Sliding Window Scheme Research Articles

Multiscale Feature-Clustering-Based Fully Convolutional Autoencoder for Fast Accurate Visual Inspection of Texture Surface Defects

Visual inspection of texture surface defects is still a challenging task in the industrial automation field due to the tremendous changes in the appearance of various surface textures. Current visual inspection methods cannot simultaneously and efficiently inspect various types of texture defects due to either the low discriminative capabilities of handcrafted features or their time-consuming sliding-window strategy. In this paper, we present a novel unsupervised multiscale feature-clustering-based fully convolutional autoencoder (MS-FCAE) method that efficiently and accurately inspects various types of texture defects based on a small number of defect-free texture samples. The proposed MS-FCAE method utilizes multiple FCAE subnetworks at different scale levels to reconstruct several textured background images. The residual images are obtained by subtracting these texture backgrounds from the input image individually; then, they are fused into one defect image. To maximize the efficiency, each FCAE subnetwork utilizes fully convolutional neural networks to extract the original feature maps directly from the input images. Meanwhile, each FCAE subnetwork performs feature clustering to improve the discriminant power of the encoded feature maps. The proposed MS-FCAE method is evaluated on several texture surface inspection data sets both qualitatively and quantitatively. This method achieves a Precision of 92.0% while requiring only 82 ms for input images of $1920\times 1080$ pixels. The extensive experimental results demonstrate that MS-FCAE achieves highly efficient and state-of-the-art inspection accuracy. Note to Practitioners-Most conventional visual inspection methods can address only one specific type of texture defect, while multiscale feature-clustering-based fully convolutional autoencoder (MS-FCAE) can simultaneously and accurately inspect various types of texture surface defects, such as those of thin-film transistor liquid crystal displays, wood, fabrics, and ceramic tiles. Furthermore, MS-FCAE requires only a small number of surface texture samples to learn a robust network model, and its training requires no defect samples. This is extremely important for industrial applications because identifying and labeling defect samples is difficult. Moreover, MS-FCAE can be applied to online visual inspection utilizing a graphics processing unit-based parallel processing strategy.

An application of backtracking search optimization–based least squares support vector machine for prediction of short-term wind speed

Wind speed prediction is an important technology in the wind power field; however, because of their chaotic nature, predicting wind speed accurately is difficult. Aims at this challenge, a backtracking search optimization–based least squares support vector machine model is proposed for short-term wind speed prediction. In this article, the least squares support vector machine is chosen as the short-term wind speed prediction model and backtracking search optimization algorithm is used to optimize the important parameters which influence the least squares support vector machine regression model. Furthermore, the optimal parameters of the model are obtained, and the short-term wind speed prediction model of least squares support vector machine is established through parameter optimization. For time-varying systems similar to short-term wind speed time series, a model updating method based on prediction error accuracy combined with sliding window strategy is proposed. When the prediction model does not match the actual short-term wind model, least squares support vector machine trains and re-establishes. This model updating method avoids the mismatch problem between prediction model and actual wind speed data. The actual collected short-term wind speed time series is used as the research object. Multi-step prediction simulation of short-term wind speed is carried out. The simulation results show that backtracking search optimization algorithm–based least squares support vector machine model has higher prediction accuracy and reliability for the short-term wind speed. At the same time, the prediction performance indicators are also improved. The prediction result is that root mean square error is 0.1248, mean absolute error is 0.1374, mean absolute percentile error is 0.1589% and R2 is 0.9648. When the short-term wind speed varies from 0 to 4 m/s, the average value of absolute prediction error is 0.1113 m/s, and average value of absolute relative prediction error is 8.7111%. The proposed prediction model in this article has high engineering application value.

Digital image forensics of non-uniform deblurring

Non-uniform image deblurring has been extensively studied, but forensics of whether an image is non-uniform deblurred is still an untouched area. In this paper, we firstly propose an approach to localize the non-uniform deblurring in digital images. Firstly, taking advantage of the properties of convolution derivation, multi-derivative gray level co-occurrence matrix (MGLCM) features are proposed to reveal the deblurring artifacts of images. The MGLCM features are extracted from the first and the second order derivative of images. Then sliding window strategy is used. For each sliding window, MGLCM features are extracted and SVMs are exploited to score the detection probability. By changing the scale of the sliding windows, a series of detection probability maps at different scales are obtained. Finally, top-down multi-scale boundary fusion (TMBF) is proposed to get the final detection map. The experimental results demonstrate that the proposed approach successfully localize the deblurred regions with a satisfactory performance.

Short Cationic Peptide Derived from Archaea with Dual Antibacterial Properties and Anti-Infective Potential.

Bacterial biofilms and associated infections represent one of the biggest challenges in the clinic, and as an alternative to counter bacterial infections, antimicrobial peptides have attracted great attention in the past decade. Here, ten short cationic antimicrobial peptides were generated through a sliding-window strategy on the basis of the 19-amino acid residue peptide, derived from a Pyrobaculum aerophilum ribosomal protein. PaDBS1R6F10 exhibited anti-infective potential as it decreased the bacterial burden in murine Pseudomonas aeruginosa cutaneous infections by more than 1000-fold. Adverse cytotoxic and hemolytic effects were not detected against mammalian cells. The peptide demonstrated structural plasticity in terms of its secondary structure in the different environments tested. PaDBS1R6F10 represents a promising antimicrobial agent against bacteria infections, without harming human cells.

Scene word recognition from pieces to whole

Convolutional neural networks (CNNs) have had great success with regard to the object classification problem. For character classification, we found that training and testing using accurately segmented character regions with CNNs resulted in higher accuracy than when roughly segmented regions were used. Therefore, we expect to extract complete character regions from scene images. Text in natural scene images has an obvious contrast with its attachments. Many methods attempt to extract characters through different segmentation techniques. However, for blurred, occluded, and complex background cases, those methods may result in adjoined or over segmented characters. In this paper, we propose a scene word recognition model that integrates words from small pieces to entire after-cluster-based segmentation. The segmented connected components are classified as four types: background, individual character proposals, adjoined characters, and stroke proposals. Individual character proposals are directly inputted to a CNN that is trained using accurately segmented character images. The sliding window strategy is applied to adjoined character regions. Stroke proposals are considered as fragments of entire characters whose locations are estimated by a stroke spatial distribution system. Then, the estimated characters from adjoined characters and stroke proposals are classified by a CNN that is trained on roughly segmented character images. Finally, a lexicon-driven integration method is performed to obtain the final word recognition results. Compared to other word recognition methods, our method achieves a comparable performance on Street View Text and the ICDAR 2003 and ICDAR 2013 benchmark databases. Moreover, our method can deal with recognizing text images of occlusion and improperly segmented text images.

Credit Card Fraud Detection using Machine Learning Algorithms

Abstract Credit card frauds are easy and friendly targets. E-commerce and many other online sites have increased the online payment modes, increasing the risk for online frauds. Increase in fraud rates, researchers started using different machine learning methods to detect and analyse frauds in online transactions. The main aim of the paper is to design and develop a novel fraud detection method for Streaming Transaction Data, with an objective, to analyse the past transaction details of the customers and extract the behavioural patterns. Where cardholders are clustered into different groups based on their transaction amount. Then using sliding window strategy [1], to aggregate the transaction made by the cardholders from different groups so that the behavioural pattern of the groups can be extracted respectively. Later different classifiers [3],[5],[6],[8] are trained over the groups separately. And then the classifier with better rating score can be chosen to be one of the best methods to predict frauds. Thus, followed by a feedback mechanism to solve the problem of concept drift [1]. In this paper, we worked with European credit card fraud dataset.

Optimal Burn-in Strategy for High Reliable Products Using Convolutional Neural Network

Burn-in test is widely used to improve the product reliability from the customer's perspective by identifying and screening out defective individuals before they are marketed. For those high reliable products whose failures are caused by gradual degradation, burn-in test not only could pick out weak units, but also increases the degradation of normal units, and hence the test duration is regarded as one key factor in the test policy optimization. In this paper, a new burn-in framework is proposed, which combines a sliding window strategy with one-dimensional convolutional neural network, completes the off-line training for classification model, and then obtains the optimal burn-in time with a group-accuracy strategy. And an online optimization algorithm is constructed to reduce the burn-in time as much as possible without deteriorating the screening effect, thereby to reduce the unnecessary lifetime loss of normal units involved in the test. The effectiveness of the presented framework is validated by the experiment. Compared to conventional strategies based on degradation models, the proposed method has better performance and robustness.

Improving the security and quality of real-time multimedia transmission in cyber-physical-social systems

Real-time multimedia transmission from smart physical terminals usually meets the challenges of quality and security problems in cyber physical social systems, such as in the Internet of things systems, social network, and industrial Internet of things. The quality of real-time video transmission is often disturbed by the instability of cyber-physical-social systems, and the security of real-time video can be compromised if fake contents are embedded into the video during the transmission. According to the video coding information and the conditions of cyber-physical-social systems like mobile network, we design a sliding window scheme to solve the out-of-order and loss problems and thus improve the quality of real-time video. In the scheme, the key video frame can be retransmitted in a dynamic range of sliding window. The scheme also simplifies the process of packet sorting and can quickly complete the process of unpacking and playing of the video stream with high accuracy. To defend the attack of inserting fake video into real-time video, we also design a detection algorithm based on time-related token. The experimental results show that the proposed methods can guarantee the continuity of video stream, improve the quality of image in the mobile network and ensure the security during transmission effectively.

A tensor framework for geosensor data forecasting of significant societal events

Geosensor data forecasting has high practical value in government affairs such as prompt response and decision making. However, the spatial correlation across distinct sites and the temporal correlation within each site pose challenges to accurate forecasting. In this paper, a geosensor data forecasting tensor framework for significant societal events is proposed. Specifically, a tensor pattern is used to model the geosensor data, based on which a tensor decomposition algorithm is then developed to estimate future values of geosensor data. The proposed approach not only combines and utilizes the multi-mode correlations, but also well extracts the underlying factors in each mode of tensor and mines the multi-dimensional structures of geosensor data. In addition, a rank increasing strategy is used to determine tensor rank automatically, and a sliding window strategy is used to improve the prediction accuracy. Extensive experimental evaluations illustrate the superiority of our approach compared with the state-of-the-arts.

A Hardware Architecture for Cell-Based Feature-Extraction and Classification Using Dual-Feature Space

Many computer-vision and machine-learning applications in robotics, mobile, wearable devices, and automotive domains are constrained by their real-time performance requirements. This paper reports a dual-feature-based object recognition coprocessor that exploits both histogram of oriented gradient (HOG) and Haar-like descriptors with a cell-based parallel sliding-window recognition mechanism. The feature extraction circuitry for HOG and Haar-like descriptors is implemented by a pixel-based pipelined architecture, which synchronizes to the pixel frequency from the image sensor. After extracting each cell feature vector, a cell-based sliding window scheme enables parallelized recognition for all windows, which contain this cell. The nearest neighbor search classifier is, respectively, applied to the HOG and Haar-like feature space. The complementary aspects of the two feature domains enable a hardware-friendly implementation of the binary classification for pedestrian detection with improved accuracy. A proof-of-concept prototype chip fabricated in a 65-nm SOI CMOS, having thin gate oxide and buried oxide layers (SOTB CMOS), with 3.22-mm2 core area achieves an energy efficiency of 1.52 nJ/pixel and a processing speed of 30 fps for $1024\times 1616$ -pixel image frames at 200-MHz recognition working frequency and 1-V supply voltage. Furthermore, multiple chips can implement image scaling, since the designed chip has image-size flexibility attributable to the pixel-based architecture.

A Hybrid Deep Learning Model for Predicting Protein Hydroxylation Sites.

Protein hydroxylation is one type of post-translational modifications (PTMs) playing critical roles in human diseases. It is known that protein sequence contains many uncharacterized residues of proline and lysine. The question that needs to be answered is: which residue can be hydroxylated, and which one cannot. The answer will not only help understand the mechanism of hydroxylation but can also benefit the development of new drugs. In this paper, we proposed a novel approach for predicting hydroxylation using a hybrid deep learning model integrating the convolutional neural network (CNN) and long short-term memory network (LSTM). We employed a pseudo amino acid composition (PseAAC) method to construct valid benchmark datasets based on a sliding window strategy and used the position-specific scoring matrix (PSSM) to represent samples as inputs to the deep learning model. In addition, we compared our method with popular predictors including CNN, iHyd-PseAAC, and iHyd-PseCp. The results for 5-fold cross-validations all demonstrated that our method significantly outperforms the other methods in prediction accuracy.

An Adaptive Sliding-Window Strategy for Outlier Detection in Wireless Sensor Networks for Smart Port Construction

ABSTRACT Yao, H.; Yang, Y.; Fu, X., and Mi, C., 2018. An adaptive sliding-window strategy for outlier detection in wireless sensor networks for smart port construction. In: Ashraf, M.A. and Chowdhu...

PCA and deep learning based myoelectric grasping control of a prosthetic hand

BackgroundFor the functional control of prosthetic hand, it is insufficient to obtain only the motion pattern information. As far as practicality is concerned, the control of the prosthetic hand force is indispensable. The application value of prosthetic hand will be greatly improved if the stable grip of prosthetic hand can be achieved. To address this problem, in this study, a bio-signal control method for grasping control of a prosthetic hand is proposed to improve patient’s sense of using prosthetic hand and the thus improving the quality of life.MethodsA MYO gesture control armband is used to collect the surface electromyographic (sEMG) signals from the upper limb. The overlapping sliding window scheme are applied for data segmentation and the correlated features are extracted from each segmented data. Principal component analysis (PCA) methods are then deployed for dimension reduction. Deep neural network is used to generate sEMG-force regression model for force prediction at different levels. The predicted force values are input to a fuzzy controller for the grasping control of a prosthetic hand. A vibration feedback device is used to feed grasping force value back to patient’s arm to improve patient’s sense of using prosthetic hand and realize accurate grasping. To test the effectiveness of the scheme, 15 able-bodied subjects participated in the experiments.ResultsThe classification results indicated that 8-channel sEMG applying all four time-domain features, with PCA reduction from 32 to 8 dimensions results in the highest classification accuracy. Based on the experimental results from 15 participants, the average recognition rate is over 95%. On the other hand, from the statistical results of standard deviation, the between-subject variations ranges from 3.58 to 1.25%, proving that the robustness and stability of the proposed approach.ConclusionsThe method proposed hereto control grasping power through the patient’s own sEMG signal, which achieves a high recognition rate to improve the success rate of grip and increases the sense of operation and also brings the gospel for upper extremity amputation patients.

Multi‐view face detection using frontal face detector

The development of frontal face detection has matured for application to everyday life. However, various profile face views are still difficult to deal with practically. A common solution is to build different detectors organized by a decision tree and ensure that each detector handles a single view or a few views. However, collecting the training images for each face view is time consuming and laborious. Moreover, many profile face detectors cannot perform as well as the frontal face detector because of insufficient training images. The management of many detectors also leads to too complex a logic structure to classify a face into its corresponding view. In this paper,we propose a novel method to reuse a frontal face detector to detect multi‐view faces, which do not need any data collection or training processes for various profile views. We focus on exploiting the potential of a general frontal face detector and aim to extend its application to multi‐view faces. We conduct a theoretical analysis to explain why our methodology works from different perspectives, and implement the proposed method based on the sliding window strategy. Furthermore, the searching process is optimized by a genetic algorithm with an original fitness function. Experimental results verify that our method can successfully detect human faces in almost all head poses in the dataset containing a complete collection of head poses in yaw and pitch axes. © 2018 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Improving the Separability of Deep Features with Discriminative Convolution Filters for RSI Classification

The extraction of activation vectors (or deep features) from the fully connected layers of a convolutional neural network (CNN) model is widely used for remote sensing image (RSI) representation. In this study, we propose to learn discriminative convolution filter (DCF) based on class-specific separability criteria for linear transformation of deep features. In particular, two types of pretrained CNN called CaffeNet and VGG-VD16 are introduced to illustrate the generality of the proposed DCF. The activation vectors extracted from the fully connected layers of a CNN are rearranged into the form of an image matrix, from which a spatial arrangement of local patches is extracted using sliding window strategy. DCF learning is then performed on each local patch individually to obtain the corresponding discriminative convolution kernel through generalized eigenvalue decomposition. The proposed DCF learning characterizes that a convolutional kernel with small size (e.g., 3 × 3 pixels) can be effectively learned on a small-size local patch (e.g., 8 × 8 pixels), thereby ensuring that the linear transformation of deep features can maintain low computational complexity. Experiments on two RSI datasets demonstrate the effectiveness of DCF in improving the classification performances of deep features without increasing dimensionality.

Simultaneous Estimation of Low- and High-Order Functional Connectivity for Identifying Mild Cognitive Impairment.

Functional connectivity (FC) network has been becoming an increasingly useful tool for understanding the cerebral working mechanism and mining sensitive biomarkers for neural/mental disease diagnosis. Currently, Pearson's Correlation (PC) is the simplest and most commonly used scheme in FC estimation. Despite its empirical effectiveness, PC only encodes the low-order (i.e., second-order) statistics by calculating the pairwise correlations between network nodes (brain regions), which fails to capture the high-order information involved in FC (e.g., the correlations among different edges in a network). To address this issue, we propose a novel FC estimation method based on Matrix Variate Normal Distribution (MVND), which can capture both low- and high-order correlations simultaneously with a clear mathematical interpretability. Specifically, we first generate a set of BOLD subseries by the sliding window scheme, and for each subseries we construct a temporal FC network by PC. Then, we employ the constructed FC networks as samples to estimate the final low- and high-order FC networks by maximizing the likelihood of MVND. To illustrate the effectiveness of the proposed method, we conduct experiments to identify subjects with Mild Cognitive Impairment (MCI) from Normal Controls (NCs). Experimental results show that the fusion of low- and high-order FCs can generally help to improve the final classification performance, even though the high-order FC may contain less discriminative information than its low-order counterpart. Importantly, the proposed method for simultaneous estimation of low- and high-order FCs can achieve better classification performance than the two baseline methods, i.e., the original PC method and a recent high-order FC estimation method.

Egocentric Temporal Action Proposals.

We present an approach to localize generic actions in egocentric videos, called temporal action proposals (TAPs), for accelerating the action recognition step. An egocentric TAP refers to a sequence of frames that may contain a generic action performed by the wearer of a head-mounted camera, e.g., taking a knife, spreading jam, pouring milk, or cutting carrots. Inspired by object proposals, this paper aims at generating a small number of TAPs, thereby replacing the popular sliding window strategy, for localizing all action events in the input video. To this end, we first propose to temporally segment the input video into action atoms, which are the smallest units that may contain an action. We then apply a hierarchical clustering algorithm with several egocentric cues to generate TAPs. Finally, we propose two actionness networks to score the likelihood of each TAP containing an action. The top ranked candidates are returned as output TAPs. Experimental results show that the proposed TAP detection framework performs significantly better than relevant approaches for egocentric action detection.

Pedestrian Detection for Android Mobile Devices

With the popularity of Android mobile devices and rapid development of hardware performance, more computer vision algorithms can be implemented on Android mobile devices, such as object detection, face recognition, etc. This paper focuses on the problem of implementing pedestrian detection on Android mobile devices. Recently, the commonly used pedestrian detection algorithm is based on sliding window strategy which uses single or multi-features, such as HOG and Channel features. Due to the resource limitation of mobile devices, in this paper, we use HOG and LBP joint feature to achieve pedestrian detection. In order to improve the efficiency of the algorithm, we use the spatial pooling algorithm to process the HOG-LBP joint feature. Experiments on the Android mobile phone shows that this method not only has competitive accuracy but also improves the pedestrian detection efficiency on both INRIA dataset and mobile phone images.

A Spatial Filtering Based Gridless DOA Estimation Method for Coherent Sources

In this paper, we investigate a covariance matrix reconstruction approach (CMRA) for direction-of-arrival (DOA) estimation in correlated/coherent sources scenario. We incorporate a spatial filtering (SF) model into our recently developed method CMRA in order to enhance its adaptation ability. In particular, a sliding window scheme is proposed to estimate the number of sources, and an iterative procedure is provided to estimate the DOAs of the signals. Since the original CMRA provides inaccurate estimate of the noise power which is undesirable during iterations, a new update rule for the noise power is proposed. Moreover, we derive a fast implementation of the SF-CMRA to accelerate the DOA estimation in each iteration. The proposed methods are suitable for both uniform and sparse linear arrays and are able to provide accurate estimates of DOAs, signal powers, and noise power. We also show that the proposed algorithmic framework can be easily extended to other gridless DOA estimation methods for accuracy improvement. Simulation results are provided to illustrate the superiority of our proposed methods.

An Accurate Framework for Arbitrary View Pedestrian Detection in Images

We consider the problem of detect pedestrian under from images collected under various viewpoints. This paper utilizes a novel framework called locality-constrained affine subspace coding (LASC). Firstly, the positive training samples are clustered into similar entities which represent similar viewpoint. Then Principal Component Analysis (PCA) is used to obtain the shared feature of each viewpoint. Finally, the samples that can be reconstructed by linear approximation using their top- k nearest shared feature with a small error are regarded as a correct detection. No negative samples are required for our method. Histograms of orientated gradient (HOG) features are used as the feature descriptors, and the sliding window scheme is adopted to detect humans in images. The proposed method exploits the sparse property of intrinsic information and the correlations among the multiple-views samples. Experimental results on the INRIA and SDL human datasets show that the proposed method achieves a higher performance than the state-of-the-art methods in form of effect and efficiency.