscale 3D convolution, we build two subnetworks to efficiently exploit hierarchical spectral and spatial features respectively, and enhance the transmission of features. Finally, to explore the discriminative features further, we design two types of attention mechanisms (AM) to build compact relationships between each position\channel and aggregation center instead of model any position\channel and position\channel relationships. After each 3D convolution layer, a compact AM is adopted to refine extracted hierarchical spectral and spatial features respectively, and boost the performance of the model. Experiments were conducted on four benchmark HSI datasets, the results demonstrate that the proposed method outperforms state-of-the-art models with the overall accuracy of 96.39%, 97.83%, 98.58%, and 97.98% over Indian Pines, Salinas Valley, Pavia University and Botswana dataset, respectively. [ABSTRACT FROM AUTHOR]

Remote sensing image analysis is a basic and practical research hotspot in remote sensing science. Remote sensing images contain abundant ground object information and it can be used in urban planning, agricultural monitoring, ecological services, geological exploration and other aspects. In this paper, we propose a lightweight model combining vgg-16 and u-net network. By combining two convolutional neural networks, we classify scenes of remote sensing images. While ensuring the accuracy of the model, try to reduce the memory of themodel. According to the experimental results of this paper, we have improved the accuracy of the model to 98%. The memory size of the model is 3.4 MB. At the same time, The classification and convergence speed of the model are greatly improved. We simultaneously take the remote sensing scene image of 64 × 64 as input into the designed model. As the accuracy of the model is 97%, it is proved that the model designed in this paper is also suitable for remote sensing images with few target feature points and low accuracy. Therefore, the model has a good application prospect in the classification of remote sensing images with few target feature points and low pixels. [ABSTRACT FROM AUTHOR]

Semantic segmentation plays a critical role in scene understanding for self-driving vehicles. A line of efforts has proven that global context matters in urban scene segmentation due to massive scale changes. However, we find that existing methods suffer from local ambiguities when dissipating continuous local context, i.e. scrambling to a huge receptive field of global cues by coarse pooling. To this end, this paper proposes a new Context Aggregation Module (CAM) that consists of two primary components: context encoding using no coarse pooling but encoder-decoders with appropriate sampling scales and gated fusion that extends gate attention mechanism to balance different-scale context during feature fusion. Weeding out coarse pooling and applying the encoder-decoder inherits the merits of exploring global context while avoiding the drawback of losing local contextual continuity. We then construct a Context Aggregation Network (CANet) and conduct extensive evaluations on challenging autonomous driving benchmarks of Cityscapes, CamVid and BDD100K. Consistently improved results evidence the effectiveness. Notably, we attain competitive mIoU 82.7% on Cityscapes and optimal mIoU 80.5% on CamVid. [ABSTRACT FROM AUTHOR]

Human pose estimation, a vital pursuit in the realm of computer vision, aims to predict the spatial coordinates of key points within images. Despite the advancements achieved by employing a Convolution Neural Network (CNN), this task still faces considerable challenges, especially in handling occlusion and overfitting issues. This paper introduces a new human pose estimation network designed to address the challenges posed by occluded and blurred images. It features a multi-scale spatial attention mechanism that zeroes in on the human body, significantly improving feature extraction for complex images. Moreover, this versatile attention module is compatible with a wide range of convolutional neural network-based pose estimation frameworks, unlike other mechanisms restricted to particular networks. Addressing the overfitting issue in human pose estimation models, this paper introduces an adversarial network-based data augmentation technique. A generator specifically tailored for pose estimation is adversarially trained to produce optimal augmentation samples, thereby reducing model overfitting. Experimental validation confirms that this augmentation method notably enhances the prediction accuracy of the pose estimation model without incurring extra computational costs. In addition, this paper introduces a streamlined Feature Pyramid Network (FPN) that enables shallow networks to assimilate extensive-scale data, addressing the issue of excessive model size. The experimental validation on the benchmark datasets MPII and MSCOCO demonstrates the efficacy of this integrated approach, showcasing significant improvements in the accuracy and the overall performance of human pose estimation and surpassing the existing methodologies. This approach effectively enhances the performance of the baseline model, achieving the best accuracy of 92.2% and 80.4% on the MPII and MSCOCO, respectively. • MSSA module refines spatial focus in complex scenes via multi-scale features. • ADA module boosts pose detection with occlusion and discriminator. • Self-distilled HRNet-MSSA optimises feature detection on compact devices. [ABSTRACT FROM AUTHOR]

Benggang is a typical erosional landform in southern and southeastern China. Since benggang poses significant risks to local ecological environments and economic infrastructure, it is vital to accurately detect benggang-eroded areas. Relying only on remote sensing imagery for benggang detection cannot produce satisfactory results. In this study, we propose integrating high-resolution Digital Orthophoto Map (DOM) and Digital Surface Model (DSM) data for efficient and automatic benggang discovery. The fusion of complementary rich information hidden in both DOM and DSM data is realized by a two-stream convolutional neural network (CNN), which integrates aggregated terrain and activation image features that are both extracted by supervised deep learning. We aggregate local low-level geomorphic features via a supervised diffusion-convolutional embedding branch for expressive representations of benggang terrain variations. Activation image features are obtained from an image-oriented convolutional neural network branch. The two sources of information (DOM and DSM) are fused via a gated neural network, which learns the most discriminative features for the detection of benggang. The evaluation of a challenging benggang dataset demonstrates that our method exceeds several baselines, even with limited training examples. The results show that the fusion of DOM and DSM data is beneficial for benggang detection via supervised convolutional and deep fusion networks. [ABSTRACT FROM AUTHOR]

Deep-learning technologies, especially convolutional neural networks (CNNs), have achieved great success in building extraction from areal images. However, shape details are often lost during the down-sampling process, which results in discontinuous segmentation or inaccurate segmentation boundary. In order to compensate for the loss of shape information, two shape-related auxiliary tasks (i.e., boundary prediction and distance estimation) were jointly learned with building segmentation task in our proposed network. Meanwhile, two consistency constraint losses were designed based on the multi-task network to exploit the duality between the mask prediction and two shape-related information predictions. Specifically, an atrous spatial pyramid pooling (ASPP) module was appended to the top of the encoder of a U-shaped network to obtain multi-scale features. Based on the multi-scale features, one regression loss and two classification losses were used for predicting the distance-transform map, segmentation, and boundary. Two inter-task consistency-loss functions were constructed to ensure the consistency between distance maps and masks, and the consistency between masks and boundary maps. Experimental results on three public aerial image data sets showed that our method achieved superior performance over the recent state-of-the-art models. [ABSTRACT FROM AUTHOR]

Remote sensing images contain complex backgrounds and multi-scale objects, which pose a challenging task for scene classification. The performance is highly dependent on the capacity of the scene representation as well as the discriminability of the classifier. Although multiple models possess better properties than a single model on these aspects, the fusion strategy for these models is a key component to maximize the final accuracy. In this paper, we construct a novel dual-model architecture with a grouping-attention-fusion strategy to improve the performance of scene classification. Specifically, the model employs two different convolutional neural networks (CNNs) for feature extraction, where the grouping-attention-fusion strategy is used to fuse the features of the CNNs in a fine and multi-scale manner. In this way, the resultant feature representation of the scene is enhanced. Moreover, to address the issue of similar appearances between different scenes, we develop a loss function which encourages small intra-class diversities and large inter-class distances. Extensive experiments are conducted on four scene classification datasets include the UCM land-use dataset, the WHU-RS19 dataset, the AID dataset, and the OPTIMAL-31 dataset. The experimental results demonstrate the superiority of the proposed method in comparison with the state-of-the-arts. [ABSTRACT FROM AUTHOR]

• Enhancing route choice modeling process with data- and knowledge-driven approach. • Conditional variational autoencoder approach to choice-set generation. • A data- and knowledge-driven route choice model with implicit availability/perception of alternatives is proposed. • Convolutional neural networks embedded into MNL to learn tastes. • Novel integration of neural networks into discrete choice models. Two central and interconnected problems arise in the specification of a "complete" path-based route choice model: choice-set generation and choice from a choice set. Choice-set generation poses a significant challenge in personalization and the enumeration of the full choice set with large size. Despite the continued prevalence of classic econometric models for modeling choices within a given set, this requirement of knowledge-driven modeling necessitates explicit model structures and intricate domain knowledge, which may result in practical biases. In this study, a Conditional Variational AutoEncoder (CVAE)-based choice set generation model is developed, which approximates the probability distribution of the underlying choice set generation process conditional on individual and OD characteristics without relying on expert knowledge. In order to facilitate a friendly integration between knowledge-driven econometric and machine learning approaches, a neural-embedded route choice model (IAP-NERCM) with implicit availability/perception (IAP) of choice alternatives is proposed to automatically capture the heterogeneity of taste parameters without assuming any a priori relationship. Results based on synthetic data show that the proposed models are capable of reproducing the pre-defined coefficients. Field data of GPS data collected in Toyota City is used to future test the proposed models compared to classical statistical models. Results indicate that IAP-NERCM exhibits the ability to recover underlying taste function and achieves the best performance in terms of goodness-of-fit, predictability, and estimation time. [ABSTRACT FROM AUTHOR]

• A novel autonomous quantitative damage evaluation approach, GW-SHMnet, by integrating PAGW technique and deep learning algorithm. • Effective quantitative damage evaluation of cracked curved plate structures with up to 100% accuracy. • Robust in quantitative damage evaluation with an accuracy of over 95% under 100% Gaussian noise. • Compared with the traditional curve-fitting method, the accuracy of GW-SHMnet has improved by about 10% in studied cases. Recent advances in phased array guided wave (PAGW) have demonstrated the potential of minor damage detection and localization in widely used curved plates, but quantitative damage evaluation remains difficult since effective features that are sensitive to damage size are hard to extract. In this study, a novel integrated framework, GW-SHMnet, is proposed, which leverages the advantages of the PAGW, finite element (FE) modeling, and deep learning algorithm. Firstly, an FE model is constructed to simulate PAGW propagation in curved plates. Secondly, PAGW experiments are performed on a curved aluminum plate to validate the FE model. Thirdly, an FE simulation database considering different sensor locations, testing frequencies, and damage sizes, is constructed and used as the training and testing data. Finally, deep learning is used to automatically extract features to determine damage size. The effectiveness, accuracy, and robustness of GW-SHMnet enable autonomous quantitative evaluation of minor damage in curved plates. [ABSTRACT FROM AUTHOR]

• The method diagnosed asymptomatic wheat scab effectively with insufficient samples. • The feasibility of transferring NIR spectral models was verified. • The generalization of model was improved by NAS and transfer learning fusion method. • CNN combined with RNN enhanced the performance of the transferred model. Wheat scab is one of the most important diseases endangering the health of wheat which severely affects the yield and quality of wheat. Thus, the diagnosis of wheat scab is very important. However, it is difficult to distinguish between asymptomatic wheats without visible spots on the surface and healthy wheats by traditional visual methods under natural conditions, which has greatly hindered the diagnosis of wheat scab. To address the challenge of poor model classification caused by the difficulty in distinguishing asymptomatic wheats from healthy wheats, we use near-infrared spectral data with healthy, symptomatic and indistinguishable asymptomatic wheats and propose a new approach Transfer Learning and Neural Architecture Search for Near-infrared with Convolutional Networks and Recurrent Networks (TranNas-NirCR). This approach integrates neural architecture search with transfer learning and employs a combination of convolutional neural networks and recurrent neural networks in the search space. Compared to other methods, the TranNas-NirCR method achieved better classification results with an accuracy of 90.42%, which is 2.68% higher than support vector machines (SVM), 5.36% higher than neural architecture search (NAS), and 4.21% higher than Transfer Learning with Neural Architecture Search (Tran_NAS). This method shows strong generalization performance in the case of only a small amount of near-infrared spectral data, which is of referential significance for diagnosing early wheat scab in real conditions. [ABSTRACT FROM AUTHOR]