Search

The features play an important role for identification of the region of interest. Different kind of features exist, it is also essential to identify the accurate class of the features, challenging dataset like MICCAI BraTs brain tumor contains many tumor images. Features are helpful to detect the region of tumor with some of their characteristic. But due to many images and their information, the issue of data complexity is raised. because the data was found to be complex. Thus, due to the complexity, higher dimension features are reduced to low dimension features. Hence, there is a need for improved feature selection method. Furthermore, it is also essential to enhance the method for the SOM Map for the selection of deterministic feature after the extraction. The goal of the work is not only to select the accurate feature of tumor but also to segment the tumor intensity with the confidence element. The objective under umbrella of this work is to improve the feature selection method by using confidence element of interest through the determination of the best feature using the SOFM with FCM. The method works with the selection of the best features with higher accuracy. Those higher accurate Features are called the deterministic Features. These deterministic features are selected through improved weighted SOM. This improved SOM is further combined with FCM to cluster the Confidence element. Evaluation is made with comparison to ground truth reality images; Results show; DOI is 0.94, JI is 0.91, MSE is 0.058db and PSNR is 17.94db; MSE with small number highlights the performance of method. It can be compared with the state of the art or it can be compared with benchmark studies. Testing parameters from benchmark studies were JI, DOI, MSE and PSNR: JI accuracy value was 31.5%, DOI accuracy value was 47.3%, MSE value was 2.5dB and PSNR value was 40dB.A better region of interest is proposed method to determine the confidence element. The average accuracy over the dataset is determined in form of confidence element (ROI), overlap is for complex cases and average value is 94 percent.

A feature-tracking algorithm algorithm utilizing the proposed K-nearest-neighbor vector field consensus (KVFC) to filter outliers is developed to monitor the dynamic changes of sea ice retrieval from synthetic aperture radar (SAR) images. The KVFC is based on vector field consensus and combines with local neighborhood correspondences to optimize the elimination of outliers while retaining inliers as many as possible. The proposed KVFC was evaluated and compared with several algorithms on three standard datasets and Sentinel-1 image pairs in the Fram Strait and the Beaufort Sea. The KVFC obtained more sea-ice drift vectors than the nearest neighbor similarity ratio (NNSR) with a 0.7 threshold and generated dense distribution of sea-ice drift vectors combined with the HH and HV channels. Using buoy datasets to calculate the sea-ice drift speed and evaluate algorithm performance, the proposed approach yielded a lower mean error (KVFC: −0.150 cm/s, NNSR: 0.407 cm/s), lower root mean square error (KVFC: 0.476 cm/s, NNSR: 1.817 cm/s), and lower angle deviation (KVFC: 3.542°, NNSR: 10.318°) compared to the NNSR.

With the proliferation of LiDAR sensors and 3-D vision cameras, 3-D point cloud analysis has attracted significant attention in recent years. In this article, we propose a novel 3-D point cloud learning network, referred to as dynamic point feature aggregation network (DPFA-Net), by selectively performing the neighborhood feature aggregation (FA) with dynamic pooling and an attention mechanism. DPFA-Net has two variants for semantic segmentation and classification of 3-D point clouds. As the core module of the DPFA-Net, we propose an FA layer, in which features of the dynamic neighborhood of each point are aggregated via a self-attention mechanism. In contrast to other segmentation models, which aggregate features from fixed neighborhoods, our approach can aggregate features from different neighbors in different layers providing a more selective and broader view to the query points and focusing more on the relevant features in a local neighborhood. In addition, to further improve the performance of semantic segmentation, we exploit the background–foreground (BF) information and present two novel approaches, namely, two-stage BF-Net and BF regularization. Experimental results show that the proposed DPFA-Net achieves the state-of-the-art overall accuracy score of 89.22% for semantic segmentation on the Stanford large-scale 3-D Indoor Spaces (S3DIS) dataset and provides consistently satisfactory performance across different tasks of semantic segmentation, part segmentation, and 3-D object classification. Our model achieves 93.1% accuracy on the ModelNet40 dataset and provides a mean shape intersection-over-union (IoU) value of 85.5% for part segmentation on the ShapeNet-Part dataset. It is also computationally more efficient compared to other methods. [ABSTRACT FROM AUTHOR]

Navigation solutions suitable for cases when both autonomous robot’s pose (i.e., attitude and position) and its environment are unknown are in great demand. Simultaneous Localization and Mapping (SLAM) fulfills this need by concurrently mapping the environment and observing robot’s pose with respect to the map. This work proposes a nonlinear observer for SLAM posed on the manifold of the Lie group of $\mathbb {SLAM}_{n}(3)$ , characterized by systematic convergence, and designed to mimic the nonlinear motion dynamics of the true SLAM problem. The system error is constrained to start within a known large set and decay systematically to settle within a known small set. The proposed estimator is guaranteed to achieve predefined transient and steady-state performance and eliminate the unknown bias inevitably present in velocity measurements by directly using measurements of angular and translational velocity, landmarks, and information collected by an inertial measurement unit (IMU). Experimental results obtained by testing the proposed solution on a real-world dataset collected by a quadrotor demonstrate the observer’s ability to estimate the six-degrees-of-freedom (6 DoF) robot pose and to position unknown landmarks in three-dimensional (3D) space. [ABSTRACT FROM AUTHOR]

Convolutional neural networks have achieved amazing success in many areas in recent years, and VGG is a widely used convolutional neural network model. However, it has some limitations, such as a large number of parameters, which take up significant memory and thus restrict its application in resource-constrained scenarios such as mobile devices and embedded systems. In convolutional neural network models, the different number of convolutional layers can extract different granularity features, which represent the different levels of importance in the image recognition process. Here, we propose a new VGG architecture with different granularity feature combinations that combine different granularity features from block1, block2, block3, block4, and block5 in VGG. Each block is followed by a local fully connected layer to reduce the dimensionality of the coarse and fine features, and five different granularity features are combined as the input of the first global fully connected layer. By combining the features of different blocks, it can increase information flow from a lower layer directly to a fully connected layer and increase feature reuse without adding too many connections. The addition of five local fully connected layers means an increase in parameters, so we reduce the neuron number in two global fully connected layers to reduce the number of parameters. The well-known datasets CIFAR-10 and MNIST were used to evaluate the network's classification performance. The experimental results show that the proposed model achieves better training and testing performance than traditional VGGs and reduces the number of parameters.

In this paper, we exploit a method for identifying flaws on product surface based on spatial connectivity domain. A number of algorithms for detecting local features exist that were established to enhance the efficiency and accuracy of identifying interest features, such as AKAZE, BFSIFT, BRIEF, BRISK, ORB, SURF, SIFT and PCA-SIFT algorithm. But the data of flaws on product surface which is similar and consistent with the background intensity became a dilemma to detect the feature of image. In terms of identifying flaws on product surface, the above algorithms are not effective and accurate. Our aim is to enhance the accuracy of detecting the feature of flaws on product surface, so that the product with flaws could be accurately identified in industrial production. Therefore, we propose a method to identify flaws on product surface based on spatial connectivity domain. Compared with some other algorithms, such as the extracting texture algorithm, the detecting local feature algorithm and the identifying edge algorithm, our proposed method is more effective and accurate in detecting the local feature flaws on product surface of auto parts in automotive manufacturing factory.

The profound impact of deep learning and particularly of convolutional neural networks (CNNs) in automatic image processing has been decisive for the progress and evolution of remote sensing (RS) hyperspectral imaging (HSI) processing. Indeed, CNNs have stated themselves as the current state of the art, reaching unparalleled results in HSI classification. However, most CNNs were designed for RGB images, and their direct application to HSI data analysis could lead to nonoptimal solutions. Moreover, CNNs perform classification based on the identification of specific features, neglecting the spatial relationships between different features (i.e., their arrangement) due to pooling techniques. The capsule network (CapsNet) architecture is an attempt to overcome this drawback by nesting several neural layers within a capsule, connected by dynamic routing, both to identify not only the presence of a feature but also its instantiation parameters and to learn the relationships between different features. Although this mechanism improves the data representations, enhancing the classification of HSI data, it still acts as a black box, without control of the most relevant features for classification purposes. Indeed, important features could be discriminated against. In this article, a new multiple attention-guided CapsNet is proposed to improve feature processing for RS-HSIs’ classification, both to improve computational efficiency (in terms of parameters) and increase accuracy. Hence, the most representative visual parts of the images are identified using a detailed feature extractor coupled with attention mechanisms. Extensive experimental results have been obtained on five real datasets, demonstrating the great potential of the proposed method compared to other state-of-the-art classifiers. [ABSTRACT FROM AUTHOR]

Recommendation systems have become an important field of research in computer science and physics. In recent years, breakthroughs have been achieved in social, biological, and research cooperation networks. With the popularization of big data and deep learning technology development, graph structures are increasingly being used to represent large-scale and complex data in the real world. In this paper, we reviewed the progress made in recommendation systems research in the past 20 years and comprehensively classified recommender systems based on the heterogeneous input features. We introduced layering in the classification of recommendation systems. Furthermore, we proposed a new hierarchical classification model of recommendation systems divided into three layers: feature input, feature learning, and output layers. In the feature learning layer, existing recommendation systems were divided into graph-based, text-based, behavior-based, spatiotemporal-based, and hybrid recommendation systems. Additionally, we provided evaluation index, open-source implementation, experimental comparison and the relative merits for each recommendation method. Subsequently, future development directions of recommendation systems are discussed.

Two novel robust nonlinear stochastic full pose (i.e., attitude and position) estimators on the Special Euclidean Group $\mathbb {SE}(3)$ are proposed using the available uncertain measurements. The resulting estimators utilize the basic structure of the deterministic pose estimators adopting it to the stochastic sense. The proposed estimators for six degrees of freedom (DOF) pose estimations consider the group velocity vectors to be contaminated with constant bias and Gaussian random noise, unlike nonlinear deterministic pose estimators which disregard the noise component in the estimator derivations. The proposed estimators ensure that the closed-loop error signals are semi-globally uniformly ultimately bounded in mean square. The efficiency and robustness of the proposed estimators are demonstrated by the numerical results which test the estimators against high levels of noise and bias associated with the group velocity and body-frame measurements and large initialization error. [ABSTRACT FROM AUTHOR]

This paper compares the three levels of data fusion with the goal of determining the optimal level of data fusion for multi-sensor human activity data. Using the data processing pipeline, gyroscope and accelerometer data was fused at the sensor-level, feature-level and decision-level. For each level of data fusion four different techniques were used with varying levels of success. This analysis was performed on four human activity publicly-available datasets along with four well-known machine learning classifiers to validate the results. The decision-level fusion (Acc = 0.7443±0.0850) outperformed the other two levels of fusion in regards to accuracy, sensor level (Acc = 0.5934 ± 0.1110) and feature level (Acc = 0.6742 ± 0.0053), but, the processing time and computational power required for training and classification were far greater than practical for a HAR system. However, Kalman filter appear to be the more efficient method, since it exhibited both good accuracy (Acc = 0.7536 ± 0.1566) and short processing time (time = 61.71ms ± 63.85); properties that play a large role in real-time applications using wearable devices. The results of this study also serve as baseline information in the HAR literature to compare future methods of data fusion. [ABSTRACT FROM AUTHOR]

Autism spectrum disorder (ASD) is a neuro dysfunction which causes the repetitive behavior and social instability of patients. Diagnosing ASD has been of great interest. However, due to the lack of discriminate differences between neuroimages of healthy persons and ASD patients, there has been no powerful diagnosis approach. In this study, we have designed brain network-based features for the diagnosis of ASD. Specifically, we have used the 264 regions based parcellation scheme to construct a brain network from a brain functional magnetic resonance imaging (fMRI). Then we have defined 264 raw brain features by the 264 eigenvalues of the Laplacian matrix of the brain network and another three features by network centralities. By applying a feature selection algorithm, we have obtained 64 discriminate features. Furthermore, we have trained several machine learning models for diagnosing ASD with our obtained features on ABIDE (Autism Brain Imaging Data Exchange) dataset. With our derived features, the linear discriminant analysis has achieved the classification accuracy of 77.7%, which is better than the state-of-the-art results.

Melanoma is considered a fatal type of skin cancer. However, it is sometimes hard to distinguish it from nevus due to their identical visual appearance and symptoms. The mortality rate because of this disease is higher than all other skin-related consolidated malignancies. The number of cases is growing among young people, but if it is diagnosed at an earlier stage, then the survival rates become very high. The cost and time required for the doctors to diagnose all patients for melanoma are very high. In this paper, we propose an intelligent system to detect and distinguish melanoma from nevus by using the state-of-the-art image processing techniques. At first, the Gaussian filter is used for removing noise from the skin lesion of the acquired images followed by the use of improved K-mean clustering to segment out the lesion. A distinctive hybrid superfeature vector is formed by the extraction of textural and color features from the lesion. Support vector machine (SVM) is utilized for the classification of skin cancer into melanoma and nevus. Our aim is to test the effectiveness of the proposed segmentation technique, extract the most suitable features, and compare the classification results with the other techniques present in the literature. The proposed methodology is tested on the DERMIS dataset having a total number of 397 skin cancer images: 146 are melanoma and 251 are nevus skin lesions. Our proposed methodology archives encouraging results having 96% accuracy.

Cascading failure may incur catastrophic consequences. The high order and non-simultaneousness properties are the leading causes for the massive power flow (PF) calculations of the online static security assessment (SSA). Current PF models fail to satisfy the requirements for accuracy, speed, and robustness. This paper proposes a novel method where line outage distribution factor (LODF) model and AC model are coordinated by a binary classifier. The classifier determines the feasibility of LODF model in a specific case by evaluating its potential error. Cases are preferred to be analyzed by LODF model, while the ones with large potential errors will be analyzed by AC model. Case studies of three IEEE systems and the Texas 2000-bus system are given to verify the effectiveness, flexibility, and robustness of the proposed method.

A joint and accurate optical signal-to-noise ratio (OSNR) estimation and modulation formats identification (MFI) scheme based on the artificial neural network (ANN) is proposed and demonstrated via both simulation and the experiment system. The proposed scheme employs ANN to estimate OSNR and modulation formats from the OSNR and modulation formats dependent features, kurtosis, and amplitude variance. Simulation results show that the proposed scheme can achieve high OSNR estimation and MFI accuracy over wide OSNR ranges for the commonly used modulation formats such as QPSK, 8 quadratic-amplitude modulation (QAM), 16QAM, and 64QAM. Meanwhile, experimental results also indicate that the mean OSNR estimation errors are 0.15 dB, 0.41 dB, and 0.49 dB for QPSK, 8QAM, and 16QAM over wide ranges OSNR of 10-17 dB, 14-20 dB, and 17-25 dB, respectively. Additionally, 100% MFI accuracy for the commonly used modulation formats in our scheme is also confirmed experimentally. Compared with the convolutional neural network and the deep neural network, the proposed scheme shows comparable estimation and identification performance, and needs less computational resource. Therefore, our scheme can be considered an attractive technique for joint OSNR estimation and MFI in future reconfigurable optical networks.

New approaches have been proposed to detect and exploit sparsity in adaptive systems. However, the sparsity is not always explicit among the system coefficients, thus requiring some tools to reveal it. By means of the so-called feature function, we propose the low-complexity feature stochastic gradient (LF-SG) algorithm to exploit hidden sparsity. The proposed algorithm aims at reducing the computational load of the learning process, as compared to the least-mean-square (LMS) algorithm. We focus on block-lowpass systems, but the proposed approach can easily be adapted to exploit other kinds of features of the unknown system, e.g., highpass and bandpass characteristics. Then, we analyze some properties of the LF-SG algorithm, namely its steady-state mean squared error (MSE), its bias, and the choice of the step-size parameter. Simulation results illustrate the competitive MSE performance of the LF-SG in comparison with the LMS, but the former algorithm requires much fewer multiplication operations to identify lowpass systems. For instance, to identify a measured room impulse response, the LF-SG algorithm realized less than half of the multiplication operations required by the LMS algorithm.

This article proposes an advanced cognition-driven electromagnetic (EM) optimization incorporating transfer function-based feature surrogate for EM optimization of microwave filters. The proposed optimization technique addresses the situations where the response of the starting point for design optimization is far away from the design specifications. This article proposes to extract transfer function-based feature parameters for optimization to address the challenge that the features cannot be clearly and explicitly identified from the filter response. Multiple transfer function-based feature parameters are extracted and used to develop the feature surrogate model for the proposed cognition-driven optimization. Furthermore, we derive new objective functions for the cognition-driven optimization directly in the feature space. The proposed cognition-driven optimization incorporating transfer function-based feature surrogate can achieve faster convergence than the existing feature-assisted EM optimization methods. Two examples of EM optimizations of microwave filters are used to demonstrate the proposed technique. [ABSTRACT FROM AUTHOR]

Effective and efficient classification of synthetic aperture radar (SAR) images represents an important step toward image interpretation and knowledge discovery. Keys to classification performance include feature extraction and availability of class labels for training. Generative adversarial networks (GANs) are a recently advanced powerful framework to offer both automatic feature extraction and semisupervised and unsupervised learning. For its known drawbacks such as mode collapse and training instability, noteworthy improvements include the deep convolutional GANs (DCGANs). In the letter, we proposed a new approach, which further advances DCGANs with autoencoder (AE) for intermediate feature extraction and reconstruction, in combination with multiclassifier (MC) for better context treatment. The resulting improvement on training stability and mode preservation over DCGAN is clearly demonstrated by the classification results across multifrequency bands (L-, C-, and X-bands). [ABSTRACT FROM AUTHOR]

There is a general agreement in the brain computer interface community that the feature extracting method called the common spatial pattern (CSP) combined with nonlinear classifiers can provide excellent results in some cases. However, CSP is also known to be very sensitive to noise and prone to over fitting, and the performance of this spatial filter is closely related to the operational frequency band of electroencephalogram data. To address this issue, we propose a new method FB-TRCSP+RF based on CSP and random forest. The FB-TRCSP is combined by the 8th-order Butterworth bandpass-filters and the CSP with Tikhonov regularization, which is a more robust feature extraction method compared to the CSP. Then, the model is applied to an experimental data set collected from 14 subjects and is compared with the non-regularization method FB-CSP+RF. The results show that the method we proposed yields relatively higher median classification accuracies and shows a stronger ability in subject-to-subject learning compared to prevailing approaches.

Currently, carbon dioxide (CO2) waveforms measured by capnography are used to estimate respiratory rate and end-tidal CO2 (EtCO2) in the clinic. However, the shape of the CO2 signal carries significant diagnostic information about the asthmatic condition. Previous studies have shown a strong correlation between various features that quantitatively characterize the shape of CO2 signal and are used to discriminate asthma from non-asthma using pulmonary function tests, but no reliable progress was made, and no translation into clinical practice has been achieved. Therefore, this paper reports a relatively simple signal processing algorithm for automatic differentiation of asthma and non-asthma. CO2 signals were recorded from 30 non-asthmatic and 43 asthmatic patients. Each breath cycle was decomposed into subcycles, and features were computationally extracted. Thereafter, feature selection was performed using the area (Az) under the receiver operating characteristics curve analysis. A classification was performed via a leave-oneout cross-validation procedure by employing a support vector machine. Our results show maximum screening capabilities for upward expiration (AR1), downward inspiration (AR2), and the sum of AR1 and AR2, with an Az of 0.892, 0.803, and 0.793, respectively. The proposed method obtained an average accuracy of 94.52%, sensitivity of 97.67%, and specificity of 90% for discrimination of asthma and non-asthma. The proposed method allows for automatic classification of asthma and non-asthma condition by analyzing the shape of the CO2 waveform. The developed method may possibly be incorporated in real-time for assessment and management of the asthmatic conditions.

Recently, many models have been developed to predict video Quality of Experience (QoE), yet the applicability of these models still faces significant challenges. Firstly, many models rely on features that are unique to a specific dataset and thus lack the capability to generalize. Due to the intricate interactions among these features, a unified representation that is independent of datasets with different modalities is needed. Secondly, existing models often lack the configurability to perform both classification and regression tasks. Thirdly, the sample size of the available datasets to develop these models is often very small, and the impact of limited data on the performance of QoE models has not been adequately addressed. To address these issues, in this work we develop a novel and end-to-end framework termed as DeepQoE. The proposed framework first uses a combination of deep learning techniques, such as word embedding and 3D convolutional neural network (C3D), to extract generalized features. Next, these features are combined and fed into a neural network for representation learning. A learned representation will then serve as input for classification or regression tasks. We evaluate the performance of DeepQoE with three datasets. The results show that for small datasets (e.g., WHU-MVQoE2016 and Live-Netflix Video Database), the performance of state-of-the-art machine learning algorithms is greatly improved by using the QoE representation from DeepQoE (e.g., 35.71% to 44.82%); while for the large dataset (e.g., VideoSet), our DeepQoE framework achieves significant performance improvement in comparison to the best baseline method (90.94% vs. 82.84%). In addition to the much improved performance, DeepQoE has the flexibility to fit different datasets, to learn QoE representation, and to perform both classification and regression problems. We also develop a DeepQoE based adaptive bitrate streaming (ABR) system to verify that our framework can be easily applied to multimedia communication service. The software package of the DeepQoE framework has been released to facilitate the current research on QoE. [ABSTRACT FROM AUTHOR]

This article proposes a multifeature-assisted neuro-transfer function (neuro-TF) surrogate-based electromagnetic (EM) optimization technique exploiting trust-region algorithms for microwave filter design. The proposed optimization technique addresses the situation where the response of the starting point is far away from the design specifications. We propose to utilize multiple feature parameters to help move the passband of the filter response into the range of design specifications. The pole–zero-based neuro-TF is introduced in this article to help extract the multiple feature parameters when the feature parameters of filter responses are not explicitly identified. Furthermore, we propose to derive new optimization objective functions to involve the multiple feature parameters. A new trust-region updating formulation for the modified optimization objective functions is derived to guarantee the optimization convergence. With the assistance of multiple feature parameters, the proposed surrogate-based EM optimization has a better capability of avoiding local minima and can reach the optimal EM solution faster than the surrogate-based EM optimizations without feature assistance. Three examples of EM optimizations of microwave filters are used to demonstrate the proposed technique. [ABSTRACT FROM AUTHOR]

Multi-scale-based image fusion is one of main fusion methods, in which multi-scale decomposition tool and feature extraction play very important roles. The quaternion wavelet transform (QWT) is one of the effective multi-scale decomposition tools. Therefore, this paper proposes a novel multimodal image fusion method using QWT and multiple features. First, we perform QWT on each source image to obtain low-frequency coefficients and high-frequency coefficients. Second, a weighted average fusion rule based on the phase and magnitude of low-frequency subband and spatial variance is proposed to fuse the low-frequency subbands. Next, a choose-max fusion rule based on the contrast and energy of coefficient is proposed to integrate the high-frequency subbands. Finally, the final fused image is constructed by inverse QWT. The proposed method is conducted on multi-focus images, medical images, infrared-visible images, and remote sensing images, respectively. Experimental results demonstrate the effectiveness of the proposed method.

For human gesture recognition applications, 3D-sensor-based approaches have received considerable attention and are crucial for future applications of an advanced body sensor network (BSN). A practical gesture recognition system is to apply active gesture patterns of a gesture-making user for body action classifications. 3D-sensor-based gesture recognition, categorized as biometric recognition in BSN applications, is greatly lack of the extensible cognition ability due to substandard recognition accuracy on the gesture-making user identity. To overcome this problem, this paper proposes an active gesture-based user identity recognition approach using a robust feature design, called user adaptation (UA) features, derived from a UA process. Two different UA features, namely Eigen Centroid-UA and Eigen Transform-UA features, were developed in this paper to accurately represent the adaptive learning tendency of gesture recognition for a specific gesture-making user. Compared with traditional 3D sensor gesture-based identity recognition approaches that employ only the feature of fixed body skeleton information without any UA designs, the presented UA-feature can exhibit fine adaptation learning continuously to the specific action user; therefore, superior identity recognition accuracy will be constantly ensured. To demonstrate the efficiency and effectiveness of the developed robust UA features in this paper, experiments on gesture-making user identification by Gaussian mixture model applying the proposed Eigen Centroid-UA feature and verification by support vector machine applying the proposed Eigen Transform-UA feature were conducted. [ABSTRACT FROM AUTHOR]

A joint and accurate optical signal-to-noise ratio (OSNR) estimation and modulation formats identification (MFI) scheme based on the artificial neural network (ANN) is proposed and demonstrated via both simulation and the experiment system. The proposed scheme employs ANN to estimate OSNR and modulation formats from the OSNR and modulation formats dependent features, kurtosis, and amplitude variance. Simulation results show that the proposed scheme can achieve high OSNR estimation and MFI accuracy over wide OSNR ranges for the commonly used modulation formats such as QPSK, 8 quadratic-amplitude modulation (QAM), 16QAM, and 64QAM. Meanwhile, experimental results also indicate that the mean OSNR estimation errors are 0.15 dB, 0.41 dB, and 0.49 dB for QPSK, 8QAM, and 16QAM over wide ranges OSNR of 10–17 dB, 14–20 dB, and 17–25 dB, respectively. Additionally, 100% MFI accuracy for the commonly used modulation formats in our scheme is also confirmed experimentally. Compared with the convolutional neural network and the deep neural network, the proposed scheme shows comparable estimation and identification performance, and needs less computational resource. Therefore, our scheme can be considered an attractive technique for joint OSNR estimation and MFI in future reconfigurable optical networks. [ABSTRACT FROM AUTHOR]

When implementing unfamiliar programming tasks, developers commonly search code examples and learn usage patterns of APIs from the code examples or reuse them by copy-pasting and modifying. For providing high-quality code examples, previous studies present several methods to recommend code snippets mainly based on information retrieval. In this paper, to provide better recommendation results, we propose ROSF, Recommending code Snippets with multi-aspect Features, a novel method combining both information retrieval and supervised learning. In our method, we recommend Top-K code snippets for a given free-form query based on two stages, i.e., coarse-grained searching and fine-grained re-ranking. First, we generate a code snippet candidate set by searching a code snippet corpus using an information retrieval method. Second, we predict probability values of the code snippets for different relevance scores in the candidate set by the learned prediction model from a training set, re-rank these candidate code snippets according to the probability values, and recommend the final results to developers. We conduct several experiments to evaluate our method in a large-scale corpus containing 921,713 real-world code snippets. The results show that ROSF is an effective method for code snippets recommendation and outperforms the-state-of-the-art methods by 20-41percent in Precision and 13-33 percent in NDCG. [ABSTRACT FROM AUTHOR]

To parse large-scale urban scenes using the supervised methods, a large amount of training data that can account for the vast visual and structural variance of urban environment is necessary. Unfortunately, such training data are mostly obtained by tedious and time-consuming manual work. To overcome the drawback, we propose a semisupervised learning framework that combines the margin, cograph, and label constraints into an objective function for point cloud parsing. Mathematically, the margin constraint is presented to learn a novel distance criterion that can effectively recognize points of different classes. The graph regularization is then employed to characterize the intrinsic geometry structure of the data manifold and explore relationships among points. The label consistency regularization is introduced to ensure the category consistency of the clustered points and single point. To classify the out-of-sample data, the framework successfully transforms the semisupervised classification results into the linear classifier by adopting a linear regression. An iterative algorithm is utilized to efficiently and effectively optimize the objective function with characteristics of multiple variables and highly nonlinear. The point clouds of four urban scenes are used to validate our method. The experimental results show that our method outperforms the state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Feature-based electromagnetic (EM) optimization techniques can help avoid local minima in microwave design. Zeros of the transfer functions are recently used to help extract the features when the features of filter responses are not explicitly identifiable. This letter proposes a feature zero-adaptation approach to enlarge the surrogate range by overcoming the problem of varying orders of the transfer function w.r.t. the changes in design variables. In this way, the proposed technique allows larger step sizes for optimization, therefore, speeding up the overall EM optimization process. During each optimization iteration, parallel techniques are proposed to be used to generate multiple EM geometrical samples simultaneously for creating the feature-based surrogate model. The proposed technique is demonstrated using two microwave filter examples. [ABSTRACT FROM AUTHOR]

Micro-expressions are brief facial movements characterized by short duration, involuntariness and low intensity. Recognition of spontaneous facial micro-expressions is a great challenge. In this paper, we propose a simple yet effective Main Directional Mean Optical-flow (MDMO) feature for micro-expression recognition. We apply a robust optical flow method on micro-expression video clips and partition the facial area into regions of interest (ROIs) based partially on action units. The MDMO is a ROI-based, normalized statistic feature that considers both local statistic motion information and its spatial location. One of the significant characteristics of MDMO is that its feature dimension is small. The length of a MDMO feature vector is $36 \times 2=72$ , where $36$ is the number of ROIs. Furthermore, to reduce the influence of noise due to head movements, we propose an optical-flow-driven method to align all frames of a micro-expression video clip. Finally, a SVM classifier with the proposed MDMO feature is adopted for micro-expression recognition. Experimental results on three spontaneous micro-expression databases, namely SMIC, CASME and CASME II, show that the MDMO can achieve better performance than two state-of-the-art baseline features, i.e., LBP-TOP and HOOF. [ABSTRACT FROM PUBLISHER]

Turn-to-turn short-circuit of stator winding due to deterioration of electrical insulation is one of the most probable faults in motor drive systems. This paper proposes two diagnostic methods to detect the fault based on Support Vector Machine. In the first method, magnitude and phase of load currents at fixed rotating speeds are considered as features reflecting winding condition. It is found that the method is effective particularly in slight insulation failure like one turn-to-turn short-circuit. Classification of winding condition, i.e. specification of number of short-circuit turns, is also achieved by this method. In the second method, only magnitude of load currents is treated as feature aiming at diagnosis available when load current varies. The method also has the advantage of low cost and short data processing time. It is found that two turn-to-turn short-circuit can be diagnosed with practically acceptable accuracy. An appropriate choice of the two methods should be recommended depending on the goal as well as circumstance of diagnosis. [ABSTRACT FROM AUTHOR]

In this paper, a new on-road vehicle detection method is presented. First, a new feature named the Position and Intensity-included Histogram of Oriented Gradients (PIHOG or $\pi$HOG) is proposed. Unlike the conventional HOG, $\pi$HOG compensates the information loss involved in the construction of a histogram with position information, and it improves the discriminative power using intensity information. Second, a new search space reduction (SSR) method is proposed to speed up the detection and reduce the computational load. The SSR additionally decreases the false positive rate. A variety of classifiers, including support vector machine, extreme learning machine, and $k$-nearest neighbor, are used to train and classify vehicles using $\pi$HOG. The validity of the proposed method is demonstrated by its application to Caltech, IR, Pittsburgh, and Kitti datasets. The experimental results demonstrate that the proposed vehicle detection method not only improves detection performance but also reduces computation time. [ABSTRACT FROM PUBLISHER]