Search

The recent work Unsupervised Linear Discriminant Analysis (Un-LDA) completes its clustering process during the alternating optimization by converting equivalently the objective and finally using the K-means algorithm. However, the K-means algorithm has its inherent drawbacks. It is hard for the K-means algorithm to deal well with some complex clustering cases where there are too many real clusters or non-convex clusters. In this paper, a novel clustering optimization method is presented to accomplish the clustering process in Un-LDA and the resulting method can be named Un-LDA(CD). Specifically, instead of the K-means algorithm, an elaborately designed coordinate descent algorithm is adopted to obtain the clusters after the objective function goes through a series of simple but deft equivalent conversions. Extensive experiments have demonstrated that the coordinate descent clustering solution for Un-LDA can outperform the original K-means based solution on the tested data sets especially those complex data sets with a pretty large number of real clusters.

Face parsing aims to assign pixel-wise semantic labels to different facial components (e.g., hair, brows, and lips) in given face images. However, directly predicting pixel-level labels for each facial component over the whole face image would obtain limited accuracy, especially for tiny facial components. To address this problem, some recent works propose to first crop tiny patches from the whole face image and then predict masks for each facial component. However, such cropping-and-segmenting strategy consists of two independent stages, which cannot be jointly optimized. Besides, as one valuable piece of information for parsing the highly structured facial components, context cues are not elaborately explored by the existing works. To address these issues, we propose a component-level refinement network (CLRNet) for precisely segmenting out each facial component. Specifically, we introduce an attention mechanism to bridge the two independent stages together and form an end-to-end trainable pipeline for face parsing. Furthermore, we incorporate the global context information into the refining process for each cropped facial component patch, providing informative cues for accurate parsing. Extensive experiments are carried out on two benchmark datasets, LFW-PL and HELEN. The results demonstrate the superiority of the proposed CLRNet over other state-of-the-art methods, especially for tiny facial components.

The unconventional machining processes are popular in machining the difficult to cut type complex contour/ profile. The key parameters are to be machined with care as they directly affects the function and non key dimensions are machined with other than consideration of functional aspect, like design, passage. That means accuracy is less important in those dimensions but finish is to be good. So the machining of engineering materials does not always require optimizing parameters based on multiple responses. This investigation focuses on the sensitiveness of Wire EDM of 12X18H10T - Stainless Steel and identifying the parameters contribution and establish mathematical model to predict values without trial. An austenitic alloy 12X18H10T - Stainless Steel exhibits its inherent mechanical properties even when it expose to extremely high temperature. Ithas potential application in aerospace engine components manufacturing. This investigation utilized the Taguchi L9 experiential Design with factors of Pulse off Time, Wire Tension, Pulse on Time, and Wire Feed Rate. This study aims to examine Maximize the rate of machining with Taguchi approach in the process of Wire EDM of 12X18H10T - Stainless Steel and establish a mathematical model to predict the highly influencing parameter(s) for desired rate of machining.

The recognition a material quality is considered as a process of finding out the constituent material present in an object and it is regarded as a vital part in various applications. Hence, it is considered as a valuable approach for the creation of a system that possesses the ability to achieve recognition of a material. In this paper, we develop a mechanism using convolutional neural networks (CNNs) for material recognition. The CNN model initially trains itself with the features extracted from the image samples. Finally, the classification is carried out with CNN model that learn the classes obtained via CNN of different category of materials. The experimental validation is conducted to test the accuracy of CNN classifiers against various deep learning classifiers. The results on various materials show that the proposed CNN classifier obtains improved recognition accuracy than other methods.

Ship-hull design is a complex process, because the small local alteration in ship hull structure may significantly change the hydrostatic and hydrodynamic performances of a ship. To find the optimum hull shape under the design requirements, the state-of-art of ship hull design combines computational fluid dynamics computation with the geometric modeling. However, this process is very computationally intensive, which is only suitable at the final stage of the design process. To narrow down the design parameter space, in this work we have developed an AI-based deep learning neural network to realize a real-time prediction of the total resistance of the ship-hull structure in its initial design process. In this work, we have demonstrated how to use the developed DNN model to carry out the initial ship hull design. The validation results showed that the deep learning model can accurately predict the ship hull’s total resistance accurately after being trained, where the average error of all samples in the testing dataset is lower than 4%. Simultaneously, the trained deep learning model can provide the information about a ship’s performances in real-time by inputting geometric modification parameters without tedious preprocessing and calculation processes. The machine learning approach in ship hull design proposed in this work is the first step towards the artificial intelligence-aided design in naval architectures.

Selective laser melting (SLM), an additive manufacturing technology, is expected to replace the traditional lost-wax casting process used in producing removable partial denture (RPD) frameworks. However, studies comparing the accuracy of RPD frameworks and the effects of process parameters are lacking.The purpose of this in vitro study was to optimize SLM process parameters and use a quantitative analysis method to improve the accuracy of 3D-printed RPD frameworks.The orientation and support structure of Kennedy Class II RPDs were designed in various ways by using 2 different software programs, CAMbridge and Magics. The optimum melt-pool parameters, including laser power, scan speed, hatch distance, and layer thickness, were determined empirically before manufacturing 12 RPD frameworks with 4 different process designs by using SLM (n=3). The accuracy of the RPD frameworks was determined by 3D scanning and comparing the 3D scan data with the original standard tessellation language (STL) RPD design with the best-fit algorithm of the Geomagic software program.Optimum melt-pool parameters were found with the function of density, surface roughness, and productivity (P=180 W, v=1200 mm/s, h=60 μm, t=30 μm). RPD frameworks fabricated by the optimized process parameters (167 ±105 μm) showed significantly better (P.05) mean ±standard deviation accuracy than the 3 other groups of RPD frameworks manufactured by using the nonoptimized process parameters (180 ±121 μm to 222 ±136 μm). The best accuracy was found with the transverse orientation and interconnected support structure.With the optimized design of process parameters, clinically acceptable RPD frameworks were produced. The accuracy of RPD frameworks fabricated by using SLM varied according to the design of the process parameters, indicating that SLM technology can replace the traditional lost-wax casting process.

The article studies Automatic Oil Spraying Machine design and manufacture for baking device. In a current baking process, the workers manually spray the oil onto the trays. The spraying cycle increases the cost of man power in small scale industries. An automated oil spraying system is designed to solve these issues, where they can automatically spray the oil on the baking tray easily and effectively. The main components of this device are the mechanical and electrical components. The mechanical portion consists of conveyor belt, roller, bearing, gear, nozzle and tank, and mechanical structure. The electrical components include the IR sensor module, the dc motor, the dc pump, Arduino, and the source of the batteries. The key advantages of this approach are high precision oil spraying; improve efficiency, cost reduction and manpower. This model system's gross weight contains 15 kg (roughly). The efficiency of this device is checked for product production under different working conditions to track the performance of the work and the time durations.

Density-Based clustering are the main clustering algorithms because they can cluster data with different shapes and densities, but some of these algorithms have high time complexity like OPTICS (Ordering Points to Identify Clustering Structure) and DBSCAN (Density-based spatial clustering of applications with noise) where the time complexity up to O( n 2 ). In this paper, we use an approach to reduce this complexity by providing fuzzy clusters to OPTICS, which make the process of finding neighbours within an only narrow region (fuzzy group) instead of searching all the state space making the algorithm faster than the original one with keeping almost the same accuracy of the innovative algorithm. The results show that there is an improvement in execution time using some synthetic and real datasets with high dimensions.

Now days, the human being are affecting by so many diseases in all over the world. If we will consider about the tumor in the encephalon, then detection of tumor often involves radiological imaging. Radiological Imaging is used to check the spread of tumor and progress of treatment and also used to monitor cancer. The manual detection of the Encephalon tumor becomes a meticulous and chaotic task for the clinical experts. If the doctors or various departmental specialists want to eliminate the spoiled cancer areas from Magnetic Resonance Images, then it becomes very hectic and takes more time. The main aim of this work is to detect the encephalon tumor regions. Once the brain is going to be scan then it offers proficient and quick identification of encephalon tumor. To modernize the performance and abbreviate the intricacy involved in the image detection process in the proposed system as well as the Fuzzy CMean predicated image segmentation processes are used. Furthermore, to expand the variance of different images, particularly when the data are worn in image is signified beside different variance of images and then another technique has been implemented in the paper that is Histogram Equalization. The proposed method is completely automatic and tested on different kinds of brain MR images and proved as robust.

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

SYN Flood Attack is one form of distributed denial of service attack that attains the handshake process of TCP. This attack consumes all available server resources and provokes legitimate traffic which aims to make the server unavailable. It causes serious damage to cloud server and networking protocols. The main objective of this research work is to train the neural network for detecting the attack and to secure network connection. A novel binary fruit fly optimization algorithm with deep learning is proposed to predict the syn flood attack. The proposed algorithm is implemented using the KDD cup dataset. DL- BFFA algorithm has achieved 99.96% detection accuracy for detecting the SYN Flood Attack. A comparison study is conducted to validate the proposed model.

This article presents an intelligent fault diagnosis method for wind turbine (WT) gearbox by using wavelet packet decomposition (WPD) and deep learning. Specifically, the vibration signals from the gearbox are decomposed using WPD and the decomposed signal components are fed into a hierarchical convolutional neural network (CNN) to extract multiscale features adaptively and classify faults effectively. The presented method combines the multiscale characteristic of WPD with the strong classification capacity of CNNs, and it does not need complex manual feature extraction steps as usually adopted in existing results. The presented CNN with multiple characteristic scales based on WPD (WPD-MSCNN) has three advantages: 1) the added WPD layer can legitimately process the nonstationary vibration data to obtain components at multiple characteristic scales adaptively, it takes full advantage of WPD and, thus, enables the CNN to extract multiscale features; 2) the WPD layer directly sends multiscale components to the hierarchical CNN to extract rich fault information effectively, and it avoids the loss of useful information due to hand-crafted feature extraction; and 3) even if the scale changes, the lengths of components remain the same, which shows that the proposed method is robust to scale uncertainties in the vibration signals. Experiments with vibration data from a production wind farm provided by a company using condition monitoring system (CMS) show that the presented WPD-MSCNN method is superior to traditional CNN and multiscale CNN (MSCNN) for fault diagnosis.

The Micro-expressions (MEs) are very important nonverbal communication clues. However, due to their local and short nature, spotting them is challenging. In this paper, we address this problem by using a dedicated local and temporal pattern (LTP) of facial movement. This pattern has a specific shape (S-pattern) when ME are displayed. Thus, by using a classical classification algorithm (SVM), MEs are distinguished from other facial movements. We also propose a global final fusion analysis on the whole face to improve the distinction between ME (local) and head (global) movements. However, the learning of S-patterns is limited by the small number of ME databases and the low volume of ME samples. Hammerstein models (HMs) are known to be a good approximation of muscle movements. By approximating each S-pattern with a HM, we can both filter outliers and generate new similar S-patterns. By this way, we perform a data augmentation for S-pattern training dataset and improve the ability to differentiate micro-expressions from other facial movements. The spotting results, performed on CASME and CASME, show that our proposed LTP outperforms the most popular spotting method in terms of F1-score. Adding the fusion process and data augmentation improve even more the spotting performance.

In the process of digital image acquisition, color inconsistency is a common issue due to the influence of photographic conditions. The inconsistency of color affects the quality of digital orthophoto products and subsequent applications significantly, and this issue remains challenging both in the academia and industry. This paper proposes a color equalization method for aerial image based on block adjustment and an improved color balance model based on the analysis of causes for color deviation. Two nonlinear functions are used to simulate the process of color deviation and correct the color of images. Due to the existence of error, the color corrected images still have minor deviations in the color of the overlapped regions. Therefore, this research introduces an error equation to address this issue and uses the least square principle to calculate the adjustment. The simulation experiments and real experiments are carried out on several groups of aerial image data respectively. The results show that the method not only achieves better visual effect but also has a higher adjustment precision. Finally, the rationality and parameters of the model proposed in this paper are discussed.

Most existing light field saliency detection methods have achieved great success by exploiting unique light field data-focus information in focal slices. However, they process light field data in a slicewise way, leading to suboptimal results because the relative contribution of different regions in focal slices is ignored. How we can comprehensively explore and integrate focused saliency regions that would positively contribute to accurate saliency detection. Answering this question inspires us to develop a new insight. In this article, we propose a patch-aware network to explore light field data in a regionwise way. First, we excavate focused salient regions with a proposed multisource learning module (MSLM), which generates a filtering strategy for integration followed by three guidances based on saliency, boundary, and position. Second, we design a sharpness recognition module (SRM) to refine and update this strategy and perform feature integration. With our proposed MSLM and SRM, we can obtain more accurate and complete saliency maps. Comprehensive experiments on three benchmark datasets prove that our proposed method achieves competitive performance over 2-D, 3-D, and 4-D salient object detection methods. The code and results of our method are available at https://github.com/OIPLab-DUT/IEEE-TCYB-PANet.

There are some factors that play a crucial role in deciding the image quality during the optical as well as the micrographic association with the technique of the shadow casting in which contrast of the image has influenced with obliquely depositing of a thin layer of the metal during the process of the microscopic preparations. Further castingadvantages of the using shadow technique are to be observed during the process if shadow casting itself. Shadow casting posses’ number of the advantages over the other casting methods. The most prominent advantage is that if the height and thickness of an object is not completely flat that can be easily notice by the shadow of that object. The lengths of the shadow of the uneven part are so differentiate from other part that can be easily determined. Shadow casting uses the scattering power or even the optical efficiency to bring out the specimens contours and will easily get notice in order to check the flatness of the surfaces.

The growth of transferring data via networks and the internet yield to safe them through communication. The image encryption process is a valuable strategy for protecting image information and it is used as an effective way of communicating several sensitive data with continuous use of images in all forms of visual media. In this paper, an efficient method of image encryption is proposed based on PRESENT-SPECK algorithms with a hybrid method for key generation called Henon and Arnold-cat map that is suggested for sequences. The proposed method petitioned the image data into the block for encryption. The total blocks joined and produced an encrypted image. The proposed method tested with various types, and the generated key also have passed various tests.

Emotional expressivity has always been a simple job for people, but computer programming is much harder to accomplish. Image emotions may be recognised by recent developments in computer vision and machine learning. In this article, we present a new method to detect face emotion. Use neural networks convolutionary (FERC). The FERC is based on a CNN network of two parts: the first portion removed the backdrop of the image, the second part removed the face vector. The expressional vector (EV) is utilised in the FERC model to detect the fve different kinds of regular facial expressions. The double-level CNN is continuous and the weights and exponent values of the final perception layer vary with each iteration. In that it improves accuracy, FERC varies from widely utilised CNN single-level technology. Moreover, EV generation prevents the development of a number of issues before a new background removal process is used (for example distance from the camera).

Deep learning is the extension of machine learning technique which attracted the researchers in the recent past and hence the literature is also limited. This research focused on the need for deep learning for effective prediction process, when compared with few other machine learning models. As a proof of concept, a dataset of students’ performance of a school is analysed by various machine learning models. The performance measurements by all such models are compared with deep learning. These are assessed by two familiar statistical measures, Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). The error rates are low for deep learning when compared with other models. It is been evident that the result obtained by the deep learning model is more effective than the other machine learning models. The results are improved by reducing the error rates at the rates of 35% by mean absolute error and 48% by root mean squared error.

MotivationUsing DNA as the medium to store information has recently been recognized as a promising solution for long-term data storage. While several system prototypes have been demonstrated, the error characteristics in DNA data storage are discussed with limited content. Due to the data and process variations from experiment to experiment, the error variation and its effect on data recovery remain to be uncovered. To close the gap, we systematically investigate the storage channel, i.e., error characteristics in the storage process.ResultsWe first propose a new concept named sequence corruption to unify the error characteristics into the sequence level, easing the channel analysis. Then we derived the formulations of the data imperfection at the decoder including both sequence loss and sequence corruption, revealing the decoding demand and monitoring the data recovery. Furthermore, we extensively explored several data-dependent unevenness observed in the base error patterns and studied a few potential factors and their impacts on the data imperfection at the decoder both theoretically and experimentally. The results presented here introduce a more comprehensive channel model and offer a new angle towards the data recovery issue in DNA data storage by further elucidating the error characteristics of the storage process.Contactpoh.chuehloo@nus.edu.sg

When agricultural goods begin to be transported by those temperatures where value added products can be transported, cold room assessments in industrial applications must be validated. Computational feedback can improve the process consistency. Data on coolant and boundary dynamics can be evaluated by means of computer fluid dynamics (CFD). Air speed and air temperature are the distribution parameters in this analysis. When the vegetable temperature is preserved with less variations in the storage room, as well as the required air distribution, the best effects of proper preservations are achieved. With proper air circulation, temperature difference is minimised. Air speed is necessary to protect the cold storage of agricultural products. This project's main objective was to observe the decrease in temperature when PCMs are used within the cold storage space. Most of the experimental results demonstrated that in the event of regular openings of doors, also in the context of electricity failures using PCM, the temperature retained inside the cool space was 1–4.5 °C lower than without PCM, even in the cases of electrical failures using PCM. The present study focuses on applying guidelines to enhance cold room technology for improved outcomes in cold storage. Present work discussed about the reference methods and the techniques used in previous literature with future recommendations.

The technological innovation among a conscience as well as configured accessibility guidance conceptual model process has been combined mostly with digital technologies that generate integrated immediate transportation with adaptability shielding mechanism. The Intelligent as well as Adaptive Accessibility Benefits Interface updates and transmits roadway data collected by roadside sensors to a powerful guide known as a dynamic map (DM), which increases programmed portability control with high accuracy and reliability. To confirm the updating and establishment of roadway data to automatic motilities, we have built up a Wireless Electronic Traffic Mirror (WETM) that procures street data, perceives street items and moves acknowledgment data with a remote correspondence framework. Furthermore, to meet the security prerequisites, the AMAS should give the remote association most appropriate for portability. To give ideal remote availability, continuous remote channel checking, which quantifies the nature of remote channels with obstruction, is a significant assignment in the AMAS. The continuous peripheral distract surveillance implementation throughout combination with the generated WETM is mentioned in this report, and unambiguous confirmation including its exactness of every situation occurrence is revealed to explain the thinking beside how some the phenomenon including its pavement was improved.

In Intelligent Transportation Systems, images are the main data sources to be analyzed for providing intelligent and precision cognitive services. Therefore, how to protect the privacy of sensitive images in the process of information transmission has become an important research issue, especially in future no non-private data era. In this article, we design the Rearrangement-Arnold Cat Map (R-ACM) to disturb the relationship between adjacent pixels and further propose an efficient Bit-level Image Encryption Algorithm(Bl-IEA) based on R-ACM. Experiments show that the correlation coefficients of two adjacent pixels are 0.0022 in the horizontal direction, -0.0105 in the vertical direction, and -0.0035 in the diagonal direction respectively, which are obviously weaker than that of the original image with high correlations of adjacent pixels. What's more, the NPCR is 0.996120172, and the UACI is 0.334613406, which indicate that Bl-IEA has stronger ability to resist different attacks compared with other solutions. Especially, the lower time complexity and only one round permutation make it particularly suitable to be used in the time-limited intelligent transportation field.

Video-based person re-identification (Re-ID) aims at retrieving the person through the video sequences across non-overlapping cameras. Some characteristics of pedestrians are not consecutive across frames due to the variations of viewpoints, postures, and occlusions over time. However, existing methods ignore such data peculiarity and the networks tend to only learn those salient consecutive characteristics among frames in video sequences. As a result, the learned representations fail to cover all the characteristics of pedestrians, thus lacking integrity and discrimination. To tackle this problem, we present a novel deep architecture termed Hierarchical Mining Network (HMN), which mines as many pedestrians’ characteristics by referring to the temporal and intra-class knowledge. It consists of a novel Attentive Temporal Module (ATM) and a Dynamic Supervising Branch (DSB), with a Balancing Triplet Loss (BTL) assisting the training. The proposed ATM, with pedestrian perceiving capacity, is capable of evaluating each activation of features through temporal analysis, so that the temporally scattered characteristics of pedestrians can be better aggregated and the contaminated ones can be eliminated. Then, the DSB along with the BTL further enhances the integrity of representations by multiple supervision. Specifically, the DSB perceives the diversities of intra-class samples in each mini-batch and generates targeted supervising signals for them, in which process the BTL guarantees the signals with smaller intra-class variations and larger inter-class variations. Comprehensive experiments on two video-based datasets, i.e., MARS, and DukeMTMC-VideoReID, demonstrate the contribution of each component and the superiority of the proposed HMN over the state-of-the-arts. Benchmarking our model on three popular image-based datasets, i.e., Market1501, DukeMTMC-Reid, and MSMT17 additionally verifies the promising generalizability of the proposed DSB and BTL.

As the controller parameter adjustment process is simplified significantly by using learning algorithms, the studies about learning-based control attract a lot of interest in recent years. This article focuses on the intelligent servo control problem using learning from desired demonstrations. Compared with the previous studies about the learning-based servo control, a control policy using the broad learning system (BLS) is developed and first applied to a microrobotic system, since the advantages of the BLS, such as simple structure and no-requirement for retraining when new demos' data is provided. Then, the Lyapunov theory is skillfully combined with the complex learning algorithm to derive the controller parameters' constraints. Thus, the final control policy not only can obtain the movement skills of the desired demonstrations but also have the strong ability of generalization and error convergence. Finally, simulation and experimental examples verify the effectiveness of the proposed strategy using MATLAB and a microswimmer trajectory tracking system.

In state-of-the-art, it has proven that multi-view ensemble learning performs better than classical machine learning algorithms, with the optimized setting of views (subsets of features) usually. Obtaining the appropriate number of views for a given dataset is a complex problem in multi-view ensemble learning (MEL). The finding of the total number of possible views is an NP-hard problem, i.e., equivalent to Bell number. Moreover, the complexity of multi-view learning increases over a higher number of views of the dataset. Therefore, it is highly required to consider a smaller number of views with higher accuracy for optimal performance of MEL. In this work, MEL using Multi-Objective Particle Swarm Optimization (MEL-MOPSO) method has been proposed. The two objectives (number of views of the data and classification accuracy of MEL) have considered where the trade-off between objectives has been performed while searching for an optimal solution using Particle Swarm Optimization (PSO) in the process of multiobjective optimization. The experiments have been done over sixteen high-dimensional datasets using four state-of-art view construction methods. The individual views of the dataset has been utilized to learn through a support vector machine algorithm. The quantitative and non-parametric statistical analyses show that the proposed method has performed effectively and efficiently.

The present study focuses on designing the memory-based sampled data control approach to achieve the desired performance from switched-type of neural networks. The overall contribution of the study can be viewed from two directions. The first part is to derive the theoretical-based sufficient stability conditions that ensure the synchronization of uncontrolled and controlled systems. The idea behind the synchronization analysis is because of time complexity and computational cost. Once, we derive the closed-loop system then user can employ them in any practical application without interruptions. The existence of time-delays during the information transmission is inevitable, and in a practical sense, the delays, which are a function of time are well-effective. Besides, the sufficient conditions are derived in terms of solvable linear matrix inequalities (LMIs), which can be solved to ensure the global asymptotical stability of the error system, derived from uncontrolled–controlled systems. Technically, the work integrates with the new type of discontinuous Lyapunov–Krasovskii functional candidate with integral terms, several delay-dependent stability conditions, and Wirtinger-based integral inequality. The second part works on utilizing the controlled system into a real-life application, say, encryption/decryption process. To an evident, the proposed controlled system is considered as crypto-system in the encryption/decryption process.

Active 3D thermography combines 3D shape and active thermography. Thus, it enables the provision of the intuitive thermographic inspection results for composite parts with a complex geometry. However, conventional 3D thermography acquires thermographic information and 3D shape through at least two independent sensors and requires complex cross-modal image registration algorithm based on the keypoint detection and matching. In this paper, an active 3D thermography system is proposed using only one thermal camera for moving objects. This system does not require an independent 3D sensor while the thermal camera acts as the 3D sensor. Nature behind that is that a mathematical model is proposed to unify the line scanning thermography and laser triangulation in the dynamic scanning process, taking a line laser as both heat excitation for active thermography and spatial coding for 3D reconstruction. Furthermore, the model enables the feature-free registration of thermogram sequence and 3D shape, so that the registration is fast and robust without keypoint features. The experiments on standard height specimens, 3D printing glass fiber composites, and carbon fiber intake tube have shown the error is calibrated within 0.25 mm in the range of 1 mm to 150 mm and evidenced the capability for sub-surface defects detection.

With the development of data plane programmable Software-Defined Networking (SDN), Distributed Denial of Service (DDoS) attacks on the data plane increasingly become fatal. Currently, traditional attack detection methods are mainly used to detect whether a DDoS attack occurs and it is difficult to find the path that the attack flow traverses the network, which makes it difficult to accurately mitigate DDoS attacks. In this paper, we propose a detection method based on Spatial-Temporal Graph Convolutional Network (ST-GCN) over the data plane programmable SDN, which maps the network into a graph. It senses the state of switches through In-band Network Telemetry (INT) with sampling, inputs the network state into the spatial-temporal graph convolutional network detection model, and finally finds out the switches through which DDoS attack flows pass. Based on this, we propose a defense method combined with an enhanced whitelist and a precise dropping strategy. The evaluation results show that our detection method can accurately detect the path that the DDoS attack flows pass through. Compared to classic methods, our method improves the detection accuracy by nearly 10%. At the same time, the southbound interface load, CPU overhead and defense process are much lower than the classic methods.

Binarized neural networks (BNNs) have drawn significant attention in recent years, owing to great potential in reducing computation and storage consumption. While it is attractive, traditional BNNs usually suffer from slow convergence speed and dramatical accuracy-degradation on large-scale classification datasets. To minimize the gap between BNNs and deep neural networks (DNNs), we propose a new framework of designing BNNs, dubbed Hyper-BinaryNet, from the aspect of enhanced information-flow. Our contributions are threefold: 1) Considering the capacity-limitation in the backward pass, we propose an 1-bit convolution module named HyperConv. By exploiting the capacity of auxiliary neural networks, BNNs gain better performance on large-scale image classification task. 2) Considering the slow convergence speed in BNNs, we rethink the gradient accumulation mechanism and propose a hyper accumulation technique. By accumulating gradients in multiple variables rather than one as before, the gradient paths for each weight increase, which escapes BNNs from the gradient bottleneck problem during training. 3) Considering the ill-posed optimization problem, a novel gradient estimation warmup strategy, dubbed STE-Warmup, is developed. This strategy prevents BNNs from the unstable optimization process by progressively transferring neural networks from 32-bit to 1-bit. We conduct evaluations with variant architectures on three public datasets: CIFAR-10/100 and ImageNet. Compared with state-of-the-art BNNs, Hyper-BinaryNet shows faster convergence speed and outperforms existing BNNs by a large margin.

In laminectomy, the laminas above the spinal nerves need to be removed by milling. This process is very dangerous because of the special lesion, and surgeons usually control the milling process by sensing the surgical status. However, in robot-assisted laminectomy, safety monitoring will be difficult because the surgeon is not directly involved. Based on the human tactile perception mechanism, this paper proposes a surgical status perception method using the acceleration sensor and force sensor installed on the robot. The Sinc-convolutional layer is introduced to process high-frequency vibration signals, and the obtained features are processed together with the averaged force signals using a 1D convolutional network. The proposed method classifies the surgical status into 5 categories, and outputs the burr clogging probability. Experiments on animal bones show the effectiveness of the proposed model. In addition, it has also been proved that the fusion of the two tactile signals can significantly improve the status recognition accuracy under changing milling parameters.

Deep neural networks have achieved breakthrough improvement in various application fields. Nevertheless, they usually suffer from a time-consuming training process because of the complicated structures of neural networks with a huge number of parameters. As an alternative, a fast and efficient discriminative broad learning system (BLS) is proposed, which takes the advantages of flat structure and incremental learning. The BLS has achieved outstanding performance in classification and regression problems. However, the previous studies ignored the reason why the BLS can generalize well. In this article, we focus on the interpretation from the viewpoint of the frequency domain. We discover the existence of the frequency principle in BLS, i.e., the BLS preferentially captures low-frequency components quickly and then fits the high frequencies during the incremental process of adding feature nodes and enhancement nodes. The frequency principle may be of great inspiration for expanding the application of BLS.

Spatial downsampling layers are favored in convolutional neural networks (CNNs) to downscale feature maps for larger receptive fields and less memory consumption. However, for discriminative tasks, there is a possibility that these layers lose the discriminative details due to improper pooling strategies, which could hinder the learning process and eventually result in suboptimal models. In this paper, we present a unified framework over the existing downsampling layers (e.g., average pooling, max pooling, and strided convolution) from a local importance view. In this framework, we analyze the issues of these widely-used pooling layers and figure out the criteria for designing an effective downsampling layer. According to this analysis, we propose a conceptually simple, general, and effective pooling layer based on local importance modeling, termed as {\em Local Importance-based Pooling} (LIP). LIP can automatically enhance discriminative features during the downsampling procedure by learning adaptive importance weights based on inputs. Experiment results show that LIP consistently yields notable gains with different depths and different architectures on ImageNet classification. In the challenging MS COCO dataset, detectors with our LIP-ResNets as backbones obtain a consistent improvement ($\ge 1.4\%$) over the vanilla ResNets, and especially achieve the current state-of-the-art performance in detecting small objects under the single-scale testing scheme., Comment: ICCV 2019 camera ready version. Code: https://github.com/sebgao/LIP

Using a unique large-scale clickstream data set from a mobile commerce retailer, we examine the effect of the mobile shopping app and mobile operating system versions on consumer search and purchase behaviors in mobile commerce. We find that the consumers with an up-to-date shopping app or mobile operating system (but not both) compared to previous versions conduct more searches in terms of increased product page views and time spent on product pages, which in turn results in higher consumer purchase probability. Surprisingly, the consumers with both an up-to-date shopping app and mobile operating system have reduced search activities and lower purchase probabilities. Our results show that the mobile commerce environment affects the consumer shopping process in a non-linear fashion and indicate that there exists a "sweet spot" with the right level of smoothness in the mobile shopping environment. Our computations show that when compared to the baseline of using previous versions of both, using an up-to-date mobile shopping app alone increases consumer search activities by 336 more product pages or 1,975 more seconds spent, and using an up-to-date mobile operating system alone increases consumer search activities by 976 more product pages or 7,258 more seconds spent. However, using both an up-to-date mobile shopping app and mobile operating system only increases consumer search activities by 270 more product pages or 1,569 more seconds, compared to the baseline. One more product page viewed or one more second spent increases the average consumer purchase probability by 4.38% or 1.35%, respectively.

Radio-Frequency IDentification (RFID) technology is an essential enabler of a multitude of intelligent applications. The robust authentication of RFID system components is critical in providing trustworthy data delivery from/to tags. In this paper, we propose an authentication protocol based on monitoring the transmissions between readers and tags in the system. The proposed authentication scheme is based on injecting decoys within the exchanged communications (between RFID readers and tags) and is used in the authentication process. Furthermore, the proposed authentication scheme is mathematically modeled and validated using extensive simulation. The simulations results show that the proposed scheme provides a 100% confidence level in the authentication of tags and detection of compromised readers.

Many vision-based systems suffer from poor levels of contrast and brightness, mainly because of inadequate and improper illumination during the image acquisition process. As a result, the required specified information from the acquired image is not available for the particular application. In general, it is hard to achieve a balance between the improvement of contrast and brightness in image enhancement. By introducing nature-inspired optimization in image enhancement, the best features of the image are utilized, and the complexity related to the nonlinearity of images can be solved with various constraints, like a balance between contrast and brightness. In this work, a novel automatic method for image enhancement to find a balance between contrast and brightness is developed by using Cuckoo Search-optimized image fusion. First, the Cuckoo Search-based optimization algorithm generates two sets of optimized parameters. These parameter sets are used to generate a pair of enhanced images, one with a high degree of sharpness and contrast, the other is bright and has been improved without losing the level of detail. Furthermore, the two enhanced images are fused by the fusion process to obtain an output image where the contrast and brightness are in balance. The effectiveness of the proposed method is verified by applying it to standard images (CVG-UGR image database) and lathe tool images. Experimental results demonstrated that the proposed method performs better with regard to both the quality of contrast and brightness, moreover, yields enhanced quality evaluation metrics compared to the other conventional techniques.

Fault prognosis based on single model is generally inaccurate due to the varying working conditions. A multistage fault prognosis methodology combining stage identification with Bayesian networks (BNs) and time series approach with particular emphasis on the autoregressive moving average (ARMA) model is proposed to solve this problem. In the first stage, degradation data are identified, and outliers are marked by the Euclidean distance. Degenerate attributes of outliers are finely identified by BNs and matched to the corresponding model. In the second stage, the ARMA model is used for prognosis according to the results of the fine identification. Subsequently, the double-precision identification and ARMA submodel prognosis are carried out alternately throughout the prognosis process. Three degradation types of PMSM are simulated to verify the applicability of the method. Result shows that it can track the changes in the degradation in time and obtains better results.

In Wireless Multimedia Sensor Networks (WMSNs), nodes capable of retrieving video, audio, images, and small scale sensor data, tend to generate immense traffic of various types. The energy-efficient transmission of such a vast amount of heterogeneous multimedia content while simultaneously ensuring the quality of service and optimal energy consumption is indispensable. Therefore, we propose a Power-Efficient Wireless Multimedia of Things (PE-WMoT), a robust and energy-efficient cluster-based mechanism to improve the overall lifetime of WMSNs. In a PE-WMoT, nodes declare themselves Cluster Heads (CHs) based on available resources. Once cluster formation and CH declaration processes are completed, the Sub-Cluster (SC) formation process triggers, in which application base nodes within close vicinity of each other organize themselves under the administration of a Sub-Cluster Head (SCH). The SCH gathers data from member nodes, removes redundancies, and forwards miniaturized data to its respective CH. PE-WMoT adopts a fuzzy-based technique named the analytical hierarchical process, which enables CHs to select an optimal SCH among available SCs. A PE-WMoT also devises a robust scheduling mechanism between CH, SCH, and child nodes to enable collision-free data transmission. Simulation results revealed that a PE-WMoT significantly reduces the number of redundant packet transmissions, improves energy consumption of the network, and effectively increases network throughput.

In addition to 3D geometry, accurate representation of texture is important when digitizing real objects in virtual worlds. Based on a single consumer RGBD sensor, accurate texture representation for static objects can be realized by fusing multi-frame information; however, extending the process to dynamic objects, which typically have time-varying textures, is difficult. Thus, to address this problem, we propose a compact keyframe-based representation that decouples a dynamic texture into a basic static texture and a set of multiplicative changing maps. With this representation, the proposed method first aligns textures recorded from multiple keyframes with the reconstructed dynamic geometry of the object. Errors in the alignment and geometry are then compensated in an innovative iterative linear optimization framework. With the reconstructed texture, we then employ a scheme to synthesize the dynamic object from arbitrary viewpoints. By considering temporal and local pose similarities jointly, dynamic textures in all keyframes are fused to guarantee high-quality image generation. Experimental results demonstrate that the proposed method handles various dynamic objects, including faces, bodies, cloth, and toys. In addition, qualitative and quantitative comparisons demonstrate that the proposed method outperforms state-of-the-art solutions.

Fetal congenital heart disease (CHD) is the most common type of fatal congenital malformation. Fetal four-chamber (FC) view is a significant and easily accessible ultrasound (US) image among fetal echocardiography images. Automatic detection of four fetal heart chambers considerably contributes to the early diagnosis of fetal CHD. Furthermore, robust and discriminative features are essential for detecting crucial visualizing medical images, especially fetal FC views. However, it is an incredibly challenging task due to several key factors, such as numerous speckles in US images, the fetal four chambers with small size and unfixed positions, and category confusion caused by the similarity of cardiac chambers. These factors hinder the process of capturing robust and discriminative features, hence destroying the fetal four chambers' precise detection. Therefore, we propose an intelligent feature learning detection system (FLDS) for FC views to detect the four chambers. A multistage residual hybrid attention module (MRHAM) presented in this paper is incorporated in the FLDS for learning powerful and robust features, helping FLDS accurately locate the four chambers in the fetal FC views. Extensive experiments demonstrate that our proposed FLDS outperforms the current state-of-the-art, including the precision of 0.919, the recall of 0.971, the F1 score of 0.944, the mAP of 0.953, and the frames per second (FPS) of 43. In addition, our proposed FLDS is also validated on other visualizing nature images such as the PASCAL VOC dataset, achieving a higher mAP of 0.878 while input size is 608 * 608.

Outlier detection in multivariate data is one of the critical challenges in preprocessing phase. Many outlier detection methods have been emerged for the past few years to perform outlier detection efficiently in multivariate datasets. The prediction accuracy cannot be improved without proper outlier analysis and the prediction model might not confirm the expected behavior. The generation of huge data in real time applications makes the outlier detection process more crucial and challenging. Most of the currently available detection methods are based on mean and covariance that are not suitable for handling large volume of datasets, they are suitable for handlind static data and simple data to detect outliers. They cannot cope up with large scale data. So, there is a need for an efficient outlier detection model to detect the outliers in multivariate datasets. The primary objective of this research work is to develop a robust model for outlier detection in multivariate data. To achieve this, the work proposed an enhanced Hidden Semi-Markov Model (HSMM) which allows arbitrary time distribution in its states to detect outliers. The proposed work utilized six benchmark datasets and the performance is compared with several outlier detection algorithms such as HMM, iForest, FastABOD, and Expose. The work achieves 98.2 % of accuracy which is significantly better for detecting outliers in multivariate dataset. The proposed work improvised the percentage of acheivements between 2% to 25% than the currently available models.. The experimental analysis shows that the proposed model performs well than the currently available models in terms of accuracy, and receiver operation curve (ROC).

In intelligent transportation system (ITS), accurate and real-time prediction of short-term multi-lane traffic flow with existing traffic data is an important part of urban traffic planning, traffic management and control. The data generated in the process of vehicle driving has the cooperative characteristics of multi-source, space-time and dynamic. Combining the data with high-performance computing or cloud computing, a new space-time information framework is designed, which is of great significance for the analysis and prediction of traffic flow, as well as intelligent traffic management, service and decision-making. This paper analyzes the statistical characteristics of urban road traffic flow from the two dimensions of time and space through the spatial-temporal correlation between multi-lane short term traffic flow in single observation point and multi observation points. We constructed a data fusion powered bi-directional long short term memory (DFBD-LSTM) model for individual lane and aggregate traffic flow, then used this model to predict multi-lane short term traffic flow. By taking individual lane traffic flow and aggregate traffic flow as different variables, the model produces more accurate predictions, which can better guide people to travel, alleviate the congestion of urban road traffic network to a certain extent, and improve the utilization rate and transportation efficiency of traffic road.

We present a novel Digital Twin Framework for a portable table-top sized temperature-controlled system constructed entirely using free and open-source software programs. By utilizing tools such as Robotic Process Automation (RPA) and a PLC program, the entire process was automated to achieve a true Digital Twin. Further, a mathematical model was constructed for the temperature-controlled system. The data from the physical model and the mathematical model was then compared and a what-if scenario was analyzed to predict and prevent a potential failure in the physical system. Additionally, a simulation model was created to visualize the results.

Feature selection aims to select strongly relevant features and discard the rest. Recently, embedded feature selection methods, which incorporate feature weights learning into the training process of a classifier, have attracted much attention. However, traditional embedded methods merely focus on the combinatorial optimality of all selected features. They sometimes select the weakly relevant features with satisfactory combination abilities and leave out some strongly relevant features, thereby degrading the generalization performance. To address this issue, we propose a novel embedded framework for feature selection, termed feature selection boosted by unselected features (FSBUF). Specifically, we introduce an extra classifier for unselected features into the traditional embedded model and jointly learn the feature weights to maximize the classification loss of unselected features. As a result, the extra classifier recycles the unselected strongly relevant features to replace the weakly relevant features in the selected feature subset. Our final objective can be formulated as a minimax optimization problem, and we design an effective gradient-based algorithm to solve it. Furthermore, we theoretically prove that the proposed FSBUF is able to improve the generalization ability of traditional embedded feature selection methods. Extensive experiments on synthetic and real-world data sets exhibit the comprehensibility and superior performance of FSBUF.

The process operating performance assessment is critical for the smelting process of electro-fused magnesium furnaces to improve quality of the magnesia product and pursue optimal comprehensive economic benefit. This paper proposes a new method of multi-source heterogeneous information deep feature fusion (MSHIDFF) to achieve higher accuracy operating performance assessment in the electro-fused magnesium smelting process. Firstly, we utilize convolutional neural network, bidirectional long short-term memory network and stacked auto-encoder to extract deep features from raw image, sound and current of different performance grades. Furthermore, those multi-source deep features are fused and the softmax regression with attention mechanism is employed to train a neural network classifier for the fused deep features of different performance grades. The simulation results show that the proposed MSHIDFF method obtains the superior assessment accuracy.

For taking out the adjustment process of sparse auto-encoder for broad learning system, S. Feng et al. proposed fuzzy broad learning system by replacing the feature nodes of broad learning system with Takagi-Sugeno fuzzy systems. In fuzzy broad learning system, artificial parameters selection of ridge regression might result in the decrease in testing accuracy. To overcome this shortcoming of fuzzy broad learning system, this paper builds a novel model of fuzzy broad learning system based on accelerating amount by introducing the accelerating amount into fuzzy broad learning system. The theoretical result on the universal approximation property of fuzzy broad learning system based on accelerating amount is presented. Three experiment studies on the regression problems of UCI, fashion MNIST and medical MNIST data-sets are performed to show the improvement in testing accuracy.To overcome this shortcoming of FBLS, this paper builds a novel model of FBLS based on accelerating amount (AA-FBLS) by introducing the accelerating amount into FBLS. The theoretical result on the universal approximation property of AA-FBLS is presented. Two experiment studies on the approximation of Fashion-MNIST and MedMNIST data-sets are performed to show the improvement in testing accuracy.

This paper proposes a novel ultra-Fast self-powered bidirectional solid-state circuit breaker (SSCB) with voltage overshoot suppression that uses gallium nitride (GaN) switches with very low conduction losses. This SSCB can be used in low and medium AC/DC networks for both short-circuit protection and overcurrent protection. An exhaustive methodology for RCD snubber design applied to SSCBs is provided. A comprehensive review of the technical literature is presented exhibiting that the proposed SSCB reduces conduction losses and detection time in comparison with other similar SSCBs. The performance features of the proposed SSCB are verified through the functional tests that are carried out employing a hardware-implemented prototype. Experimental results confirm that our proposal is able to detect and process the short-circuit failure in 282ns; the authors have not found any faster results in the literature.

In pull-based development process, contributors submit their code to open-source projects by pull requests, which are accepted or rejected by reviewers. Contributors may modify their code, which causes several iterations of code review process, and makes code reviews time-consuming for both contributors and reviewers. In this article, we set out to study pull request modifications in a code review process. We collect nine projects on GitHub with 104 307 pull requests, and investigate pull request modifications through analyzing added commits after pull requests’ submission. By studying four research questions, we conclude our major findings as follow. First, 34.56% of collected pull requests have modifications. Pull requests with modifications have longer lifetime but higher pass rates. Second, we conclude eight modification types indicating why pull requests are modified. Third, we propose a novel method called MClassify to automatically classify pull request modifications, which achieves the accuracy of 0.807. Fourth, various modification types affect code review differently from the perspective of lifetime and pass rate. Pull requests with source control system management modifications have the longest lifetime. These findings enable developers and researchers to understand a pull-based code review process better and make improvements.

Fake news is inaccurate information that is intentionally disseminated for a specific purpose. If allowed to spread, fake news can harm the political and social spheres, so several studies are conducted to detect fake news. This study uses a deep learning method with several architectures such as CNN, Bidirectional LSTM, and ResNet, combined with pre-trained word embedding, trained using four different datasets. Each data goes through a data augmentation process using the back-translation method to reduce data imbalances between classes. The results showed that the Bidirectional LSTM architecture outperformed CNN and ResNet on all tested datasets.

In this paper, we propose to use the data augmentation of batch drop-block with varying dropping ratios for constructing diversity-achieving branches in person re-identification. Since a considerable portion of input images may be dropped, this reinforces feature learning of the un-dropped region but makes the training process hard to converge. Hence, we propose a novel double-batch-split co-training approach for remedying this problem. In particular, we show that the feature diversity can be well achieved with the use of multiple dropping branches by setting individual dropping ratio for each branch. Empirical evidence demonstrates that the proposed method performs competitively on popular person Re-ID datasets, including Market-1501, DukeMTMC-reID and CUHK03, and the use of more dropping branches can further boost the performance. Source code is available at url.