scispace - formally typeset
Search or ask a question

Showing papers on "Filter (video) published in 2018"


Proceedings ArticleDOI
29 Jul 2018
TL;DR: This paper proposes a novel CNN architecture, called SincNet, that encourages the first convolutional layer to discover more meaningful filters, based on parametrized sinc functions, which implement band-pass filters.
Abstract: Deep learning is progressively gaining popularity as a viable alternative to i-vectors for speaker recognition. Promising results have been recently obtained with Convolutional Neural Networks (CNNs) when fed by raw speech samples directly. Rather than employing standard hand-crafted features, the latter CNNs learn low-level speech representations from waveforms, potentially allowing the network to better capture important narrow-band speaker characteristics such as pitch and formants. Proper design of the neural network is crucial to achieve this goal.This paper proposes a novel CNN architecture, called SincNet, that encourages the first convolutional layer to discover more meaningful filters. SincNet is based on parametrized sinc functions, which implement band-pass filters. In contrast to standard CNNs, that learn all elements of each filter, only low and high cutoff frequencies are directly learned from data with the proposed method. This offers a very compact and efficient way to derive a customized filter bank specifically tuned for the desired application.Our experiments, conducted on both speaker identification and speaker verification tasks, show that the proposed architecture converges faster and performs better than a standard CNN on raw waveforms.

605 citations


Posted Content
TL;DR: The proposed Soft Filter Pruning (SFP) method enables the pruned filters to be updated when training the model after pruning, which has two advantages over previous works: larger model capacity and less dependence on the pretrained model.
Abstract: This paper proposed a Soft Filter Pruning (SFP) method to accelerate the inference procedure of deep Convolutional Neural Networks (CNNs). Specifically, the proposed SFP enables the pruned filters to be updated when training the model after pruning. SFP has two advantages over previous works: (1) Larger model capacity. Updating previously pruned filters provides our approach with larger optimization space than fixing the filters to zero. Therefore, the network trained by our method has a larger model capacity to learn from the training data. (2) Less dependence on the pre-trained model. Large capacity enables SFP to train from scratch and prune the model simultaneously. In contrast, previous filter pruning methods should be conducted on the basis of the pre-trained model to guarantee their performance. Empirically, SFP from scratch outperforms the previous filter pruning methods. Moreover, our approach has been demonstrated effective for many advanced CNN architectures. Notably, on ILSCRC-2012, SFP reduces more than 42% FLOPs on ResNet-101 with even 0.2% top-5 accuracy improvement, which has advanced the state-of-the-art. Code is publicly available on GitHub: this https URL

518 citations


Proceedings ArticleDOI
01 Jan 2018
TL;DR: A novel Gradient-based method, Soft Taylor Pruning (STP), is proposed to reduce the network complexity in dynamic way by allowing simultaneous pruning on multiple layers by controlling the opening and closing of multiple mask layers.
Abstract: This paper proposed a Soft Filter Pruning (SFP) method to accelerate the inference procedure of deep Convolutional Neural Networks (CNNs). Specifically, the proposed SFP enables the pruned filters to be updated when training the model after pruning. SFP has two advantages over previous works: (1) Larger model capacity. Updating previously pruned filters provides our approach with larger optimization space than fixing the filters to zero. Therefore, the network trained by our method has a larger model capacity to learn from the training data. (2) Less dependence on the pre-trained model. Large capacity enables SFP to train from scratch and prune the model simultaneously. In contrast, previous filter pruning methods should be conducted on the basis of the pre-trained model to guarantee their performance. Empirically, SFP from scratch outperforms the previous filter pruning methods. Moreover, our approach has been demonstrated effective for many advanced CNN architectures. Notably, on ILSCRC-2012, SFP reduces more than 42% FLOPs on ResNet-101 with even 0.2% top-5 accuracy improvement, which has advanced the state-of-the-art. Code is publicly available on GitHub: this https URL

517 citations


Proceedings ArticleDOI
18 Jun 2018
TL;DR: The authors proposed a bank of convolutional filters to capture class-specific discriminative patches without extra part or bounding box annotations, which achieves state-of-the-art performance on three publicly available fine-grained recognition datasets (CUB-200-2011, Stanford Cars and FGVC-Aircraft).
Abstract: Compared to earlier multistage frameworks using CNN features, recent end-to-end deep approaches for fine-grained recognition essentially enhance the mid-level learning capability of CNNs. Previous approaches achieve this by introducing an auxiliary network to infuse localization information into the main classification network, or a sophisticated feature encoding method to capture higher order feature statistics. We show that mid-level representation learning can be enhanced within the CNN framework, by learning a bank of convolutional filters that capture class-specific discriminative patches without extra part or bounding box annotations. Such a filter bank is well structured, properly initialized and discriminatively learned through a novel asymmetric multi-stream architecture with convolutional filter supervision and a non-random layer initialization. Experimental results show that our approach achieves state-of-the-art on three publicly available fine-grained recognition datasets (CUB-200-2011, Stanford Cars and FGVC-Aircraft). Ablation studies and visualizations are provided to understand our approach.

366 citations


Proceedings ArticleDOI
01 Jun 2018
TL;DR: In this article, steerable filter convolutional neural networks (SFCNNs) are proposed to achieve joint equivariance under translations and rotations by design, which achieves state-of-the-art performance on the rotated MNIST benchmark and on the ISBI 2012 2D EM segmentation challenge.
Abstract: In many machine learning tasks it is desirable that a model's prediction transforms in an equivariant way under transformations of its input. Convolutional neural networks (CNNs) implement translational equivariance by construction; for other transformations, however, they are compelled to learn the proper mapping. In this work, we develop Steerable Filter CNNs (SFCNNs) which achieve joint equivariance under translations and rotations by design. The proposed architecture employs steerable filters to efficiently compute orientation dependent responses for many orientations without suffering interpolation artifacts from filter rotation. We utilize group convolutions which guarantee an equivariant mapping. In addition, we generalize He's weight initialization scheme to filters which are defined as a linear combination of a system of atomic filters. Numerical experiments show a substantial enhancement of the sample complexity with a growing number of sampled filter orientations and confirm that the network generalizes learned patterns over orientations. The proposed approach achieves state-of-the-art on the rotated MNIST benchmark and on the ISBI 2012 2D EM segmentation challenge.

320 citations


Proceedings ArticleDOI
18 Jun 2018
TL;DR: Wang et al. as mentioned in this paper proposed an Appearance and Relation Network (ARTNet) to simultaneously model appearance and relation from RGB input in a separate and explicit manner, which decouple the spatiotemporal learning module into an appearance branch for spatial modeling and a relation branch for temporal modeling.
Abstract: Spatiotemporal feature learning in videos is a fundamental problem in computer vision. This paper presents a new architecture, termed as Appearance-and-Relation Network (ARTNet), to learn video representation in an end-to-end manner. ARTNets are constructed by stacking multiple generic building blocks, called as SMART, whose goal is to simultaneously model appearance and relation from RGB input in a separate and explicit manner. Specifically, SMART blocks decouple the spatiotemporal learning module into an appearance branch for spatial modeling and a relation branch for temporal modeling. The appearance branch is implemented based on the linear combination of pixels or filter responses in each frame, while the relation branch is designed based on the multiplicative interactions between pixels or filter responses across multiple frames. We perform experiments on three action recognition benchmarks: Kinetics, UCF101, and HMDB51, demonstrating that SMART blocks obtain an evident improvement over 3D convolutions for spatiotemporal feature learning. Under the same training setting, ARTNets achieve superior performance on these three datasets to the existing state-of-the-art methods.1

308 citations


Journal ArticleDOI
TL;DR: In this article, the authors introduce the channel and spatial reliability concepts to discriminative correlation filters (DCF) and provide a learning algorithm for its efficient and seamless integration in the filter update and the tracking process.
Abstract: Short-term tracking is an open and challenging problem for which discriminative correlation filters (DCF) have shown excellent performance. We introduce the channel and spatial reliability concepts to DCF tracking and provide a learning algorithm for its efficient and seamless integration in the filter update and the tracking process. The spatial reliability map adjusts the filter support to the part of the object suitable for tracking. This both allows to enlarge the search region and improves tracking of non-rectangular objects. Reliability scores reflect channel-wise quality of the learned filters and are used as feature weighting coefficients in localization. Experimentally, with only two simple standard feature sets, HoGs and colornames, the novel CSR-DCF method--DCF with channel and spatial reliability--achieves state-of-the-art results on VOT 2016, VOT 2015 and OTB100. The CSR-DCF runs close to real-time on a CPU.

228 citations


Proceedings ArticleDOI
13 Jul 2018
TL;DR: This paper proposes a novel global & dynamic pruning (GDP) scheme to prune redundant filters for CNN acceleration that achieves superior performance to accelerate several cutting-edge CNNs on the ILSVRC 2012 benchmark.
Abstract: Accelerating convolutional neural networks has recently received ever-increasing research focus. Among various approaches proposed in the literature, filter pruning has been regarded as a promising solution, which is due to its advantage in significant speedup and memory reduction of both network model and intermediate feature maps. To this end, most approaches tend to prune filters in a layerwise fixed manner, which is incapable to dynamically recover the previously removed filter, as well as jointly optimize the pruned network across layers. In this paper, we propose a novel global & dynamic pruning (GDP) scheme to prune redundant filters for CNN acceleration. In particular, GDP first globally prunes the unsalient filters across all layers by proposing a global discriminative function based on prior knowledge of each filter. Second, it dynamically updates the filter saliency all over the pruned sparse network, and then recovers the mistakenly pruned filter, followed by a retraining phase to improve the model accuracy. Specially, we effectively solve the corresponding nonconvex optimization problem of the proposed GDP via stochastic gradient descent with greedy alternative updating. Extensive experiments show that the proposed approach achieves superior performance to accelerate several cutting-edge CNNs on the ILSVRC 2012 benchmark, comparing to the state-of-the-art filter pruning methods.

209 citations


Journal ArticleDOI
TL;DR: Wang et al. as mentioned in this paper proposed a new discriminative correlation filter (DCF) based tracking method with adaptive spatial feature selection and temporal consistent constraints, with which the new tracker enables joint spatial-temporal filter learning.
Abstract: With efficient appearance learning models, Discriminative Correlation Filter (DCF) has been proven to be very successful in recent video object tracking benchmarks and competitions. However, the existing DCF paradigm suffers from two major issues, i.e., spatial boundary effect and temporal filter degradation. To mitigate these challenges, we propose a new DCF-based tracking method. The key innovations of the proposed method include adaptive spatial feature selection and temporal consistent constraints, with which the new tracker enables joint spatial-temporal filter learning in a lower dimensional discriminative manifold. More specifically, we apply structured spatial sparsity constraints to multi-channel filers. Consequently, the process of learning spatial filters can be approximated by the lasso regularisation. To encourage temporal consistency, the filter model is restricted to lie around its historical value and updated locally to preserve the global structure in the manifold. Last, a unified optimisation framework is proposed to jointly select temporal consistency preserving spatial features and learn discriminative filters with the augmented Lagrangian method. Qualitative and quantitative evaluations have been conducted on a number of well-known benchmarking datasets such as OTB2013, OTB50, OTB100, Temple-Colour, UAV123 and VOT2018. The experimental results demonstrate the superiority of the proposed method over the state-of-the-art approaches.

200 citations


Journal ArticleDOI
TL;DR: This paper investigates the problem of event-triggered fault detection (FD) filter design for nonlinear networked systems in the framework of interval type-2 fuzzy systems and proposes an augmented FD system with imperfectly matched MFs, which hampers the stability analysis and FD.
Abstract: This paper investigates the problem of event-triggered fault detection (FD) filter design for nonlinear networked systems in the framework of interval type-2 fuzzy systems. In the system model, the parameter uncertainty is captured effectively by the membership functions (MFs) with upper and lower bounds. For reducing the utilization of limited communication bandwidth, an event-triggered communication mechanism is applied. A novel FD filter subject to event-triggered communication mechanism, data quantization, and communication delay is designed to generate a residual signal and detect system faults, where the premise variables are different from those of the system model. Consequently, the augmented FD system is with imperfectly matched MFs, which hampers the stability analysis and FD. To relax the stability analysis and achieve a better FD performance, the information of MFs and slack matrices are utilized in the stability analysis. Finally, two examples are employed to demonstrate the effectiveness of the proposed scheme.

199 citations


Journal ArticleDOI
TL;DR: The performance of the individual S-Trap filter and 96-well filter plate is evaluated in the context of global protein identification and quantitation using whole-cell lysate and clinically relevant sputum samples to suggest an ultrafast sample-preparation approach for shotgun proteomics.
Abstract: The success of shotgun proteomic analysis depends largely on how samples are prepared. Current approaches (such as those that are gel-, solution-, or filter-based), although being extensively employed in the field, are time-consuming and less effective with respect to the repetitive sample processing, recovery, and overall yield. As an alternative, the suspension trapping (S-Trap) filter has been commercially available very recently in the format of a single or 96-well filter plate. In contrast to the conventional filter-aided sample preparation (FASP) approach, which utilizes a molecular weight cut-off (MWCO) membrane as the filter and requires hours of processing before digestion-ready proteins can be obtained, the S-Trap employs a three-dimensional porous material as filter media and traps particulate protein suspensions with the subsequent depletion of interfering substances and in-filter digestion. Due to the large (submicron) pore size, each centrifugation cycle of the S-Trap filter only takes 1 min, which significantly reduces the total processing time from approximately 3 h by FASP to less than 15 min, suggesting an ultrafast sample-preparation approach for shotgun proteomics. Here, we comprehensively evaluate the performance of the individual S-Trap filter and 96-well filter plate in the context of global protein identification and quantitation using whole-cell lysate and clinically relevant sputum samples.

Journal ArticleDOI
TL;DR: A novel fuzzy finite-time command filtered backstepping approach is proposed by introducing the fuzzy finite -time command filter, designing the new virtual control signals and the modified error compensation signals.
Abstract: This paper considers the fuzzy finite-time tracking control problem for a class of nonlinear systems with input saturation. A novel fuzzy finite-time command filtered backstepping approach is proposed by introducing the fuzzy finite-time command filter, designing the new virtual control signals and the modified error compensation signals. The proposed approach not only holds the advantages of the conventional command-filtered backstepping control, but also guarantees the finite-time convergence. A practical example is included to show the effectiveness of the proposed method.

Proceedings ArticleDOI
24 Apr 2018
TL;DR: This work proposes a novel CF-based optimization problem to jointly model the discrimination and reliability information and introduces a local response consistency regular term to emphasize equal contributions of different regions and avoid the tracker being dominated by unreliable regions.
Abstract: For visual tracking, an ideal filter learned by the correlation filter (CF) method should take both discrimination and reliability information. However, existing attempts usually focus on the former one while pay less attention to reliability learning. This may make the learned filter be dominated by the unexpected salient regions on the feature map, thereby resulting in model degradation. To address this issue, we propose a novel CF-based optimization problem to jointly model the discrimination and reliability information. First, we treat the filter as the element-wise product of a base filter and a reliability term. The base filter is aimed to learn the discrimination information between the target and backgrounds, and the reliability term encourages the final filter to focus on more reliable regions. Second, we introduce a local response consistency regular term to emphasize equal contributions of different regions and avoid the tracker being dominated by unreliable regions. The proposed optimization problem can be solved using the alternating direction method and speeded up in the Fourier domain. We conduct extensive experiments on the OTB-2013, OTB-2015 and VOT-2016 datasets to evaluate the proposed tracker. Experimental results show that our tracker performs favorably against other state-of-the-art trackers.

Journal ArticleDOI
TL;DR: The new solutions improve the FCS-MPC with an active damping algorithm by eliminating low-order grid current harmonics and decreasing sensitivity to grid voltage distortion and extending the prediction horizon in PCi1 i2uc-2steps.
Abstract: This paper presents two new implementations of finite control set model predictive control (FCS-MPC) methods applied to ac–dc converters with an inductive.capacitive.inductive (LCL) filter. The LCL filter, despite its advantages, can cause a strong resonance in the grid current and also pose a substantially more complex control problem. The new solutions improve the FCS-MPC with an active damping algorithm by eliminating low-order grid current harmonics and decreasing sensitivity to grid voltage distortion. The new methods, i.e., PCi1i2uc and PCi1i2uc-2steps propose multivariable approaches using converter-side current, line-side current, and capacitor voltage. Another improvement involves extending the prediction horizon in PCi1 i2uc-2steps. Both methods have been tested and compared in steady and transient states as well as under grid voltage disturbances. The methods were also compared with a predictive control algorithm with active damping. The simulation results and experimental measurements validate the developed control schemes and show high quality of grid current (low THDi value), high dynamic performance, and immunity under distorted grid voltage conditions.

Journal ArticleDOI
TL;DR: Bivalve molluscs have epifaunal or infaunal lifestyles but are largely filter feeders that coup... as mentioned in this paper, but are abundant in marine and freshwater ecosystems and perform important ecological functions.
Abstract: Bivalve molluscs are abundant in marine and freshwater ecosystems and perform important ecological functions. Bivalves have epifaunal or infaunal lifestyles but are largely filter feeders that coup...

Journal ArticleDOI
TL;DR: A teleoperation scheme using integrated tremor attenuation with a variable gain control algorithm involving surface electromyogram is proposed to achieve personalized control performance and to reduce reliance on operator’s skill.
Abstract: Teleoperated robot systems are able to support humans to accomplish their tasks in many applications. However, the performance of teleoperation largely depends on motor functionality and human operator’s skill, especially when a human operator is short of skill training. In order to adapt to various unstructured environments for the robot system and the human operator, in this paper, a teleoperation scheme using integrated tremor attenuation with a variable gain control algorithm involving surface electromyogram is proposed to achieve personalized control performance and to reduce reliance on operator’s skill. For attenuating tremor, a filter based on support vector machine is developed to guarantee normal operation. This filter depends on the machine learning scheme and does not rely on a priori filter parameters. Semiphysical experiments have been performed to demonstrate the effectiveness of the proposed methods.

Proceedings ArticleDOI
10 Jan 2018
TL;DR: In this article, the authors propose the Net2Vec framework, in which semantic concepts are mapped to vectorial embeddings based on corresponding filter responses and show that in most cases, multiple filters are required to code for a concept, and that often filters are not concept specific and help encode multiple concepts.
Abstract: In an effort to understand the meaning of the intermediate representations captured by deep networks, recent papers have tried to associate specific semantic concepts to individual neural network filter responses, where interesting correlations are often found, largely by focusing on extremal filter responses. In this paper, we show that this approach can favor easy-to-interpret cases that are not necessarily representative of the average behavior of a representation. A more realistic but harder-to-study hypothesis is that semantic representations are distributed, and thus filters must be studied in conjunction. In order to investigate this idea while enabling systematic visualization and quantification of multiple filter responses, we introduce the Net2Vec framework, in which semantic concepts are mapped to vectorial embeddings based on corresponding filter responses. By studying such embeddings, we are able to show that 1., in most cases, multiple filters are required to code for a concept, that 2., often filters are not concept specific and help encode multiple concepts, and that 3., compared to single filter activations, filter embeddings are able to better characterize the meaning of a representation and its relationship to other concepts.

Journal ArticleDOI
TL;DR: In this paper, a derivation of the Poisson multi-Bernoulli mixture (PMBM) filter for multi-target tracking with the standard point target measurements without using probability generating functionals or functional derivatives is provided.
Abstract: We provide a derivation of the Poisson multi-Bernoulli mixture (PMBM) filter for multitarget tracking with the standard point target measurements without using probability generating functionals or functional derivatives. We also establish the connection with the $\delta$ -generalized labeled multi-Bernoulli ( $\delta$ -GLMB) filter, showing that a $\delta$ -GLMB density represents a multi-Bernoulli mixture with labeled targets so it can be seen as a special case of PMBM. In addition, we propose an implementation for linear/Gaussian dynamic and measurement models and how to efficiently obtain typical estimators in the literature from the PMBM. The PMBM filter is shown to outperform other filters in the literature in a challenging scenario.

Book ChapterDOI
02 Dec 2018
TL;DR: A computationally efficient, asynchronous filter that continuously fuses image frames and events into a single high-temporal-resolution, high-dynamic-range image state is proposed that outperforms existing state-of-the-art methods.
Abstract: Event cameras provide asynchronous, data-driven measurements of local temporal contrast over a large dynamic range with extremely high temporal resolution. Conventional cameras capture low-frequency reference intensity information. These two sensor modalities provide complementary information. We propose a computationally efficient, asynchronous filter that continuously fuses image frames and events into a single high-temporal-resolution, high-dynamic-range image state. In absence of conventional image frames, the filter can be run on events only. We present experimental results on high-speed, high-dynamic-range sequences, as well as on new ground truth datasets we generate to demonstrate the proposed algorithm outperforms existing state-of-the-art methods.

Journal ArticleDOI
TL;DR: A novel 2-D/3-D symmetry filter that considers local phase features by using a quadrature filter to distinguish between lines and edges, but also uses the weighted geometric mean of the blurred and shifted responses of the quadratures filter, which allows more tolerance of vessels with irregular appearance.
Abstract: Automated detection of vascular structures is of great importance in understanding the mechanism, diagnosis, and treatment of many vascular pathologies. However, automatic vascular detection continues to be an open issue because of difficulties posed by multiple factors, such as poor contrast, inhomogeneous backgrounds, anatomical variations, and the presence of noise during image acquisition. In this paper, we propose a novel 2-D/3-D symmetry filter to tackle these challenging issues for enhancing vessels from different imaging modalities. The proposed filter not only considers local phase features by using a quadrature filter to distinguish between lines and edges, but also uses the weighted geometric mean of the blurred and shifted responses of the quadrature filter, which allows more tolerance of vessels with irregular appearance. As a result, this filter shows a strong response to the vascular features under typical imaging conditions. Results based on eight publicly available datasets (six 2-D data sets, one 3-D data set, and one 3-D synthetic data set) demonstrate its superior performance to other state-of-the-art methods.

Proceedings ArticleDOI
27 May 2018
TL;DR: This work proposes a meta-framework, called Cold Filter (CF), that enables faster and more accurate stream processing, and can accurately estimate both cold and hot items, giving it a genericity that makes it applicable to many stream processing tasks.
Abstract: Approximate stream processing algorithms, such as Count-Min sketch, Space-Saving, etc., support numerous applications in databases, storage systems, networking, and other domains. However, the unbalanced distribution in real data streams poses great challenges to existing algorithms. To enhance these algorithms, we propose a meta-framework, called Cold Filter (CF), that enables faster and more accurate stream processing. Different from existing filters that mainly focus on hot items, our filter captures cold items in the first stage, and hot items in the second stage. Also, existing filters require two-direction communication - with frequent exchanges between the two stages; our filter on the other hand is one-direction - each item enters one stage at most once. Our filter can accurately estimate both cold and hot items, giving it a genericity that makes it applicable to many stream processing tasks. To illustrate the benefits of our filter, we deploy it on three typical stream processing tasks and experimental results show speed improvements of up to 4.7 times, and accuracy improvements of up to 51 times. All source code is made publicly available at Github.

Journal ArticleDOI
TL;DR: A new method to remove salt and pepper noise, which is based on pixel density filter (BPDF), which shows that BPDF produces better results than the above-mentioned methods at low and medium noise density.
Abstract: In this paper, we deliver a new method to remove salt and pepper noise, which we refer to as based on pixel density filter (BPDF) The first step of the method is to determine whether or not a pixel is noisy, and then we decide on an adaptive window size that accepts the noisy pixel as the center The most repetitive noiseless pixel value within the window is set as the new pixel value By using 18 test images, we give the results of peak signal-to-noise ratio (PSNR), structural similarity (SSIM), image enhancement factor (IEF), standard median filter (SMF), adaptive median filter (AMF), adaptive fuzzy filter (AFM), progressive switching median filter (PSMF), decision-based algorithm (DBA), modified decision-based unsymmetrical trimmed median filter (MDBUTMF), noise adaptive fuzzy switching median filter (NAFSM), and BPDF The results show that BPDF produces better results than the above-mentioned methods at low and medium noise density

Journal ArticleDOI
TL;DR: Simulation as well as practical results prove the feasibility of FCS-MPC application in HAPF reactive power control and allow tracking fluctuations and abrupt changes in load reactive power.
Abstract: This paper applies finite control set model predictive control (FCS-MPC) for dynamic reactive power compensation using a hybrid active power filter (HAPF). The FCS-MPC uses a model based on LCL -filter equations to predict the system behavior and optimize the control action. In fact, the application of FCS-MPC in grid-connected converters with LCL -Filter is quite recent. This algorithm is a very promising control technique for power electronics converters and its use for reactive power control of hybrid filter has not been reported in the literature yet. This paper uses the FCS-MPC in a multivariable structure along with an adaptive notch filter to damp resonance. The main purpose is to improve the dynamic response of the HAPF. Simulation as well as practical results prove the feasibility of FCS-MPC application in HAPF reactive power control. The dynamic response of the equipment was significantly improved and represents the main contribution of this paper. As a result, the FCS-MPC allows tracking fluctuations and abrupt changes in load reactive power.

Journal ArticleDOI
TL;DR: A novel stacked CNN model with multiple convolutional layers of decreasing filter sizes is proposed to improve the performance of CNN models with either log-mel feature input or raw waveform input to build the ensemble DS-CNN model for ESC.
Abstract: Convolutional neural networks (CNNs) with log-mel audio representation and CNN-based end-to-end learning have both been used for environmental event sound recognition (ESC). However, log-mel features can be complemented by features learned from the raw audio waveform with an effective fusion method. In this paper, we first propose a novel stacked CNN model with multiple convolutional layers of decreasing filter sizes to improve the performance of CNN models with either log-mel feature input or raw waveform input. These two models are then combined using the Dempster–Shafer (DS) evidence theory to build the ensemble DS-CNN model for ESC. Our experiments over three public datasets showed that our method could achieve much higher performance in environmental sound recognition than other CNN models with the same types of input features. This is achieved by exploiting the complementarity of the model based on log-mel feature input and the model based on learning features directly from raw waveforms.

Journal ArticleDOI
TL;DR: The quantitative and qualitative results of experiments demonstrate that the proposed DDF performs well on contrast enhancement, structure preservation, and noise reduction, and its satisfactory computation time resulting from its simple implementation makes it suitable for extensive application.
Abstract: Enhancement and denoising have always been a pair of conflicting problems in image processing of computer vision. Inspired by an earlier dual-domain filter (DDF), this letter proposes a progressive DDF to simultaneously enhance and denoise low-quality optical remote-sensing images. The main procedure of the proposed enhancement filter has two parts. First, a bilateral filter is exploited as a guide filter to obtain high-contrast images, which are enhanced by a histogram modification method. Then, low-contrast useful structures are restored by a short-time Fourier transform and are enhanced using an adaptive correction parameter. Both the quantitative and qualitative results of experiments on synthetic and real-world low-quality remote-sensing images demonstrate that the proposed method performs well on contrast enhancement, structure preservation, and noise reduction. Moreover, its satisfactory computation time resulting from its simple implementation makes it suitable for extensive application.

Journal ArticleDOI
TL;DR: The maximum correntropy criterion (MCC) is utilized to improve the robust performance instead of traditional minimum mean square error (MMSE) criterion, and a new square-root nonlinear filter is proposed in this study, named as the maximum Correntropysquare-root cubature Kalman filter (MCSCKF).
Abstract: For a nonlinear system, the cubature Kalman filter (CKF) and its square-root version are useful methods to solve the state estimation problems, and both can obtain good performance in Gaussian noises. However, their performances often degrade significantly in the face of non-Gaussian noises, particularly when the measurements are contaminated by some heavy-tailed impulsive noises. By utilizing the maximum correntropy criterion (MCC) to improve the robust performance instead of traditional minimum mean square error (MMSE) criterion, a new square-root nonlinear filter is proposed in this study, named as the maximum correntropy square-root cubature Kalman filter (MCSCKF). The new filter not only retains the advantage of square-root cubature Kalman filter (SCKF), but also exhibits robust performance against heavy-tailed non-Gaussian noises. A judgment condition that avoids numerical problem is also given. The results of two illustrative examples, especially the SINS/GPS integrated systems, demonstrate the desirable performance of the proposed filter.

Journal ArticleDOI
TL;DR: A predictive current control scheme is presented, implemented in addition to common PI current control and used for a three-phase four-wire LCL-filter-based active power filter, where inaccuracies in the model caused by parameter variations are compensated by an additional control structure.
Abstract: Voltage source inverter based active power filters are widely used for the compensation of harmonic currents, especially in industrial environments. Loads such as three-phase diode rectifiers, thyristor-based rectifiers or variable-speed drives can cause relevant harmonics up to the 50th order or even higher, depending on the line filter. Usually, the control of the harmonics is either done in multiple rotating frames, where the harmonics appear as dc values, or by resonant controllers in a stationary frame. At this, the calculation effort is rising with the number of harmonics to compensate. In this paper, a predictive current control scheme is presented. It is implemented in addition to common PI current control and used for a three-phase four-wire LCL -filter-based active power filter. Inaccuracies in the model caused by parameter variations are compensated by an additional control structure. Both structures are related only with low computational effort and are not depending on the desired number of the harmonics to compensate. Experimental tests are demonstrating, that both structures individually, and especially the combination of both, are offering a very good behavior in the steady state and in the case of load changes.

Proceedings Article
03 Dec 2018
TL;DR: In this article, a pre-filter is used to enhance Bloom filters by applying machine learning to determine a function that models the data set the Bloom filter is meant to represent, with the following outcomes: (1) they clarify what guarantees can and cannot be associated with such a structure; (2) they show how to estimate what size the learning function must obtain in order to obtain improved performance; and (3) they provide a simple method, sandwiching, for optimizing learned Bloom filters.
Abstract: Recent work has suggested enhancing Bloom filters by using a pre-filter, based on applying machine learning to determine a function that models the data set the Bloom filter is meant to represent. Here we model such learned Bloom filters, with the following outcomes: (1) we clarify what guarantees can and cannot be associated with such a structure; (2) we show how to estimate what size the learning function must obtain in order to obtain improved performance; (3) we provide a simple method, sandwiching, for optimizing learned Bloom filters; and (4) we propose a design and analysis approach for a learned Bloomier filter, based on our modeling approach.

Journal ArticleDOI
TL;DR: In this paper, a symmetric and asymmetric plasmonic bandpass filter (BPF) topology based on the metal-insulator-metal (MIM) configuration is proposed.

Journal ArticleDOI
TL;DR: The converter current ripple is thoroughly analyzed to generalize the current ripple behavior and find the maximum current ripple for sinusoidal pulse width modulation (PWM) and third-harmonic injection PWM.
Abstract: This paper proposes a comprehensive analytical LCL filter design method for three-phase two-level power factor correction rectifiers (PFCs) The high-frequency converter current ripple generates the high-frequency current harmonics that need to be attenuated with respect to the grid standards Studying the high-frequency current of each element proposes a noniterative solution for designing an LCL filter In this paper, the converter current ripple is thoroughly analyzed to generalize the current ripple behavior and find the maximum current ripple for sinusoidal pulse width modulation (PWM) and third-harmonic injection PWM Consequently, the current ripple is used to accurately determine the required filter capacitance based on the maximum charge of the filter capacitor To choose the grid-side inductance, two methods are investigated First method uses the structure of the damping to express the grid-side filter inductance as a function of the converter current ripple Reducing the power loss in the filter and optimizing the grid-side filter inductance is the main focus of the second method which is achieved by employing line impedance stabilization network (LISN) Accordingly, two LCL filters are designed for a 5 kW silicon-carbide-based three-phase PFC Various experimental scenarios are performed to verify the filters attenuation and performance