Application of graph neural network and feature information enhancement in relation inference of sparse knowledge graph

At present,knowledge embedding methods are widely used in the field of knowledge graph(KG)reasoning,and have been successfully applied to those with large entities and relationships.However,in research and production environments,there are a large number of KGs with a small number of entities and relations,which are called sparse KGs.Limited by the performance of knowledge extraction methods or some other reasons(some common-sense information does not appear in the natural corpus),the relation between entities is often incomplete.To solve this problem,a method of the graph neural network and information enhancement is proposed.The improved method increases the mean reciprocal rank(MRR)and Hit@3 by 1.6%and 1.7%,respectively,when the sparsity of the FB15K-237 dataset is 10%.When the sparsity is 50%,the evaluation indexes MRR and Hit@10 are increased by 0.8%and 1.8%,respectively.

Hyperspectral image classification based on spatial and spectral features and sparse representation

To minimize the low classification accuracy and low utilization of spatial information in traditional hyperspectral image classification methods, we propose a new hyperspectral image classification method, which is based on the Gabor spatial texture features and nonparametric weighted spectral features, and the sparse representation classification method（Gabor–NWSF and SRC）, abbreviated GNWSF–SRC. The proposed（GNWSF–SRC） method first combines the Gabor spatial features and nonparametric weighted spectral features to describe the hyperspectral image, and then applies the sparse representation method. Finally, the classification is obtained by analyzing the reconstruction error. We use the proposed method to process two typical hyperspectral data sets with different percentages of training samples. Theoretical analysis and simulation demonstrate that the proposed method improves the classification accuracy and Kappa coefficient compared with traditional classification methods and achieves better classification performance.

SRMD:Sparse Random Mode Decomposition

Signal decomposition and multiscale signal analysis provide many useful tools for timefrequency analysis.We proposed a random feature method for analyzing time-series data by constructing a sparse approximation to the spectrogram.The randomization is both in the time window locations and the frequency sampling,which lowers the overall sampling and computational cost.The sparsification of the spectrogram leads to a sharp separation between time-frequency clusters which makes it easier to identify intrinsic modes,and thus leads to a new data-driven mode decomposition.The applications include signal representation,outlier removal,and mode decomposition.On benchmark tests,we show that our approach outperforms other state-of-the-art decomposition methods.

Ship recognition based on HRRP via multi-scale sparse preserving method

In order to extract the richer feature information of ship targets from sea clutter, and address the high dimensional data problem, a method termed as multi-scale fusion kernel sparse preserving projection(MSFKSPP) based on the maximum margin criterion(MMC) is proposed for recognizing the class of ship targets utilizing the high-resolution range profile(HRRP). Multi-scale fusion is introduced to capture the local and detailed information in small-scale features, and the global and contour information in large-scale features, offering help to extract the edge information from sea clutter and further improving the target recognition accuracy. The proposed method can maximally preserve the multi-scale fusion sparse of data and maximize the class separability in the reduced dimensionality by reproducing kernel Hilbert space. Experimental results on the measured radar data show that the proposed method can effectively extract the features of ship target from sea clutter, further reduce the feature dimensionality, and improve target recognition performance.

Night Vehicle Detection Using Variable Haar-Like Feature

This paper proposes a night-time vehicle detection method using variable Haar-like feature.The specific features of front vehicle cannot be obtained in road image at night-time because of light reflection and ambient light,and it is also difficult to define optimal brightness and color of rear lamp according to road conditions.In comparison,the difference of vehicle region and road surface is more robust for road illumination environment.Thus,we select the candidates of vehicles by analysing the difference,and verify the candidates using those brightness and complexity to detect vehicle correctly.The feature of brightness difference is detected using variable horizontal Haar-like mask according to vehicle size in the location of image.And the region occurring rapid change is selected as the candidate.The proposed method is evaluated by testing on the various real road conditions.

Nonlinear industrial process fault diagnosis with latent label consistency and sparse Gaussian feature learning

With the increasing complexity of industrial processes, the high-dimensional industrial data exhibit a strong nonlinearity, bringing considerable challenges to the fault diagnosis of industrial processes. To efficiently extract deep meaningful features that are crucial for fault diagnosis, a sparse Gaussian feature extractor(SGFE) is designed to learn a nonlinear mapping that projects the raw data into the feature space with the fault label dimension. The feature space is described by the one-hot encoding of the fault category label as an orthogonal basis. In this way, the deep sparse Gaussian features related to fault categories can be gradually learned from the raw data by SGFE. In the feature space,the sparse Gaussian(SG) loss function is designed to constrain the distribution of features to multiple sparse multivariate Gaussian distributions. The sparse Gaussian features are linearly separable in the feature space, which is conducive to improving the accuracy of the downstream fault classification task. The feasibility and practical utility of the proposed SGFE are verified by the handwritten digits MNIST benchmark and Tennessee-Eastman(TE) benchmark process,respectively.

Rare Bird Sparse Recognition via Part-Based Gist Feature Fusion and Regularized Intraclass Dictionary Learning

Rare bird has long been considered an important in the field of airport security,biological conservation,environmental monitoring,and so on.With the development and popularization of IOT-based video surveillance,all day and weather unattended bird monitoring becomes possible.However,the current mainstream bird recognition methods are mostly based on deep learning.These will be appropriate for big data applications,but the training sample size for rare bird is usually very short.Therefore,this paper presents a new sparse recognition model via improved part detection and our previous dictionary learning.There are two achievements in our work:(1)after the part localization with selective search,the gist feature of all bird image parts will be fused as data description;(2)the fused gist feature needs to be learned through our proposed intraclass dictionary learning with regularized K-singular value decomposition.According to above two innovations,the rare bird sparse recognition will be implemented by solving one l1-norm optimization.In the experiment with Caltech-UCSD Birds-200-2011 dataset,results show the proposed method can have better recognition performance than other SR methods for rare bird task with small sample size.

Underground Coal Mine Target Tracking via Multi-Feature Joint Sparse Representation

Single-feature methods are unable to effectively track a target in an underground coal mine video due to the high background noise, low and uneven illumination, and drastic light fluctuation in the video. In this study, we propose an underground coal mine personnel target tracking method using multi-feature joint sparse representation. First, with a particle filter framework, the global and local multiple features of the target template and candidate particles are extracted. Second, each of the candidate particles is sparsely represented by a dictionary template, and reconstruction is achieved after solving the sparse coefficient. Last, the particle with the lowest reconstruction error is deemed the tracking result. To validate the effectiveness of the proposed algorithm, we compare the proposed method with three commonly employed tracking algorithms. The results show that the proposed method is able to reliably track the target in various scenarios, such as occlusion and illumination change, which generates better tracking results and validates the feasibility and effectiveness of the proposed method.

Underwater Noise Target Recognition Based on Sparse Adversarial Co-Training Model with Vertical Line Array

The automatic identification of underwater noncooperative targets without label records remains an arduous task considering the marine noise interference and the shortage of labeled samples.In particular,the data-driven mechanism of deep learning cannot identify false samples,aggravating the difficulty in noncooperative underwater target recognition.A semi-supervised ensemble framework based on vertical line array fusion and the sparse adversarial co-training algorithm is proposed to identify noncooperative targets effectively.The sound field cross-correlation compression(SCC)feature is developed to reduce noise and computational redundancy.Starting from an incomplete dataset,a joint adversarial autoencoder is constructed to extract the sparse features with source depth sensitivity,aiming to discover the unknown underwater targets.The adversarial prediction label is converted to initialize the joint co-forest,whose evaluation function is optimized by introducing adaptive confidence.The experiments prove the strong denoising performance,low mean square error,and high separability of SCC features.Compared with several state-of-the-art approaches,the numerical results illustrate the superiorities of the proposed method due to feature compression,secondary recognition,and decision fusion.

A weighted block cooperative sparse representation algorithm based on visual saliency dictionary

Unconstrained face images are interfered by many factors such as illumination,posture,expression,occlusion,age,accessories and so on,resulting in the randomness of the noise pollution implied in the original samples.In order to improve the sample quality,a weighted block cooperative sparse representation algorithm is proposed based on visual saliency dictionary.First,the algorithm uses the biological visual attention mechanism to quickly and accurately obtain the face salient target and constructs the visual salient dictionary.Then,a block cooperation framework is presented to perform sparse coding for different local structures of human face,and the weighted regular term is introduced in the sparse representation process to enhance the identification of information hidden in the coding coefficients.Finally,by synthesising the sparse representation results of all visual salient block dictionaries,the global coding residual is obtained and the class label is given.The experimental results on four databases,that is,AR,extended Yale B,LFW and PubFig,indicate that the combination of visual saliency dictionary,block cooperative sparse representation and weighted constraint coding can effectively enhance the accuracy of sparse representation of the samples to be tested and improve the performance of unconstrained face recognition.

Adaptive multiscale wavelet-guided periodic sparse representation for bearing incipient fault feature extraction

Currently, accurately extracting early-stage bearing incipient fault features is urgent and challenging. This paper introduces a novel method called adaptive multiscale wavelet-guided periodic sparse representation(AMWPSR) to address this issue. For the first time, the dual-tree complex wavelet transform is applied to construct the linear transformation for the AMWPSR model.This transform offers superior shift invariance and minimizes spectrum aliasing. By integrating this linear transformation with the generalized minimax concave penalty term, a new sparse representation model is developed to recover faulty impulse components from heavily disturbed vibration signals. During each iteration of the AMWPSR process, the impulse periods of sparse signals are adaptively estimated, and the periodicity of the latest sparse signal is augmented using the final estimated period. Simulation studies demonstrate that AMWPSR can effectively estimate periodic impulses even in noisy environments, demonstrating greater accuracy and robustness in recovering faulty impulse components than existing techniques.Further validation through research on two sets of bearing life cycle data shows that AMWPSR delivers superior fault diagnosis results.

Sparseness-controlled non-negative tensor factorization and its application in machinery fault diagnosis

Aiming at the problems of bispectral analysis when applied to machinery fault diagnosis, a machinery fault feature extraction method based on sparseness-controlled non-negative tensor factorization （SNTF） is proposed. First, a non-negative tensor factorization（NTF） algorithm is improved by imposing sparseness constraints on it. Secondly, the bispectral images of mechanical signals are obtained and stacked to form a third-order tensor. Thirdly, the improved algorithm is used to extract features, which are represented by a series of basis images from this tensor. Finally, coefficients indicating these basis images＇ weights in constituting original bispectral images are calculated for fault classification. Experiments on fault diagnosis of gearboxes show that the extracted features can not only reveal some nonlinear characteristics of the system, but also have intuitive meanings with regard to fault characteristic frequencies. These features provide great convenience for the interpretation of the relationships between machinery faults and corresponding bispectra.

基于Haar-like特征的实时L1-跟踪算法

稀疏表示技术已成功应用于视觉跟踪,但是仍然存在跟踪算法效率低的问题。提出一种基于Haar-like特征的视频跟踪算法,该算法是基于粒子滤波框架的L1-跟踪算法,其特点是运用Haar-like特征及特征块的思想对完备基进行重新构造。将正负小模板由单个像素改为像素块,降低稀疏表示中过完备基的维数,大幅减少稀疏矩阵的计算量;同时,在保证跟踪质量的前提下适当减少目标模板数量,减少稀疏计算的次数,并控制模板更新频率。实验结果表明,所提算法能大幅提高跟踪的实时性,同时很好地解决了跟踪问题中的短时间遮挡、目标物体的形变以及光照变化等问题。

基于改进Sparse Indexing的多负载消冗方法

针对现有的Sparse Indexing方法不能有效处理小文件备份负载的问题,提出了一种以Broder扩展定理为理论依据的最小特征采样算法,该算法可以对不同形式的备份负载进行有效的特征采样。在此算法的基础上,设计了一种多负载重复数据消除方法,该方法通过对备份负载进行特征采样,仅在内存中维护完整索引的一个很小的子集,并通过批量读入分块标识符,摊销了磁盘访问开销,提高了吞吐量。实验结果表明,该方法对混合备份负载的压缩比是Sparse Indexing的2.04倍,而吞吐量与Sparse Indexing相当。该方法适用于需要处理多种形式备份负载的高性能重复数据消除系统。

基于Sparse ICP的三维点云耳廓识别

提出一种新颖的三维耳廓识别方法,首先基于PCA和SVD分解对三维耳廓点云模型进行归一化预处理,以统一数据库中所有耳廓点云模型的位置与姿态;然后基于Iannarelli分类系统提取三维耳廓的4个局部特征区域,并利用Sparse ICP算法对局部特征区域进行匹配;最后根据局部特征区域中对应点间的距离判断耳廓之间的差异测度,实现耳廓形状识别。实验证明,本文算法与其他算法相比具有较高的识别精度和识别效率。

Human Action Recognition Based on Supervised Class-Specific Dictionary Learning with Deep Convolutional Neural Network Features

Human action recognition under complex environment is a challenging work.Recently,sparse representation has achieved excellent results of dealing with human action recognition problem under different conditions.The main idea of sparse representation classification is to construct a general classification scheme where the training samples of each class can be considered as the dictionary to express the query class,and the minimal reconstruction error indicates its corresponding class.However,how to learn a discriminative dictionary is still a difficult work.In this work,we make two contributions.First,we build a new and robust human action recognition framework by combining one modified sparse classification model and deep convolutional neural network(CNN)features.Secondly,we construct a novel classification model which consists of the representation-constrained term and the coefficients incoherence term.Experimental results on benchmark datasets show that our modified model can obtain competitive results in comparison to other state-of-the-art models.

D-SS Frame:deep spectral-spatial feature extraction and fusion for classification of panchromatic and multispectral images

Facing the very high-resolution( VHR) image classification problem,a feature extraction and fusion framework is presented for VHR panchromatic and multispectral image classification based on deep learning techniques. The proposed approach combines spectral and spatial information based on the fusion of features extracted from panchromatic( PAN) and multispectral( MS) images using sparse autoencoder and its deep version. There are three steps in the proposed method,the first one is to extract spatial information of PAN image,and the second one is to describe spectral information of MS image. Finally,in the third step,the features obtained from PAN and MS images are concatenated directly as a simple fusion feature. The classification is performed using the support vector machine( SVM) and the experiments carried out on two datasets with very high spatial resolution. MS and PAN images from WorldView-2 satellite indicate that the classifier provides an efficient solution and demonstrate that the fusion of the features extracted by deep learning techniques from PAN and MS images performs better than that when these techniques are used separately. In addition,this framework shows that deep learning models can extract and fuse spatial and spectral information greatly,and have huge potential to achieve higher accuracy for classification of multispectral and panchromatic images.

Abnormal behavior detection by causality analysis and sparse reconstruction

A new approach for abnormal behavior detection was proposed using causality analysis and sparse reconstruction. To effectively represent multiple-object behavior, low level visual features and causality features were adopted. The low level visual features, which included trajectory shape descriptor, speeded up robust features and histograms of optical flow, were used to describe properties of individual behavior, and causality features obtained by causality analysis were introduced to depict the interaction information among a set of objects. In order to cope with feature noisy and uncertainty, a method for multiple-object anomaly detection was presented via a sparse reconstruction. The abnormality of the testing sample was decided by the sparse reconstruction cost from an atomically learned dictionary. Experiment results show the effectiveness of the proposed method in comparison with other state-of-the-art methods on the public databases for abnormal behavior detection.

Vehicle Representation and Classification of Surveillance Video Based on Sparse Learning

We cast vehicle recognition as problem of feature representation and classification, and introduce a sparse learning based framework for vehicle recognition and classification in this paper. After objects captured with a GMM background subtraction program, images are labeled with vehicle type for dictionary learning and decompose the images with sparse coding （SC）, a linear SVM trained with the SC feature for vehicle classification. A simple but efficient active learning stategy is adopted by adding the false positive samples into previous training set for dictionary and SVM model retraining. Compared with traditional feature representation and classification realized with SVM, SC method achieves dramatically improvement on classification accuracy and exhibits strong robustness. The work is also validated on real-world surveillance video.

A Scheme of s EMG Feature Extraction for Improving Myoelectric Pattern Recognition

This paper proposed a feature extraction scheme based on sparse representation considering the nonstationary property of surface electromyography( sEMG). Sparse Bayesian Learning( SBL) algorithm was introduced to extract the feature with optimal class separability to improve recognition accuracies of multimovement patterns. The SBL algorithm exploited the compressibility( or weak sparsity) of sEMG signal in some transformed domains. The proposed feature extracted by using the SBL algorithm was named SRC. The feature SRC represented time-varying characteristics of sEMG signal very effectively. We investigated the effect of the feature SRC by comparing with other fourteen individual features and eighteen multi-feature sets in offline recognition. The results demonstrated the feature SRC revealed the important dynamic information in the sEMG signals. And the multi-feature sets formed by the feature SRC and other single features yielded more superior performance on recognition accuracy. The best average recognition accuracy of 91. 67% was gained by using SVM classifier with the multi-feature set combining the feature SRC and the feature wavelength( WL). The proposed feature extraction scheme is promising for multi-movement recognition with high accuracy.