In this paper,we propose hierarchical attention dual network(DNet)for fine-grained image classification.The DNet can randomly select pairs of inputs from the dataset and compare the differences between them through hi...In this paper,we propose hierarchical attention dual network(DNet)for fine-grained image classification.The DNet can randomly select pairs of inputs from the dataset and compare the differences between them through hierarchical attention feature learning,which are used simultaneously to remove noise and retain salient features.In the loss function,it considers the losses of difference in paired images according to the intra-variance and inter-variance.In addition,we also collect the disaster scene dataset from remote sensing images and apply the proposed method to disaster scene classification,which contains complex scenes and multiple types of disasters.Compared to other methods,experimental results show that the DNet with hierarchical attention is robust to different datasets and performs better.展开更多
Bird monitoring and protection are essential for maintaining biodiversity,and fine-grained bird classification has become a key focus in this field.Audio-visual modalities provide critical cues for this task,but robus...Bird monitoring and protection are essential for maintaining biodiversity,and fine-grained bird classification has become a key focus in this field.Audio-visual modalities provide critical cues for this task,but robust feature extraction and efficient fusion remain major challenges.We introduce a multi-stage fine-grained audiovisual fusion network(MSFG-AVFNet) for fine-grained bird species classification,which addresses these challenges through two key components:(1) the audiovisual feature extraction module,which adopts a multi-stage finetuning strategy to provide high-quality unimodal features,laying a solid foundation for modality fusion;(2) the audiovisual feature fusion module,which combines a max pooling aggregation strategy with a novel audiovisual loss function to achieve effective and robust feature fusion.Experiments were conducted on the self-built AVB81and the publicly available SSW60 datasets,which contain data from 81 and 60 bird species,respectively.Comprehensive experiments demonstrate that our approach achieves notable performance gains,outperforming existing state-of-the-art methods.These results highlight its effectiveness in leveraging audiovisual modalities for fine-grained bird classification and its potential to support ecological monitoring and biodiversity research.展开更多
The deep learning technology has shown impressive performance in various vision tasks such as image classification, object detection and semantic segmentation. In particular, recent advances of deep learning technique...The deep learning technology has shown impressive performance in various vision tasks such as image classification, object detection and semantic segmentation. In particular, recent advances of deep learning techniques bring encouraging performance to fine-grained image classification which aims to distinguish subordinate-level categories, such as bird species or dog breeds. This task is extremely challenging due to high intra-class and low inter-class variance. In this paper, we review four types of deep learning based fine-grained image classification approaches, including the general convolutional neural networks （CNNs）, part detection based, ensemble of networks based and visual attention based fine-grained image classification approaches. Besides, the deep learning based semantic segmentation approaches are also covered in this paper. The region proposal based and fully convolutional networks based approaches for semantic segmentation are introduced respectively.展开更多
Fine-grained image classification, which aims to distinguish images with subtle distinctions, is a challenging task for two main reasons: lack of sufficient training data for every class and difficulty in learning dis...Fine-grained image classification, which aims to distinguish images with subtle distinctions, is a challenging task for two main reasons: lack of sufficient training data for every class and difficulty in learning discriminative features for representation. In this paper, to address the two issues, we propose a two-phase framework for recognizing images from unseen fine-grained classes, i.e., zeroshot fine-grained classification. In the first feature learning phase, we finetune deep convolutional neural networks using hierarchical semantic structure among fine-grained classes to extract discriminative deep visual features. Meanwhile, a domain adaptation structure is induced into deep convolutional neural networks to avoid domain shift from training data to test data. In the second label inference phase, a semantic directed graph is constructed over attributes of fine-grained classes. Based on this graph, we develop a label propagation algorithm to infer the labels of images in the unseen classes. Experimental results on two benchmark datasets demonstrate that our model outperforms the state-of-the-art zero-shot learning models. In addition, the features obtained by our feature learning model also yield significant gains when they are used by other zero-shot learning models, which shows the flexility of our model in zero-shot finegrained classification.展开更多
In modern wireless communication and electromagnetic control,automatic modulationclassification(AMC)of orthogonal frequency division multiplexing(OFDM)signals plays animportant role.However,under Doppler frequency shi...In modern wireless communication and electromagnetic control,automatic modulationclassification(AMC)of orthogonal frequency division multiplexing(OFDM)signals plays animportant role.However,under Doppler frequency shift and complex multipath channel conditions,extracting discriminative features from high-order modulation signals and ensuring model inter-pretability remain challenging.To address these issues,this paper proposes a Fourier attention net-work(FAttNet),which combines an attention mechanism with a Fourier analysis network(FAN).Specifically,the method directly converts the input signal to the frequency domain using the FAN,thereby obtaining frequency features that reflect the periodic variations in amplitude and phase.Abuilt-in attention mechanism then automatically calculates the weights for each frequency band,focusing on the most discriminative components.This approach improves both classification accu-racy and model interpretability.Experimental validation was conducted via high-order modulationsimulation using an RF testbed.The results show that under three different Doppler frequencyshifts and complex multipath channel conditions,with a signal-to-noise ratio of 10 dB,the classifi-cation accuracy can reach 89.1%,90.4%and 90%,all of which are superior to the current main-stream methods.The proposed approach offers practical value for dynamic spectrum access and sig-nal security detection,and it makes important theoretical contributions to the application of deeplearning in complex electromagnetic signal recognition.展开更多
Automatic modulation classification(AMC)is an essential technique in both civil and military applications.While deep learning has surpassed traditional methods in accuracy,distinguishing high-order modulations remain ...Automatic modulation classification(AMC)is an essential technique in both civil and military applications.While deep learning has surpassed traditional methods in accuracy,distinguishing high-order modulations remain challenging.Current efforts prioritize complex network designs,neglecting the integration of deep features and tailored feature engineering to reslove high-order ambiguities.Therefore,a multi-feature extraction framework is proposed,which directly concatenates the deep feature extracted by a newly designed lightweight neural network and the proposed spectrum secondary features or de-noised high-order statistical features.The proposed features and lightweight network both demonstrate superior overall accuracy than other competing features or networks.Furthermore,the effectiveness of the feature extraction framework is also validated.The average classification accuracy on high-order modulation sets reaches 67.39% on the RadioML2018.01A dataset,increasing more than 2%compared with the other competitive networks under the framework.The results indicate the effectiveness of the proposed feature extraction framework for its representational ability by combing the deep features with the proposed domain features.展开更多
Fine-grained sedimentary rocks are defined as rocks which mainly compose of fine grains（〈62.5 μm）. The detailed studies on these rocks have revealed the need of a more unified, comprehensive and inclusive classifi...Fine-grained sedimentary rocks are defined as rocks which mainly compose of fine grains（〈62.5 μm）. The detailed studies on these rocks have revealed the need of a more unified, comprehensive and inclusive classification. The study focuses on fine-grained rocks has turned from the differences of inorganic mineral components to the significance of organic matter and microorganisms. The proposed classification is based on mineral composition, and it is noted that organic matters have been taken as a very important parameter in this classification scheme. Thus, four parameters, the TOC content, silica（quartz plus feldspars）, clay minerals and carbonate minerals, are considered to divide the fine-grained sedimentary rocks into eight categories, and the further classification within every category is refined depending on subordinate mineral composition. The nomenclature consists of a root name preceded by a primary adjective. The root names reflect mineral constituent of the rock, including low organic（TOC〈2%）, middle organic（2%4%） claystone, siliceous mudstone, limestone, and mixed mudstone. Primary adjectives convey structure and organic content information, including massive or limanited. The lithofacies are closely related to the reservoir storage space, porosity, permeability, hydrocarbon potential and shale oil/gas sweet spot, and are the key factor for the shale oil and gas exploration. The classification helps to systematically and practicably describe variability within fine-grained sedimentary rocks, what＇s more, it helps to guide the hydrocarbon exploration.展开更多
Based on reviews and summaries of the naming schemes of fine-grained sedimentary rocks, and analysis of characteristics of fine-grained sedimentary rocks, the problems existing in the classification and naming of fine...Based on reviews and summaries of the naming schemes of fine-grained sedimentary rocks, and analysis of characteristics of fine-grained sedimentary rocks, the problems existing in the classification and naming of fine-grained sedimentary rocks are discussed. On this basis, following the principle of three-level nomenclature, a new scheme of rock classification and naming for fine-grained sedimentary rocks is determined from two perspectives: First, fine-grained sedimentary rocks are divided into 12 types in two major categories, mudstone and siltstone, according to particle size(sand, silt and mud). Second,fine-grained sedimentary rocks are divided into 18 types in four categories, carbonate rock, fine-grained felsic sedimentary rock,clay rock and mixed fine-grained sedimentary rock according to mineral composition(carbonate minerals, felsic detrital minerals and clay minerals as three end elements). Considering the importance of organic matter in unconventional oil and gas generation and evaluation, organic matter is taken as the fourth element in the scheme. Taking the organic matter contents of 0.5% and 2% as dividing points, fine grained sedimentary rocks are divided into three categories, organic-poor, organic-bearing,and organic-rich ones. The new scheme meets the requirement of unconventional oil and gas exploration and development today and solves the problem of conceptual confusion in fine-grained sedimentary rocks, providing a unified basic term system for the research of fine-grained sedimentology.展开更多
Automatic modulation classification(AMC) technology is one of the cutting-edge technologies in cognitive radio communications. AMC based on deep learning has recently attracted much attention due to its superior perfo...Automatic modulation classification(AMC) technology is one of the cutting-edge technologies in cognitive radio communications. AMC based on deep learning has recently attracted much attention due to its superior performances in classification accuracy and robustness. In this paper, we propose a novel, high resolution and multi-scale feature fusion convolutional neural network model with a squeeze-excitation block, referred to as HRSENet,to classify different kinds of modulation signals.The proposed model establishes a parallel computing mechanism of multi-resolution feature maps through the multi-layer convolution operation, which effectively reduces the information loss caused by downsampling convolution. Moreover, through dense skipconnecting at the same resolution and up-sampling or down-sampling connection at different resolutions, the low resolution representation of the deep feature maps and the high resolution representation of the shallow feature maps are simultaneously extracted and fully integrated, which is benificial to mine signal multilevel features. Finally, the feature squeeze and excitation module embedded in the decoder is used to adjust the response weights between channels, further improving classification accuracy of proposed model.The proposed HRSENet significantly outperforms existing methods in terms of classification accuracy on the public dataset “Over the Air” in signal-to-noise(SNR) ranging from-2dB to 20dB. The classification accuracy in the proposed model achieves 85.36% and97.30% at 4dB and 10dB, respectively, with the improvement by 9.71% and 5.82% compared to LWNet.Furthermore, the model also has a moderate computation complexity compared with several state-of-the-art methods.展开更多
The remote sensing ships’fine-grained classification technology makes it possible to identify certain ship types in remote sensing images,and it has broad application prospects in civil and military fields.However,th...The remote sensing ships’fine-grained classification technology makes it possible to identify certain ship types in remote sensing images,and it has broad application prospects in civil and military fields.However,the current model does not examine the properties of ship targets in remote sensing images with mixed multi-granularity features and a complicated backdrop.There is still an opportunity for future enhancement of the classification impact.To solve the challenges brought by the above characteristics,this paper proposes a Metaformer and Residual fusion network based on Visual Attention Network(VAN-MR)for fine-grained classification tasks.For the complex background of remote sensing images,the VAN-MR model adopts the parallel structure of large kernel attention and spatial attention to enhance the model’s feature extraction ability of interest targets and improve the classification performance of remote sensing ship targets.For the problem of multi-grained feature mixing in remote sensing images,the VAN-MR model uses a Metaformer structure and a parallel network of residual modules to extract ship features.The parallel network has different depths,considering both high-level and lowlevel semantic information.The model achieves better classification performance in remote sensing ship images with multi-granularity mixing.Finally,the model achieves 88.73%and 94.56%accuracy on the public fine-grained ship collection-23(FGSC-23)and FGSCR-42 datasets,respectively,while the parameter size is only 53.47 M,the floating point operations is 9.9 G.The experimental results show that the classification effect of VAN-MR is superior to that of traditional CNNs model and visual model with Transformer structure under the same parameter quantity.展开更多
Medical image classification has played an important role in the medical field, and the related method based on deep learning has become an important and powerful technique in medical image classification. In this art...Medical image classification has played an important role in the medical field, and the related method based on deep learning has become an important and powerful technique in medical image classification. In this article, we propose a simplified inception module based Hadamard attention (SI + HA) mechanism for medical image classification. Specifically, we propose a new attention mechanism: Hadamard attention mechanism. It improves the accuracy of medical image classification without greatly increasing the complexity of the model. Meanwhile, we adopt a simplified inception module to improve the utilization of parameters. We use two medical image datasets to prove the superiority of our proposed method. In the BreakHis dataset, the AUCs of our method can reach 98.74%, 98.38%, 98.61% and 97.67% under the magnification factors of 40×, 100×, 200× and 400×, respectively. The accuracies can reach 95.67%, 94.17%, 94.53% and 94.12% under the magnification factors of 40×, 100×, 200× and 400×, respectively. In the KIMIA Path 960 dataset, the AUCs and accuracy of our method can reach 99.91% and 99.03%. It is superior to the currently popular methods and can significantly improve the effectiveness of medical image classification.展开更多
A nonparametric Bayesian method is presented to classify the MPSK （M-ary phase shift keying） signals. The MPSK signals with unknown signal noise ratios （SNRs） are modeled as a Gaussian mixture model with unknown m...A nonparametric Bayesian method is presented to classify the MPSK （M-ary phase shift keying） signals. The MPSK signals with unknown signal noise ratios （SNRs） are modeled as a Gaussian mixture model with unknown means and covariances in the constellation plane, and a clustering method is proposed to estimate the probability density of the MPSK signals. The method is based on the nonparametric Bayesian inference, which introduces the Dirichlet process as the prior probability of the mixture coefficient, and applies a normal inverse Wishart （NIW） distribution as the prior probability of the unknown mean and covariance. Then, according to the received signals, the parameters are adjusted by the Monte Carlo Markov chain （MCMC） random sampling algorithm. By iterations, the density estimation of the MPSK signals can be estimated. Simulation results show that the correct recognition ratio of 2/4/8PSK is greater than 95% under the condition that SNR 〉5 dB and 1 600 symbols are used in this method.展开更多
Deep Learning(DL)is such a powerful tool that we have seen tremendous success in areas such as Computer Vision,Speech Recognition,and Natural Language Processing.Since Automated Modulation Classification(AMC)is an imp...Deep Learning(DL)is such a powerful tool that we have seen tremendous success in areas such as Computer Vision,Speech Recognition,and Natural Language Processing.Since Automated Modulation Classification(AMC)is an important part in Cognitive Radio Networks,we try to explore its potential in solving signal modulation recognition problem.It cannot be overlooked that DL model is a complex model,thus making them prone to over-fitting.DL model requires many training data to combat with over-fitting,but adding high quality labels to training data manually is not always cheap and accessible,especially in real-time system,which may counter unprecedented data in dataset.Semi-supervised Learning is a way to exploit unlabeled data effectively to reduce over-fitting in DL.In this paper,we extend Generative Adversarial Networks(GANs)to the semi-supervised learning will show it is a method can be used to create a more dataefficient classifier.展开更多
To make the modulation classification system more suitable for signals in a wide range of signal to noise rate （SNR）, a feature extraction method based on signal wavelet packet transform modulus maxima matrix （WPT...To make the modulation classification system more suitable for signals in a wide range of signal to noise rate （SNR）, a feature extraction method based on signal wavelet packet transform modulus maxima matrix （WPTMMM） and a novel support vector machine fuzzy network （SVMFN） classifier is presented. The WPTMMM feature extraction method has less computational complexity, more stability, and has the preferable advantage of robust with the time parallel moving and white noise. Further, the SVMFN uses a new definition of fuzzy density that incorporates accuracy and uncertainty of the classifiers to improve recognition reliability to classify nine digital modulation types （i.e. 2ASK, 2FSK, 2PSK, 4ASK, 4FSK, 4PSK, 16QAM, MSK, and OQPSK）. Computer simulation shows that the proposed scheme has the advantages of high accuracy and reliability （success rates are over 98% when SNR is not lower than 0dB）, and it adapts to engineering applications.展开更多
Automatic modulation classification(AMC)aims at identifying the modulation of the received signals,which is a significant approach to identifying the target in military and civil applications.In this paper,a novel dat...Automatic modulation classification(AMC)aims at identifying the modulation of the received signals,which is a significant approach to identifying the target in military and civil applications.In this paper,a novel data-driven framework named convolutional and transformer-based deep neural network(CTDNN)is proposed to improve the classification performance.CTDNN can be divided into four modules,i.e.,convolutional neural network(CNN)backbone,transition module,transformer module,and final classifier.In the CNN backbone,a wide and deep convolution structure is designed,which consists of 1×15 convolution kernels and intensive cross-layer connections instead of traditional 1×3 kernels and sequential connections.In the transition module,a 1×1 convolution layer is utilized to compress the channels of the previous multi-scale CNN features.In the transformer module,three self-attention layers are designed for extracting global features and generating the classification vector.In the classifier,the final decision is made based on the maximum a posterior probability.Extensive simulations are conducted,and the result shows that our proposed CTDNN can achieve superior classification performance than traditional deep models.展开更多
An automatic method for classifying frequency shift keying(FSK),minimum shift keying(MSK),phase shift keying(PSK),quadrature amplitude modulation(QAM),and orthogonal frequency division multiplexing(OFDM)is proposed by...An automatic method for classifying frequency shift keying(FSK),minimum shift keying(MSK),phase shift keying(PSK),quadrature amplitude modulation(QAM),and orthogonal frequency division multiplexing(OFDM)is proposed by simultaneously using normality test,spectral analysis,and geometrical characteristics of in-phase-quadrature(I-Q)constellation diagram.Since the extracted features are unique for each modulation,they can be considered as a fingerprint of each modulation.We show that the proposed algorithm outperforms the previously published methods in terms of signal-to-noise ratio(SNR)and success rate.For example,the success rate of the proposed method for 64-QAM modulation at SNR=11 dB is 99%.Another advantage of the proposed method is its wide SNR range;such that the probability of classification for 16-QAM at SNR=3 dB is almost 1.The proposed method also provides a database for geometrical features of I-Q constellation diagram.By comparing and correlating the data of the provided database with the estimated I-Q diagram of the received signal,the processing gain of 4 dB is obtained.Whatever can be mentioned about the preference of the proposed algorithm are low complexity,low SNR,wide range of modulation set,and enhanced recognition at higher-order modulations.展开更多
Automatic Modulation Classification(AMC) is an important technology used to recognize the modulation type.A dictionary set was trained via signals with known modulation schemes in cooperative scenarios.Then we classif...Automatic Modulation Classification(AMC) is an important technology used to recognize the modulation type.A dictionary set was trained via signals with known modulation schemes in cooperative scenarios.Then we classify the modulation scheme of the signals received in the non-cooperative environment according to its sparse representation.Furthermore,we proposed a novel approach called Fast Block Coordinate descent Dictionary Learning(FBCDL).Moreover,the convergence of FBCDL was proved and we find that our proposed method achieves lower complexity.Experimental results indicate that our proposed FBCDL achieves better classification accuracy than traditional methods.展开更多
Radio modulation classification has always been an important technology in the field of communications.The difficulty of incremental learning in radio modulation classification is that learning new tasks will lead to ...Radio modulation classification has always been an important technology in the field of communications.The difficulty of incremental learning in radio modulation classification is that learning new tasks will lead to catastrophic forgetting of old tasks.In this paper,we propose a sample memory and recall framework for incremental learning of radio modulation classification.For data with different signal-to-noise ratios,we use a partial memory strategy by selecting appropriate samples for memorizing.We compare the performance of our proposed method with three baselines through a large number of simulation experiments.Results show that our method achieves far higher classification accuracy than finetuning method and feature extraction method.Furthermore,it performs closely to joint training method which uses all old data in terms of classification accuracy which validates the effectiveness of our method against catastrophic forgetting.展开更多
Modulation signal classification in communication systems can be considered a pattern recognition problem.Earlier works have focused on several feature extraction approaches such as fractal feature,signal constellatio...Modulation signal classification in communication systems can be considered a pattern recognition problem.Earlier works have focused on several feature extraction approaches such as fractal feature,signal constellation reconstruction,etc.The recent advent of deep learning(DL)models makes it possible to proficiently classify the modulation signals.In this view,this study designs a chaotic oppositional satin bowerbird optimization(COSBO)with bidirectional long term memory(BiLSTM)model for modulation signal classification in communication systems.The proposed COSBO-BiLSTM technique aims to classify the different kinds of digitally modulated signals.In addition,the fractal feature extraction process takes place by the use of Sevcik Fractal Dimension(SFD)approach.Moreover,the modulation signal classification process takes place using BiLSTM with fully convolutional network(BiLSTM-FCN).Furthermore,the optimal hyperparameter adjustment of the BiLSTM-FCN technique takes place by the use of COSBO algorithm.In order to ensure the enhanced classification performance of the COSBO-BiLSTM model,a wide range of simulations were carried out.The experimental results highlighted that the COSBO-BiLSTM technique has accomplished improved performance over the existing techniques.展开更多
Automatic modulation classification is the process of identification of the modulation type of a signal in a general environment. This paper proposes a new method to evaluate the tracking performance of large margin c...Automatic modulation classification is the process of identification of the modulation type of a signal in a general environment. This paper proposes a new method to evaluate the tracking performance of large margin classifier against signal-tonoise ratio （SNR）, and classifies all forms of primary user＇s signals in a cognitive radio environment. For achieving this objective, two structures of a large margin are developed in additive white Gaussian noise （AWGN） channels with priori unknown SNR. A combination of higher order statistics and instantaneous characteristics is selected as effective features. Simulation results show that the classification rates of the proposed structures are well robust against environmental SNR changes.展开更多
基金Supported by the National Natural Science Foundation of China(61601176)。
文摘In this paper,we propose hierarchical attention dual network(DNet)for fine-grained image classification.The DNet can randomly select pairs of inputs from the dataset and compare the differences between them through hierarchical attention feature learning,which are used simultaneously to remove noise and retain salient features.In the loss function,it considers the losses of difference in paired images according to the intra-variance and inter-variance.In addition,we also collect the disaster scene dataset from remote sensing images and apply the proposed method to disaster scene classification,which contains complex scenes and multiple types of disasters.Compared to other methods,experimental results show that the DNet with hierarchical attention is robust to different datasets and performs better.
基金supported by the Beijing Natural Science Foundation(No.5252014)the Open Fund of The Key Laboratory of Urban Ecological Environment Simulation and Protection,Ministry of Ecology and Environment of the People's Republic of China (No.UEESP-202502)the National Natural Science Foundation of China (No.62303063&32371874)。
文摘Bird monitoring and protection are essential for maintaining biodiversity,and fine-grained bird classification has become a key focus in this field.Audio-visual modalities provide critical cues for this task,but robust feature extraction and efficient fusion remain major challenges.We introduce a multi-stage fine-grained audiovisual fusion network(MSFG-AVFNet) for fine-grained bird species classification,which addresses these challenges through two key components:(1) the audiovisual feature extraction module,which adopts a multi-stage finetuning strategy to provide high-quality unimodal features,laying a solid foundation for modality fusion;(2) the audiovisual feature fusion module,which combines a max pooling aggregation strategy with a novel audiovisual loss function to achieve effective and robust feature fusion.Experiments were conducted on the self-built AVB81and the publicly available SSW60 datasets,which contain data from 81 and 60 bird species,respectively.Comprehensive experiments demonstrate that our approach achieves notable performance gains,outperforming existing state-of-the-art methods.These results highlight its effectiveness in leveraging audiovisual modalities for fine-grained bird classification and its potential to support ecological monitoring and biodiversity research.
基金supported by the National Natural Science Foundation of China(Nos.61373121 and 61328205)Program for Sichuan Provincial Science Fund for Distinguished Young Scholars(No.13QNJJ0149)+1 种基金the Fundamental Research Funds for the Central UniversitiesChina Scholarship Council(No.201507000032)
文摘The deep learning technology has shown impressive performance in various vision tasks such as image classification, object detection and semantic segmentation. In particular, recent advances of deep learning techniques bring encouraging performance to fine-grained image classification which aims to distinguish subordinate-level categories, such as bird species or dog breeds. This task is extremely challenging due to high intra-class and low inter-class variance. In this paper, we review four types of deep learning based fine-grained image classification approaches, including the general convolutional neural networks （CNNs）, part detection based, ensemble of networks based and visual attention based fine-grained image classification approaches. Besides, the deep learning based semantic segmentation approaches are also covered in this paper. The region proposal based and fully convolutional networks based approaches for semantic segmentation are introduced respectively.
基金supported by National Basic Research Program of China (973 Program) (No. 2015CB352502)National Nature Science Foundation of China (No. 61573026)Beijing Nature Science Foundation (No. L172037)
文摘Fine-grained image classification, which aims to distinguish images with subtle distinctions, is a challenging task for two main reasons: lack of sufficient training data for every class and difficulty in learning discriminative features for representation. In this paper, to address the two issues, we propose a two-phase framework for recognizing images from unseen fine-grained classes, i.e., zeroshot fine-grained classification. In the first feature learning phase, we finetune deep convolutional neural networks using hierarchical semantic structure among fine-grained classes to extract discriminative deep visual features. Meanwhile, a domain adaptation structure is induced into deep convolutional neural networks to avoid domain shift from training data to test data. In the second label inference phase, a semantic directed graph is constructed over attributes of fine-grained classes. Based on this graph, we develop a label propagation algorithm to infer the labels of images in the unseen classes. Experimental results on two benchmark datasets demonstrate that our model outperforms the state-of-the-art zero-shot learning models. In addition, the features obtained by our feature learning model also yield significant gains when they are used by other zero-shot learning models, which shows the flexility of our model in zero-shot finegrained classification.
基金supported by the National Natural Science Foundation of China(No.62027801).
文摘In modern wireless communication and electromagnetic control,automatic modulationclassification(AMC)of orthogonal frequency division multiplexing(OFDM)signals plays animportant role.However,under Doppler frequency shift and complex multipath channel conditions,extracting discriminative features from high-order modulation signals and ensuring model inter-pretability remain challenging.To address these issues,this paper proposes a Fourier attention net-work(FAttNet),which combines an attention mechanism with a Fourier analysis network(FAN).Specifically,the method directly converts the input signal to the frequency domain using the FAN,thereby obtaining frequency features that reflect the periodic variations in amplitude and phase.Abuilt-in attention mechanism then automatically calculates the weights for each frequency band,focusing on the most discriminative components.This approach improves both classification accu-racy and model interpretability.Experimental validation was conducted via high-order modulationsimulation using an RF testbed.The results show that under three different Doppler frequencyshifts and complex multipath channel conditions,with a signal-to-noise ratio of 10 dB,the classifi-cation accuracy can reach 89.1%,90.4%and 90%,all of which are superior to the current main-stream methods.The proposed approach offers practical value for dynamic spectrum access and sig-nal security detection,and it makes important theoretical contributions to the application of deeplearning in complex electromagnetic signal recognition.
基金supported by the National Natural Science Foundation of China(12273054).
文摘Automatic modulation classification(AMC)is an essential technique in both civil and military applications.While deep learning has surpassed traditional methods in accuracy,distinguishing high-order modulations remain challenging.Current efforts prioritize complex network designs,neglecting the integration of deep features and tailored feature engineering to reslove high-order ambiguities.Therefore,a multi-feature extraction framework is proposed,which directly concatenates the deep feature extracted by a newly designed lightweight neural network and the proposed spectrum secondary features or de-noised high-order statistical features.The proposed features and lightweight network both demonstrate superior overall accuracy than other competing features or networks.Furthermore,the effectiveness of the feature extraction framework is also validated.The average classification accuracy on high-order modulation sets reaches 67.39% on the RadioML2018.01A dataset,increasing more than 2%compared with the other competitive networks under the framework.The results indicate the effectiveness of the proposed feature extraction framework for its representational ability by combing the deep features with the proposed domain features.
基金supported by the Certificate of China Postdoctoral Science Foundation (No. 2015M582165)the National Natural Science Foundation of China (Nos. 41602142, 41772090)the National Science and Technology Special (No. 2017ZX05009-002)
文摘Fine-grained sedimentary rocks are defined as rocks which mainly compose of fine grains（〈62.5 μm）. The detailed studies on these rocks have revealed the need of a more unified, comprehensive and inclusive classification. The study focuses on fine-grained rocks has turned from the differences of inorganic mineral components to the significance of organic matter and microorganisms. The proposed classification is based on mineral composition, and it is noted that organic matters have been taken as a very important parameter in this classification scheme. Thus, four parameters, the TOC content, silica（quartz plus feldspars）, clay minerals and carbonate minerals, are considered to divide the fine-grained sedimentary rocks into eight categories, and the further classification within every category is refined depending on subordinate mineral composition. The nomenclature consists of a root name preceded by a primary adjective. The root names reflect mineral constituent of the rock, including low organic（TOC〈2%）, middle organic（2%4%） claystone, siliceous mudstone, limestone, and mixed mudstone. Primary adjectives convey structure and organic content information, including massive or limanited. The lithofacies are closely related to the reservoir storage space, porosity, permeability, hydrocarbon potential and shale oil/gas sweet spot, and are the key factor for the shale oil and gas exploration. The classification helps to systematically and practicably describe variability within fine-grained sedimentary rocks, what＇s more, it helps to guide the hydrocarbon exploration.
基金Supported by the National Natural Science Foundation of China (41872166)。
文摘Based on reviews and summaries of the naming schemes of fine-grained sedimentary rocks, and analysis of characteristics of fine-grained sedimentary rocks, the problems existing in the classification and naming of fine-grained sedimentary rocks are discussed. On this basis, following the principle of three-level nomenclature, a new scheme of rock classification and naming for fine-grained sedimentary rocks is determined from two perspectives: First, fine-grained sedimentary rocks are divided into 12 types in two major categories, mudstone and siltstone, according to particle size(sand, silt and mud). Second,fine-grained sedimentary rocks are divided into 18 types in four categories, carbonate rock, fine-grained felsic sedimentary rock,clay rock and mixed fine-grained sedimentary rock according to mineral composition(carbonate minerals, felsic detrital minerals and clay minerals as three end elements). Considering the importance of organic matter in unconventional oil and gas generation and evaluation, organic matter is taken as the fourth element in the scheme. Taking the organic matter contents of 0.5% and 2% as dividing points, fine grained sedimentary rocks are divided into three categories, organic-poor, organic-bearing,and organic-rich ones. The new scheme meets the requirement of unconventional oil and gas exploration and development today and solves the problem of conceptual confusion in fine-grained sedimentary rocks, providing a unified basic term system for the research of fine-grained sedimentology.
基金supported by the Beijing Natural Science Foundation (L202003)National Natural Science Foundation of China (No. 31700479)。
文摘Automatic modulation classification(AMC) technology is one of the cutting-edge technologies in cognitive radio communications. AMC based on deep learning has recently attracted much attention due to its superior performances in classification accuracy and robustness. In this paper, we propose a novel, high resolution and multi-scale feature fusion convolutional neural network model with a squeeze-excitation block, referred to as HRSENet,to classify different kinds of modulation signals.The proposed model establishes a parallel computing mechanism of multi-resolution feature maps through the multi-layer convolution operation, which effectively reduces the information loss caused by downsampling convolution. Moreover, through dense skipconnecting at the same resolution and up-sampling or down-sampling connection at different resolutions, the low resolution representation of the deep feature maps and the high resolution representation of the shallow feature maps are simultaneously extracted and fully integrated, which is benificial to mine signal multilevel features. Finally, the feature squeeze and excitation module embedded in the decoder is used to adjust the response weights between channels, further improving classification accuracy of proposed model.The proposed HRSENet significantly outperforms existing methods in terms of classification accuracy on the public dataset “Over the Air” in signal-to-noise(SNR) ranging from-2dB to 20dB. The classification accuracy in the proposed model achieves 85.36% and97.30% at 4dB and 10dB, respectively, with the improvement by 9.71% and 5.82% compared to LWNet.Furthermore, the model also has a moderate computation complexity compared with several state-of-the-art methods.
文摘The remote sensing ships’fine-grained classification technology makes it possible to identify certain ship types in remote sensing images,and it has broad application prospects in civil and military fields.However,the current model does not examine the properties of ship targets in remote sensing images with mixed multi-granularity features and a complicated backdrop.There is still an opportunity for future enhancement of the classification impact.To solve the challenges brought by the above characteristics,this paper proposes a Metaformer and Residual fusion network based on Visual Attention Network(VAN-MR)for fine-grained classification tasks.For the complex background of remote sensing images,the VAN-MR model adopts the parallel structure of large kernel attention and spatial attention to enhance the model’s feature extraction ability of interest targets and improve the classification performance of remote sensing ship targets.For the problem of multi-grained feature mixing in remote sensing images,the VAN-MR model uses a Metaformer structure and a parallel network of residual modules to extract ship features.The parallel network has different depths,considering both high-level and lowlevel semantic information.The model achieves better classification performance in remote sensing ship images with multi-granularity mixing.Finally,the model achieves 88.73%and 94.56%accuracy on the public fine-grained ship collection-23(FGSC-23)and FGSCR-42 datasets,respectively,while the parameter size is only 53.47 M,the floating point operations is 9.9 G.The experimental results show that the classification effect of VAN-MR is superior to that of traditional CNNs model and visual model with Transformer structure under the same parameter quantity.
文摘Medical image classification has played an important role in the medical field, and the related method based on deep learning has become an important and powerful technique in medical image classification. In this article, we propose a simplified inception module based Hadamard attention (SI + HA) mechanism for medical image classification. Specifically, we propose a new attention mechanism: Hadamard attention mechanism. It improves the accuracy of medical image classification without greatly increasing the complexity of the model. Meanwhile, we adopt a simplified inception module to improve the utilization of parameters. We use two medical image datasets to prove the superiority of our proposed method. In the BreakHis dataset, the AUCs of our method can reach 98.74%, 98.38%, 98.61% and 97.67% under the magnification factors of 40×, 100×, 200× and 400×, respectively. The accuracies can reach 95.67%, 94.17%, 94.53% and 94.12% under the magnification factors of 40×, 100×, 200× and 400×, respectively. In the KIMIA Path 960 dataset, the AUCs and accuracy of our method can reach 99.91% and 99.03%. It is superior to the currently popular methods and can significantly improve the effectiveness of medical image classification.
基金Cultivation Fund of the Key Scientific and Technical Innovation Project of Ministry of Education of China(No.3104001014)
文摘A nonparametric Bayesian method is presented to classify the MPSK （M-ary phase shift keying） signals. The MPSK signals with unknown signal noise ratios （SNRs） are modeled as a Gaussian mixture model with unknown means and covariances in the constellation plane, and a clustering method is proposed to estimate the probability density of the MPSK signals. The method is based on the nonparametric Bayesian inference, which introduces the Dirichlet process as the prior probability of the mixture coefficient, and applies a normal inverse Wishart （NIW） distribution as the prior probability of the unknown mean and covariance. Then, according to the received signals, the parameters are adjusted by the Monte Carlo Markov chain （MCMC） random sampling algorithm. By iterations, the density estimation of the MPSK signals can be estimated. Simulation results show that the correct recognition ratio of 2/4/8PSK is greater than 95% under the condition that SNR 〉5 dB and 1 600 symbols are used in this method.
基金This work is supported by the National Natural Science Foundation of China(Nos.61771154,61603239,61772454,6171101570).
文摘Deep Learning(DL)is such a powerful tool that we have seen tremendous success in areas such as Computer Vision,Speech Recognition,and Natural Language Processing.Since Automated Modulation Classification(AMC)is an important part in Cognitive Radio Networks,we try to explore its potential in solving signal modulation recognition problem.It cannot be overlooked that DL model is a complex model,thus making them prone to over-fitting.DL model requires many training data to combat with over-fitting,but adding high quality labels to training data manually is not always cheap and accessible,especially in real-time system,which may counter unprecedented data in dataset.Semi-supervised Learning is a way to exploit unlabeled data effectively to reduce over-fitting in DL.In this paper,we extend Generative Adversarial Networks(GANs)to the semi-supervised learning will show it is a method can be used to create a more dataefficient classifier.
文摘To make the modulation classification system more suitable for signals in a wide range of signal to noise rate （SNR）, a feature extraction method based on signal wavelet packet transform modulus maxima matrix （WPTMMM） and a novel support vector machine fuzzy network （SVMFN） classifier is presented. The WPTMMM feature extraction method has less computational complexity, more stability, and has the preferable advantage of robust with the time parallel moving and white noise. Further, the SVMFN uses a new definition of fuzzy density that incorporates accuracy and uncertainty of the classifiers to improve recognition reliability to classify nine digital modulation types （i.e. 2ASK, 2FSK, 2PSK, 4ASK, 4FSK, 4PSK, 16QAM, MSK, and OQPSK）. Computer simulation shows that the proposed scheme has the advantages of high accuracy and reliability （success rates are over 98% when SNR is not lower than 0dB）, and it adapts to engineering applications.
基金supported in part by the National Natural Science Foundation of China under Grant(62171045,62201090)in part by the National Key Research and Development Program of China under Grants(2020YFB1807602,2019YFB1804404).
文摘Automatic modulation classification(AMC)aims at identifying the modulation of the received signals,which is a significant approach to identifying the target in military and civil applications.In this paper,a novel data-driven framework named convolutional and transformer-based deep neural network(CTDNN)is proposed to improve the classification performance.CTDNN can be divided into four modules,i.e.,convolutional neural network(CNN)backbone,transition module,transformer module,and final classifier.In the CNN backbone,a wide and deep convolution structure is designed,which consists of 1×15 convolution kernels and intensive cross-layer connections instead of traditional 1×3 kernels and sequential connections.In the transition module,a 1×1 convolution layer is utilized to compress the channels of the previous multi-scale CNN features.In the transformer module,three self-attention layers are designed for extracting global features and generating the classification vector.In the classifier,the final decision is made based on the maximum a posterior probability.Extensive simulations are conducted,and the result shows that our proposed CTDNN can achieve superior classification performance than traditional deep models.
文摘An automatic method for classifying frequency shift keying(FSK),minimum shift keying(MSK),phase shift keying(PSK),quadrature amplitude modulation(QAM),and orthogonal frequency division multiplexing(OFDM)is proposed by simultaneously using normality test,spectral analysis,and geometrical characteristics of in-phase-quadrature(I-Q)constellation diagram.Since the extracted features are unique for each modulation,they can be considered as a fingerprint of each modulation.We show that the proposed algorithm outperforms the previously published methods in terms of signal-to-noise ratio(SNR)and success rate.For example,the success rate of the proposed method for 64-QAM modulation at SNR=11 dB is 99%.Another advantage of the proposed method is its wide SNR range;such that the probability of classification for 16-QAM at SNR=3 dB is almost 1.The proposed method also provides a database for geometrical features of I-Q constellation diagram.By comparing and correlating the data of the provided database with the estimated I-Q diagram of the received signal,the processing gain of 4 dB is obtained.Whatever can be mentioned about the preference of the proposed algorithm are low complexity,low SNR,wide range of modulation set,and enhanced recognition at higher-order modulations.
基金supported in part by the National Natural Science Foundation of China with grants 61525101,91746301,61631003,61601055the Shenzhen Fundamental Research Fund with grant KQTD2015033114415450
文摘Automatic Modulation Classification(AMC) is an important technology used to recognize the modulation type.A dictionary set was trained via signals with known modulation schemes in cooperative scenarios.Then we classify the modulation scheme of the signals received in the non-cooperative environment according to its sparse representation.Furthermore,we proposed a novel approach called Fast Block Coordinate descent Dictionary Learning(FBCDL).Moreover,the convergence of FBCDL was proved and we find that our proposed method achieves lower complexity.Experimental results indicate that our proposed FBCDL achieves better classification accuracy than traditional methods.
文摘Radio modulation classification has always been an important technology in the field of communications.The difficulty of incremental learning in radio modulation classification is that learning new tasks will lead to catastrophic forgetting of old tasks.In this paper,we propose a sample memory and recall framework for incremental learning of radio modulation classification.For data with different signal-to-noise ratios,we use a partial memory strategy by selecting appropriate samples for memorizing.We compare the performance of our proposed method with three baselines through a large number of simulation experiments.Results show that our method achieves far higher classification accuracy than finetuning method and feature extraction method.Furthermore,it performs closely to joint training method which uses all old data in terms of classification accuracy which validates the effectiveness of our method against catastrophic forgetting.
文摘Modulation signal classification in communication systems can be considered a pattern recognition problem.Earlier works have focused on several feature extraction approaches such as fractal feature,signal constellation reconstruction,etc.The recent advent of deep learning(DL)models makes it possible to proficiently classify the modulation signals.In this view,this study designs a chaotic oppositional satin bowerbird optimization(COSBO)with bidirectional long term memory(BiLSTM)model for modulation signal classification in communication systems.The proposed COSBO-BiLSTM technique aims to classify the different kinds of digitally modulated signals.In addition,the fractal feature extraction process takes place by the use of Sevcik Fractal Dimension(SFD)approach.Moreover,the modulation signal classification process takes place using BiLSTM with fully convolutional network(BiLSTM-FCN).Furthermore,the optimal hyperparameter adjustment of the BiLSTM-FCN technique takes place by the use of COSBO algorithm.In order to ensure the enhanced classification performance of the COSBO-BiLSTM model,a wide range of simulations were carried out.The experimental results highlighted that the COSBO-BiLSTM technique has accomplished improved performance over the existing techniques.
文摘Automatic modulation classification is the process of identification of the modulation type of a signal in a general environment. This paper proposes a new method to evaluate the tracking performance of large margin classifier against signal-tonoise ratio （SNR）, and classifies all forms of primary user＇s signals in a cognitive radio environment. For achieving this objective, two structures of a large margin are developed in additive white Gaussian noise （AWGN） channels with priori unknown SNR. A combination of higher order statistics and instantaneous characteristics is selected as effective features. Simulation results show that the classification rates of the proposed structures are well robust against environmental SNR changes.
