Low-dimensional Feature Vector Research Articles

A Passive Approach for Detecting Image Splicing Based on Deep Learning and Wavelet Transform

Splicing image forgery detection has become a significant research subject in multimedia forensics and security due to its widespread use and its hard detection. Many algorithms have already been executed on the image splicing. The existing algorithms may be affected by some problems, such as high feature dimensionality and low accuracy with high false positive rates. In this paper, an algorithm based on deep learning approach and wavelet transform is proposed to detect the spliced image. In the deep learning approach, convolutional neural network (CNN) is employed to automatically extract features from the spliced image. CNN is applied and then discrete wavelet transform (DWT) is used. Support vector machine is used later for classification. Additional experiments are performed. That is, discrete cosine transform replaces DWT and then principal component analysis is applied. The proposed algorithm is evaluated on a publicly available image splicing datasets (CASIA v1.0 and CASIA v2.0). It achieves high accuracy while using a relatively low-dimensional feature vector. Our results demonstrate that the proposed algorithm is effective and accomplishes better performance for detecting the spliced image.

Trembr

In this article, we propose a novel representation learning framework, namely TRajectory EMBedding via Road networks (Trembr) , to learn trajectory embeddings (low-dimensional feature vectors) for use in a variety of trajectory applications. The novelty of Trembr lies in (1) the design of a recurrent neural network--(RNN) based encoder--decoder model, namely Traj2Vec , that encodes spatial and temporal properties inherent in trajectories into trajectory embeddings by exploiting the underlying road networks to constrain the learning process in accordance with the matched road segments obtained using road network matching techniques (e.g., Barefoot [24, 27]), and (2) the design of a neural network--based model, namely Road2Vec , to learn road segment embeddings in road networks that captures various relationships amongst road segments in preparation for trajectory representation learning. In addition to model design, several unique technical issues raising in Trembr, including data preparation in Road2Vec, the road segment relevance-aware loss, and the network topology constraint in Traj2Vec, are examined. To validate our ideas, we learn trajectory embeddings using multiple large-scale real-world trajectory datasets and use them in three tasks, including trajectory similarity measure, travel time prediction, and destination prediction. Empirical results show that Trembr soundly outperforms the state-of-the-art trajectory representation learning models, trajectory2vec and t2vec , by at least one order of magnitude in terms of mean rank in trajectory similarity measure, 23.3% to 41.7% in terms of mean absolute error (MAE) in travel time prediction, and 39.6% to 52.4% in terms of MAE in destination prediction.

Human gait recognition based on histogram of oriented gradients and Haralick texture descriptor

Gait recognition is an evolving technology in the biometric domain; it aims to recognize people through an analysis of their walking pattern. One of the significant challenges of the appearance-based gait recognition system is to augment its performance by using a distinctive low-dimensional feature vector. Therefore, this study proposes the low-dimensional features that are capable of effectively capturing the spatial, gradient, and texture information in this context. These features are obtained by the computation of histogram of oriented gradients, followed by sum variance Haralick texture descriptor from nine cells of gait gradient magnitude image. Further, the performance of the proposed method is validated on five widely used gait databases. They include CASIA A gait database, CASIA B gait database, OU-ISIR D gait database, CMU MoBo database, and KTH video database. The experimental results demonstrated that the proposed approach could choose significant discriminatory features for individual identification and consequently, outperform certain state-of-the-art methods in terms of recognition performance.

Using high-dimensional features for high-accuracy pulse diagnosis.

Accurate pulse diagnosis is often based on extensive clinical experience. Recently, modern computer-aided pulse diagnostic methods have been developed to help doctors to quickly determine patients' physiological conditions. Most pulse diagnostic methods used low-dimensional feature vectors to classify pulse types. Therefore, some important but subtle pulse information might be ignored. In this study, a novel high-dimensional pulse classification method was developed to improve pulse diagnosis accuracy. To understand the underlying physical meaning or implications hidden in pulse discrimination, 71 pulse features were extracted from the time, spatial, and frequency domains to cover as much pulse information as possible. Then, Principal Component Analysis (PCA) was applied to extract the most representative components. Artificial neural networks were trained to classify 10 different pulse types. The results showed that PCA accounted for 95% of the total variances achieved the highest accuracy of 98.2% in pulse classification. The results also showed that pulse energy, local instantaneous characteristics, main frequency, and waveform complexity were the major factors determining pulse discriminability. This study demonstrated that using high-dimensional features could retain more pulse information and thus, effectively improve pulse diagnostic accuracy.

The Research on Face Recognition and Segmentation Based on Intelligent Background

Affected by factors such as attitude, light, expression, etc., it is impossible to accurately identify the identity in a wireless visual sensor network in an uncontrollable environment. In traditional visual identity recognition, it is necessary to convert uncontrollable factors into controllable and stable feature factors for identity recognition in a relatively uncontrollable environment where the node distribution is relatively complicated. The conversion process leads to long recognition time and low efficiency. An adaptive recognition method for identity features in wireless visual sensing networks based on LBP face recognition is proposed. A strong classifier is obtained for cascading, and the underlying features are extracted. The final Harr face cascade classifier is applied to the face Check it out. The PCA dimensionality reduction processing of the facial area feature vector is performed to obtain the low-dimensional feature vector, the dimensionality reduction coefficient, and the average face of the person. For the face image in the wireless local area, its LBP operation is given. Perform histogram statistics on face feature information, obtain face LBP histograms, and perform feature matching on the face feature database to complete recognition. The improved algorithm has improved the cumulative matching score of traditional algorithms by 17.8%; the accuracy rate has improved It is 32.7%, and the recognition time is shortened by 3.9s. Simulation results show that the proposed algorithm has high accuracy and recognition efficiency.

Breast Cancer Detection with Low-dimension Ordered Orthogonal Projection in Terahertz Imaging.

This paper proposes a new dimension reduction algorithm based on low-dimension ordered orthogonal projection (LOOP), which is used for cancer detection with terahertz (THz) images of freshly excised human breast cancer tissues. A THz image can be represented by a data cube with each pixel containing a high dimension spectrum vector covering several THz frequencies, where each frequency represents a different dimension in the vector. The proposed algorithm projects the high-dimension spectrum vector of each pixel within the THz image into a low-dimension subspace that contains the majority of the unique features embedded in the image. The low-dimension subspace is constructed by sequentially identifying its orthonormal basis vectors, such that each newly chosen basis vector represents the most unique information not contained by existing basis vectors. A multivariate Gaussian mixture model is used to represent the statistical distributions of the low-dimension feature vectors obtained from the proposed dimension reduction algorithm. The model parameters are iteratively learned by using unsupervised learning methods such as Markov chain Monte Carlo or expectation maximization, and the results are used to classify the various regions within a tumor sample. Experiment results demonstrate that the proposed method achieves apparent performance improvement in human breast cancer tissue over existing approaches such as one-dimension Markov chain Monte Carlo. The results confirm that the dimension reduction algorithm presented in this paper is a promising technique for breast cancer detection with THz images, and the classification results present a good correlation with respect to the histopathology results of the analyzed samples.

Multi-frame super-resolution reconstruction based on global motion estimation using a novel CNN descriptor

In this paper, we introduce a novel feature descriptor based on deep learning that trains a model to match the patches of images on scenes captured under different viewpoints and lighting conditions for Multi-frame super-resolution. The patch matching of images capturing the same scene in varied circumstances and diverse manners is challenging. We develop a model which maps the raw image patch to a low dimensional feature vector. As our experiments show, the proposed approach is much better than state-of-the-art descriptors and can be considered as a direct replacement of SURF. The results confirm that these techniques further improve the performance of the proposed descriptor. Then we propose an improved Random Sample Consensus algorithm for removing false matching points. Finally, we show that our neural network based image descriptor for image patch matching outperforms state-of-the-art methods on a number of benchmark datasets and can be used for image registration with high quality in multi-frame super-resolution reconstruction.

Research on image classification method of features of combinatorial convolution

In image classification, shallow convolutional features and deep convolutional features are not fully utilized by many network frameworks. To solve this problem, we propose a combinatorial convolutional network (CCNet) that integrates convolutional features of all levels. According to its own structure, the convolutional features of shallow, medium, and deep levels are extracted. These features are combined by weighted concatenation and convolutional fusion, and the coefficients of each channel of final combination feature are again weighted to improve the identification degree of features. CCNet can improve the single case where most network only add or concatenate shallow and deep features, so that the network can achieve lower classification error rate while generating low-dimensional features. Extensive experiments are performed on CIFAR-10 and CIFAR-100 respectively. The experimental results show that the low-dimensional image feature vectors generated by CCNet effectively reduce the classification error rate when the number of convolutional layers does not exceed 100 layers.

Low-Complexity Failed Element Diagnosis for Radar-Communication mmWave Antenna Array with Low SNR

The millimeter-wave (mmWave) antenna array plays an important role in the excellent performance of wireless sensors networks (WSN) or unmanned aerial vehicle (UAV) clusters. However, the array elements are easily damaged in its harsh working environment but hard to be repaired or exchanged timely, resulting in a serious decline in the beamforming performance. Thus, accurate self-diagnosis of the failed elements is of great importance. In previous studies, there are still significant difficulties for large-scale arrays under extremely low SNR. In this paper, a diagnosis algorithm with low complexity and high reliability for the failed elements is proposed, which is based on a joint decision of communication signal and sensing echoes. Compared with the previous studies, the complexity of the algorithm is reduced by the construction of low-dimensional feature vectors for classification, the decoupling of the degree of arrival (DOA) estimation and the failed pattern diagnosis, with the help of the sub-array division. Simulation results show that, under an ultra-low SNR of −12.5 dB for communication signals and −16 dB for sensing echoes, an accurate self-diagnosis with a block error rate lower than 8% can be realized. The study in this paper will effectively promote the long-term and reliable operation of the mmWave antenna array in WSN, UAV clusters and other similar fields.

Asynchronous Averaging of Gait Cycles for Classification of Gait and Device Modes

The problem of joint classification of gait and device mode from inertial measurement units (IMU) measurements is considered. For this, an approach for computing unique gait signature using measurements collected from body-worn inertial measurement units (IMUs) is proposed.The gait signature represents one full cycle of the human gait, and is suitable for off-line or on-line classification of the gait mode. The signature can also be used to jointly classify the gait mode and the device mode. The device mode identifies how the IMU-equipped device is being carried by the user. The method is based on precise segmentation and resampling of the measured IMU signal, as an initial step, further tuned by minimizing the variability of the obtained signature within each gait cycle. Finally, a Fourier series expansion of the gait signature is introduced which provides a low-dimensional feature vector well suited for classification purposes. The proposed method is evaluated on a large dataset involving several subjects, each one containing two different gait modes and four different device modes.

A hybrid regularized extreme learning machine for automated detection of pathological brain

This paper presents an automated method for detection of pathological brain using magnetic resonance (MR) images. The proposed method suggests to derive features using fast discrete curvelet transform. A combined feature reduction algorithm principal component analysis+linear discriminant analysis (PCA+LDA) is then applied to generate a low-dimensional and discriminant feature vector. Finally, the classification is carried out using a hybrid regularized extreme learning machine (RELM). The proposed hybrid classifier combines RELM and sine cosine algorithm to not only overcome the drawbacks of conventional learning algorithms but also provide good generalization performance with a compact and well-conditioned network. Besides root mean square error, the norm of the output weights and the condition value of the hidden layer output matrix have been separately taken into consideration to optimize the parameters of RELM. Extensive simulations on three benchmark datasets demonstrate that the proposed scheme obtains promising results as compared to state-of-the-art approaches. The effectiveness of the proposed hybrid classifier is compared with its counterparts. Moreover, the impact of noise in the training data is analyzed using the suggested scheme. The proposed approach can aid the radiologists for screening pathological brain at a larger scale.

Data-Driven Microstructure Property Relations

An image based prediction of the effective heat conductivity for highly heterogeneous microstructured materials is presented. The synthetic materials under consideration show different inclusion morphology, orientation, volume fraction and topology. The prediction of the effective property is made exclusively based on image data with the main emphasis being put on the 2-point spatial correlation function. This task is implemented using both unsupervised and supervised machine learning methods. First, a snapshot proper orthogonal decomposition (POD) is used to analyze big sets of random microstructures and, thereafter, to compress significant characteristics of the microstructure into a low-dimensional feature vector. In order to manage the related amount of data and computations, three different incremental snapshot POD methods are proposed. In the second step, the obtained feature vector is used to predict the effective material property by using feed forward neural networks. Numerical examples regarding the incremental basis identification and the prediction accuracy of the approach are presented. A Python code illustrating the application of the surrogate is freely available.

Automatic Land Cover Reconstruction From Historical Aerial Images: An Evaluation of Features Extraction and Classification Algorithms.

The land cover reconstruction from monochromatic historical aerial images is a challenging task that has recently attracted an increasing interest from the scientific community with the proliferation of large-scale epidemiological studies involving retrospective analysis of spatial patterns. However, the efforts made by the computer vision community in remote-sensing applications are mostly focused on prospective approaches through the analysis of high-resolution multi-spectral data acquired by the advanced spatial programs. Hence, four contributions are proposed in this paper. They aim at providing a comparison basis for the future development of computer vision algorithms applied to the automation of the land cover reconstruction from monochromatic historical aerial images. First, a new multi-scale multi-date dataset composed of 4.9 million non-overlapping annotated patches of the France territory between 1970 and 1990 has been created with the help of geography experts. This dataset has been named HistAerial. Second, an extensive comparison study of the state-of-the-art texture features extraction and classification algorithms, including deep convolutional neural networks (DCNNs), has been performed. It is presented in the form of an evaluation. Third, a novel low-dimensional local texture filter named rotated-corner local binary pattern (R-CRLBP) is presented as a simplification of the binary gradient contours filter through the use of an orthogonal combination representation. Finally, a novel combination of low-dimensional texture descriptors, including the R-CRLBP filter, is introduced as a light combination of local binary patterns (LCoLBPs). The LCoLBP filter achieved state-of-the-art results on the HistAerial dataset while conserving a relatively low-dimensional feature vector space compared with the DCNN approaches (17 times shorter).

New Texture Descriptor Based on Modified Fractional Entropy for Digital Image Splicing Forgery Detection.

Forgery in digital images is immensely affected by the improvement of image manipulation tools. Image forgery can be classified as image splicing or copy-move on the basis of the image manipulation type. Image splicing involves creating a new tampered image by merging the components of one or more images. Moreover, image splicing disrupts the content and causes abnormality in the features of a tampered image. Most of the proposed algorithms are incapable of accurately classifying high-dimension feature vectors. Thus, the current study focuses on improving the accuracy of image splicing detection with low-dimension feature vectors. This study also proposes an approximated Machado fractional entropy (AMFE) of the discrete wavelet transform (DWT) to effectively capture splicing artifacts inside an image. AMFE is used as a new fractional texture descriptor, while DWT is applied to decompose the input image into a number of sub-images with different frequency bands. The standard image dataset CASIA v2 was used to evaluate the proposed approach. Superior detection accuracy and positive and false positive rates were achieved compared with other state-of-the-art approaches with a low-dimension of feature vectors.

Deep Learning for Fast and Spatially Constrained Tissue Quantification From Highly Accelerated Data in Magnetic Resonance Fingerprinting.

Magnetic resonance fingerprinting (MRF) is a quantitative imaging technique that can simultaneously measure multiple important tissue properties of human body. Although MRF has demonstrated improved scan efficiency as compared to conventional techniques, further acceleration is still desired for translation into routine clinical practice. The purpose of this paper is to accelerate MRF acquisition by developing a new tissue quantification method for MRF that allows accurate quantification with fewer sampling data. Most of the existing approaches use the MRF signal evolution at each individual pixel to estimate tissue properties, without considering the spatial association among neighboring pixels. In this paper, we propose a spatially constrained quantification method that uses the signals at multiple neighboring pixels to better estimate tissue properties at the central pixel. Specifically, we design a unique two-step deep learning model that learns the mapping from the observed signals to the desired properties for tissue quantification, i.e.: 1) with a feature extraction module for reducing the dimension of signals by extracting a low-dimensional feature vector from the high-dimensional signal evolution and 2) a spatially constrained quantification module for exploiting the spatial information from the extracted feature maps to generate the final tissue property map. A corresponding two-step training strategy is developed for network training. The proposed method is tested on highly undersampled MRF data acquired from human brains. Experimental results demonstrate that our method can achieve accurate quantification for T1 and T2 relaxation times by using only 1/4 time points of the original sequence (i.e., four times of acceleration for MRF acquisition).

A novel binary feature descriptor to discriminate normal and abnormal chest CT images using dissimilarity measures

In this paper, a new feature descriptor local diagonal Laplacian pattern (LDLP) is proposed to separate the normal and emphysematous lesions in computed tomographic (CT) images containing pulmonary emphysema. LDLP employs a diagonal element approach in which the relationship of the centre pixel with the diagonal elements is obtained using the second-order derivatives (Laplacian). This results in a low-dimensional feature vector with richer information about the local structure. In the proposed framework, the feature histograms of chest CT image slices of EMPHYSEMA database are first determined. Further, the distance between the features of normal and emphysematous tissues is measured and analysed using ANOVA statistical method. Furthermore, a four-class classification is performed using an artificial neural network classifier. The classification performance of the proposed LDLP approach is compared with the prominent methods like local binary pattern, local tetra pattern, and local diagonal extrema pattern. The observational results show that the LDLP outperforms the existing binary feature descriptors in the segregation and classification of healthy and abnormal chest CT images.

A Hierarchical Attention Fused Descriptor for 3D Point Matching

Motivated by recent successes on learning 3D feature representations, we present a Siamese network to generate representative 3D descriptors for 3D point matching in point cloud registration. Our system, dubbed HAF-Net, consists of feature extraction module, hierarchical feature reweighting and recalibration module (HRR), as well as feature aggregation and compression module. The HRR module is proposed to adaptively integrate multi-level features through learning, acting as a hierarchical attention fusion mechanism. The learnable feature pooling technique VLAD is extended into our aggregation module, which is further utilized to extract principal components of features and compress them into a low dimensional feature vector. To train our model, we amass a large dataset for 3D point matching. The dataset is composed of matched and unmatched point block pairs, which are automatically searched from existing reconstruction datasets with known poses. The experiments demonstrate that the proposed HAF-Net not only outperforms other state-of-the-art approaches in 3D feature representation but also has a good generalization ability in various tasks and datasets.

Herb Target Prediction Based on Representation Learning of Symptom related Heterogeneous Network

Traditional Chinese Medicine (TCM) has received increasing attention as a complementary approach or alternative to modern medicine. However, experimental methods for identifying novel targets of TCM herbs heavily relied on the current available herb-compound-target relationships. In this work, we present an Herb-Target Interaction Network (HTINet) approach, a novel network integration pipeline for herb-target prediction mainly relying on the symptom related associations. HTINet focuses on capturing the low-dimensional feature vectors for both herbs and proteins by network embedding, which incorporate the topological properties of nodes across multi-layered heterogeneous network, and then performs supervised learning based on these low-dimensional feature representations. HTINet obtains performance improvement over a well-established random walk based herb-target prediction method. Furthermore, we have manually validated several predicted herb-target interactions from independent literatures. These results indicate that HTINet can be used to integrate heterogeneous information to predict novel herb-target interactions.

Impact of reduction in descriptor size on object detection and classification

Extraction of distinctive and robust features in image/video analysis and processing has attracted the attention of researchers in the recent years. Elimination of irrelevant and less important features reduces the computational complexity to a great extent at the cost of a very marginal reduction in accuracy. This paper presents a framework for dimensionality reduction of the binary features to obtain a low dimension feature vector for object detection. The process of identification and selection of the most relevant feature is performed in three steps: extraction of features using binary descriptors; Selection of best feature subset using Sequential Forward Selection (SFS) and Principal Component Analysis; classification using SVM classifier. The experimental results show that BRISK and LATCH descriptors perform better even when the dimensionality is reduced from 256, 128, and 64 to 32 bits with an acceptable classification accuracy and significant reduction in run time. However, there is slight decrease in the classification accuracy of LBP, FREAK, BRIEF, and ORB. A classification rate of 84.93% is obtained with LATCH descriptor for a descriptor size of 32 bits.

Deep Network Embedding for Graph Representation Learning in Signed Networks.

Network embedding has attracted an increasing attention over the past few years. As an effective approach to solve graph mining problems, network embedding aims to learn a low-dimensional feature vector representation for each node of a given network. The vast majority of existing network embedding algorithms, however, are only designed for unsigned networks, and the signed networks containing both positive and negative links, have pretty distinct properties from the unsigned counterpart. In this paper, we propose a deep network embedding model to learn the low-dimensional node vector representations with structural balance preservation for the signed networks. The model employs a semisupervised stacked auto-encoder to reconstruct the adjacency connections of a given signed network. As the adjacency connections are overwhelmingly positive in the real-world signed networks, we impose a larger penalty to make the auto-encoder focus more on reconstructing the scarce negative links than the abundant positive links. In addition, to preserve the structural balance property of signed networks, we design the pairwise constraints to make the positively connected nodes much closer than the negatively connected nodes in the embedding space. Based on the network representations learned by the proposed model, we conduct link sign prediction and community detection in signed networks. Extensive experimental results in real-world datasets demonstrate the superiority of the proposed model over the state-of-the-art network embedding algorithms for graph representation learning in signed networks.