Filter Network Research Articles

Salt-and-pepper Noise Reduction for Medical Images Based on Image Fusion.

During the collection process, the prostate capsule is prone to introduce salt and pepper noise due to gastrointestinal peristalsis, which will affect the precision of subsequent object detection. A cascade optimization scheme for image denoising based on image fusion was proposed to improve the peak signal-to-noise ratio(PSNR) and contour protection performance of heterogeneous medical images after image denoising. Anisotropic diffusion fusion (ADF) was used to decompose the images denoised by adaptive median filter, non-local adaptive median filter and artificial neural network to generate the base layer and detail layer, which were fused by weighted average and Karhunen-Loeve Transform respectively. Finally, the image was reconstructed by linear superposition. Compared with the traditional denoising method, the image denoised by this method has a higher PSNR while maintaining the image edge contour. Using the denoised dataset for object detection, the detection precision of the obtained model is higher.

A Lumped-Element Directional Coupler for Bandwidth Enhancement, Impedance Matching, and Harmonic Suppressions

This paper presents a lumped-element wideband directional coupler that enables arbitrary impedance matching and harmonic suppressions from 2f0. The coupler consists of four filtering networks (FNs) and an asymmetric branch-line hybrid serving as the matching network. The use of lumped-element topology helps reduce circuit size and improve harmonic suppressions. Each FN, composed of two lumped-element resonators, contributes two additional transmission poles, enabling bandwidth scaling, bandwidth enhancement, and harmonic suppressions. To demonstrate the effectiveness of the proposed design, a wideband coupler with various terminal impedances operating at 0.5 GHz is designed, simulated, and fabricated. The measured size is 0.039×0.039 λg2, the fractional bandwidth for –14.3 dB return loss is 46%, and the harmonic suppression is more than –30 dB from 0.69 to 4.5 GHz.

Echo cancellation using Artificial Intelligence technique

To improve the audio quality in telecommunication, echo cancellation was proposed to prevent signals from re-appearing. Generally speaking, the echo canceler first recognizes the original transmitted signal that appears again due to delay in the transmitted or recieved signal. Then, the echo canceler subtracts the echo in the signals. Commonly, echo cancelers are calledacoustic echo cancellation (AEC) in general, whether they aim for acoustic echo, line echo, or both. However, in modern times, new problems have been emerging which require more innovative, more advanced methods to tackle the echo issue. Thus, the paper reviews methods regarding Artificial Intelligence (AI) to improve the current means for echo cancellation. In this paper, we mainly discuss the approaches to echo cancellation, the identification of echo using adaptive AI and the usage of adaptive filters and neural network for echo cancellation. For instance, deep learning approaches can facilitate the issues of AEC (active echo control) and ANC (active noise control),the optimization of LMS algorithm can be achieved by combining Swarm Intelligence (SI), combining adaptive digital filter and recurrent neural network for acoustic echo cancellation etc. Using AI in echo cancellation can help us achieve high-quality, full-duplex telecommunication in modern telephony system.

BGF-Net: Boundary guided filter network for medical image segmentation

How to fuse low-level and high-level features effectively is crucial to improving the accuracy of medical image segmentation. Most CNN-based segmentation models on this topic usually adopt attention mechanisms to achieve the fusion of different level features, but they have not effectively utilized the guided information of high-level features, which is often highly beneficial to improve the performance of the segmentation model, to guide the extraction of low-level features. To address this problem, we design multiple guided modules and develop a boundary-guided filter network (BGF-Net) to obtain more accurate medical image segmentation. To the best of our knowledge, this is the first time that boundary guided information is introduced into the medical image segmentation task. Specifically, we first propose a simple yet effective channel boundary guided module to make the segmentation model pay more attention to the relevant channel weights. We further design a novel spatial boundary guided module to complement the channel boundary guided module and aware of the most important spatial positions. Finally, we propose a boundary guided filter to preserve the structural information from the previous feature map and guide the model to learn more important feature information. Moreover, we conduct extensive experiments on skin lesion, polyp, and gland segmentation datasets including ISIC 2016, CVC-EndoSceneStil and GlaS to test the proposed BGF-Net. The experimental results demonstrate that BGF-Net performs better than other state-of-the-art methods.

An Integrated Detection and Multi-Object Tracking Pipeline for Satellite Video Analysis of Maritime and Aerial Objects

Optical remote sensing videos, as a new source of remote sensing data that has emerged in recent years, have significant potential in remote sensing applications, especially national defense. In this paper, a tracking pipeline named TDNet (tracking while detecting based on a neural network) is proposed for optical remote sensing videos based on a correlation filter and deep neural networks. The pipeline is used to simultaneously track ships and planes in videos. There are many target tracking methods for general video data, but they suffer some difficulties in remote sensing videos with low resolution and those influenced by weather conditions. The tracked targets are usually misty. Therefore, in TDNet, we propose a new multi-target tracking method called MT-KCF and a detecting-assisted tracking (i.e., DAT) module to improve tracking accuracy and precision. Meanwhile, we also design a new target recognition (i.e., NTR) module to recognise newly emerged targets. In order to verify the performance of TDNet, we compare our method with several state-of-the-art tracking methods on optical video remote sensing data sets acquired from the Jilin No. 1 satellite. The experimental results demonstrate the effectiveness and the state-of-the-art performance of the proposed method. The proposed method can achieve more than 90% performance in terms of precision for single-target tracking tasks and more than 85% performance in terms of MOTA for multi-object tracking tasks.

Self-supervised context-aware correlation filter for robust landmark tracking in liver ultrasound sequences.

Respiratory motion-induced displacement of internal organs poses a significant challenge in image-guided radiation therapy, particularly affecting liver landmark tracking accuracy. Addressing this concern, we propose a self-supervised method for robust landmark tracking in long liver ultrasound sequences. Our approach leverages a Siamese-based context-aware correlation filter network, trained by using the consistency loss between forward tracking and back verification. By effectively utilizing both labeled and unlabeled liver ultrasound images, our model, Siam-CCF , mitigates the impact of speckle noise and artifacts on ultrasonic image tracking by a context-aware correlation filter. Additionally, a fusion strategy for template patch feature helps the tracker to obtain rich appearance information around the point-landmark. Siam-CCF achieves a mean tracking error of 0.79±0.83 mm at a frame rate of 118.6 fps, exhibiting a superior speed-accuracy trade-off on the public MICCAI 2015 Challenge on Liver Ultrasound Tracking (CLUST2015) 2D dataset. This performance won the 5th place on the CLUST2015 2D point-landmark tracking task. Extensive experiments validate the effectiveness of our proposed approach, establishing it as one of the top-performing techniques on the CLUST2015 online leaderboard at the time of this submission.

A 0.5 V Nanowatt Biquadratic Low-Pass Filter with Tunable Quality Factor for Electronic Cochlea Applications

A novel implementation of an analogue low-power, second-order, low-pass filter with tunable quality factor (Q) is presented and discussed. The filter feature is a relatively simple, buffer-based, circuit network consisting of eleven transistors operating in a subthreshold region. Q tuning is accomplished by injecting direct current into a network node, which changes the output resistance of the transistors and, as a result, modifies the filter network’s loss, and thus its Q. Q tuning is independent of a filter cut-off frequency (ω0). The filter, with a nominal ω0 of 1 kHz, was fabricated using a 0.18 µm CMOS technology, and features a Q range of 2–11, power consumption of up to 52 nW, and a 59 dB dynamic range when using a 0.5 V supply. The ω0 can be tuned from 0.5 to 2.5 kHz using a traditional method by changing the transistor transconductances, but this process partially affects the quality factor.

Robust state of charge estimation of LiFePO4 batteries based on Sage_Husa adaptive Kalman filter and dynamic neural network

Robust and accurate state of charge estimation for LiFePO4 batteries has tremendous significance, since they are constituted the basis of reliable operation for battery management system. However, the nature flat voltage curve poses challenges in both the accuracy and robustness of state of charge estimation. Here, we present a state of charge estimation framework for LiFePO4 batteries with integrating physics calculation, adaptive Kalman filter, and machine learning, where nonlinearity of flat voltage curve can be captured with strong robustness by optimized machine learning. To address the parameter sensitivity problem for dynamic neural network, the sine cosine algorithm is adopted to obtain the optimal parameters. Then, the offline training model is established based on nonlinear autoregressive models with exogenous inputs and sine cosine algorithm, and the adaptive Kalman filter is selected for online estimation. The observed data from 26,650 and 18,650 LiFePO4 batteries at 25 and 40 °C has been used to verify the accuracy and robustness of the proposed method under four working conditions. Compared to the existing methods, the proposed method can achieve strong robustness and higher accuracy in state of charge estimation with a maximum relative error below 2 %.

TFCSRec: Time–frequency consistency based contrastive learning for sequential recommendation

Sequential recommendation aims to predict future user interactions by analyzing dynamic patterns within their historical behavior sequences. Deep neural networks have recently become popular for learning representations of these sequences in the time domain. However, representing users’ intentions in the time domain faces challenges such as noise in interactions and sparsity of data. Contrastive learning and representation learning in the frequency domain can mitigate these issues from different perspectives. In this paper, to fully integrate time-domain sequence representations, frequency-domain sequence representations and contrastive learning based on them, we propose a model called Time–Frequency Consistency based contrastive learning for Sequential Recommendation (TFCSRec). TFCSRec utilizes a time-domain encoder with a fully connected network and a filter network to extract high-order features and catch pure sequential patterns. Then, a learnable frequency-domain encoder with a recurrent neural network is designed to capture sequential characteristics in the frequency-domain space. Finally, TFCSRec combines a recommendation task and two contrastive learning tasks to optimize the two user representation encoders. Its contrastive learning is designed to minimize a contrastive regularization loss and a time–frequency consistency loss, which for the first time is constructed directly on the time-domain sequence representation and the frequency-domain sequence representation. Experiments on five benchmark datasets show that the proposed TFCSRec model outperforms other sequential recommendation models based on deep neural networks.

Convolution Filter Compression via Sparse Linear Combinations of Quantized Basis.

Convolutional neural networks (CNNs) have achieved significant performance on various real-life tasks. However, the large number of parameters in convolutional layers requires huge storage and computation resources, making it challenging to deploy CNNs on memory-constrained embedded devices. In this article, we propose a novel compression method that generates the convolution filters in each layer using a set of learnable low-dimensional quantized filter bases. The proposed method reconstructs the convolution filters by stacking the linear combinations of these filter bases. By using quantized values in weights, the compact filters can be represented using fewer bits so that the network can be highly compressed. Furthermore, we explore the sparsity of coefficients through L1 -ball projection when conducting linear combination to further reduce the storage consumption and prevent overfitting. We also provide a detailed analysis of the compression performance of the proposed method. Evaluations of image classification and object detection tasks using various network structures demonstrate that the proposed method achieves a higher compression ratio with comparable accuracy compared with the existing state-of-the-art filter decomposition and network quantization methods.

A Novel Proportional Multi-Resonant Current Controller Strategy for Reduced DC Voltage fed D-STATCOM with Internal LCL Resonance Damping

Abstract This work focuses on a new topology-control-based D-STATCOM solution with reduced DC bus voltage requirement and with an excellent grid side performance. The proposed solution consists of a main inverter and auxiliary inverter along with a transformer and LCL filter network to achieve the required DC bus reduction. A new controller structure with two proportional-multi resonant controller for the converters with only one of the inductors current as a controlled variable ensures the active damping of the LCL resonance. The power circuit configuration assists the controller to generate a difference in the modulation signal due to non-equal gains in two controllers and helps to achieve the resonance damping without capacitor current sensor. Hence, the corresponding capacitor current sensor can be eliminated. The converter operates for any point of common coupling (PCC) loading conditions and the performance of the controller is immune to the grid impedance variation. A detailed stability study is carried out for the proposed controller. The proposed controller can achieve a very fast dynamic response with an excellent stability margin. The proposed solution is verified through simulation studies and through a scaled-down experimental prototype.

Multi-level filter network for low-overlap point cloud registration

Multi-level filter network for low-overlap point cloud registration

Adaptive Blind Super-Resolution Network for Spatial-Specific and Spatial-Agnostic Degradations.

Prior methodologies have disregarded the diversities among distinct degradation types during image reconstruction, employing a uniform network model to handle multiple deteriorations. Nevertheless, we discover that prevalent degradation modalities, including sampling, blurring, and noise, can be roughly categorized into two classes. We classify the first class as spatial-agnostic dominant degradations, less affected by regional changes in image space, such as downsampling and noise degradation. The second class degradation type is intimately associated with the spatial position of the image, such as blurring, and we identify them as spatial-specific dominant degradations. We introduce a dynamic filter network integrating global and local branches to address these two degradation types. This network can greatly alleviate the practical degradation problem. Specifically, the global dynamic filtering layer can perceive the spatial-agnostic dominant degradation in different images by applying weights generated by the attention mechanism to multiple parallel standard convolution kernels, enhancing the network's representation ability. Meanwhile, the local dynamic filtering layer converts feature maps of the image into a spatially specific dynamic filtering operator, which performs spatially specific convolution operations on the image features to handle spatial-specific dominant degradations. By effectively integrating both global and local dynamic filtering operators, our proposed method outperforms state-of-the-art blind super-resolution algorithms in both synthetic and real image datasets.

Multiscale Progressive Fusion Filter Network for Infrared Small Target Detection

Multiscale Progressive Fusion Filter Network for Infrared Small Target Detection

Towards fair and personalized federated recommendation

Recommender systems have gained immense popularity in recent years for predicting users’ interests by learning embeddings. The majority of existing recommendation approaches, represented by graph neural network-based recommendation algorithms, rely on centralized storage of user-item graphs for model learning, which raises privacy issues in the process of collecting and sharing user data. Federated recommendation can mitigate privacy concerns by preventing the server from collecting sensitive data from clients while there still exist unfairness and personalization issues. To address these challenges, we propose a novel framework named Fair and Personalized Federated Recommendation (FPFR). On the client-side, the soft attention mechanism is designed to learn the representation of user/item by combining interaction and attribute information, and the filter network is combined to better characterize user preferences. On the server-side, we cluster users into different groups and learn personalized models for each user. Then, we select representative users from each group to participate in the global model parameters update. Finally, the fairness of federated recommendation is implemented by adding the fairness constraint to recommendation loss. We conduct experiments on five real-world recommendation datasets, and the results demonstrate that the proposed FPFR not only balances group fairness and recommendation accuracy but also improves personalization.

Semantic-Layout-Guided Image Synthesis for High-Quality Synthetic-Aperature Radar Detection Sample Generation

With the widespread application and functional complexity of deep neural networks (DNNs), the demand for training samples is increasing. This elevated requirement also extends to DNN-based SAR object detection. Most public SAR object detection datasets are oriented to marine targets such as ships, while data sets oriented to land targets are relatively rare, though they are an effective way to improve the land object detection capability of deep models through SAR sample generation. In this paper, a synthesis generation collaborative SAR sample augmentation framework is proposed to achieve flexible and diverse high-quality sample augmentation. First, a semantic-layout-guided image synthesis strategy is proposed to generate diverse detection samples. The issues of object location rationality and object layout diversity are also addressed. Meanwhile, a pix2pixGAN network guided by layout maps is utilized to achieve diverse background augmentation. Second, a progressive training strategy of diffusion models is proposed to achieve semantically controllable SAR sample generation to further improve the diversity of scene clutter. Finally, a sample cleaning method considering distribution migration and network filtering is employed to further improve the quality of detection samples. The experimental results show that this semantic synthesis generation method can outperform existing sample augmentation methods, leading to a comprehensive improvement in the accuracy metrics of classical detection networks.

Low-Frequency Current Ripple Minimization of Single-Star Bridge Cells-Based Battery Energy Storage System Using Optimal Third Harmonic Voltage Injection

Single-star bridge cell (SSBC) based multilevel converters are a promising solution for constructing high-voltage and large-capacity battery energy storage systems (BESSs) in power systems. Nevertheless, an undesirable second harmonic current (SHC) will be generated in the battery units since the single-phase instantaneous power of the SSBC pulsates at twice the line frequency. Such SHC-dominated low-frequency current ripple not only shortens the lifespan of the BUs but also degrades the system's efficiency. However, this intrinsic issue is overlooked in recent studies. To this end, an SHC suppression scheme that makes full use of the residual modulation index of the system to inject an optimal third harmonic voltage (THV) is proposed. The phase of the THV is determined first, followed by a closed-loop control strategy to regulate its magnitude dynamically for achieving a minimum SHC within the modulation constraints. The scheme is implemented with the variables that already existed in the main control system, resulting in a low increase in cost. The adverse impact of the technique on the AC-side output characteristics is investigated quantitatively, and a modified design procedure for the DC filter network is presented to address this issue. Hardware-in-the-loop experiments are carried out to validate the technique.

Pairing grid-forming VSC filter topologies with voltage control structures

This paper focuses on the voltage control of grid-forming Voltage Source Converters (VSCs) connected to AC networks. The work aims to identify the best pairing between voltage control structures and usual ac filter topologies. The two filter topologies considered are the inductive (L) and the inductive–capacitive (LC) filter. The control structures studied are the Cascaded Control (CC, cascaded voltage and current loop) and the direct Voltage Control (VC, without current loop). To select the best control-filter pair, the following methodology is suggested. First, for each pair, a voltage control parameter sweeping methodology is developed to find suitable designs considering the following limits: minimum bandwidth, maximum converter voltage, point of connection voltage deviation and maximum current. Second, this action is repeated for different filter parameters and network Short-Circuit Ratios (SCRs). Finally, based of the obtained results, the most relevant control-filter pairs can be selected, prioritizing them according to the fulfilment of the defined limits. To conclude, time-domain simulations are carried out to validate pairing selection.

Reconfigurable bandpass filter with flexible tuning of both center frequency and bandwidth

This paper proposes a novel reconfigurable filter that can flexibly tune both the center frequency and bandwidth simultaneously. This reconfigurable filter consists of a filter network with Chebyshev properties and a parallel resonator. These two structures are symmetrical and can be analyzed using the odd-even mode analysis method for the filtering network. With the addition of a tuning factor, the Chebyshev filter network provides a shift in passband and coupling coefficients that can be adjusted to maintain a stable bandwidth; the parallel resonators will produce adjustable transmission zeros on either side of the passband, providing good out-of-band rejection when adjusting the bandwidth and center frequency. By using the transformation of the resonant frequency of two filtering networks, filters of different orders can also be achieved. This reconfigurable filter can achieve continuous bandwidth and center frequency adjustment and can maintain a constant absolute bandwidth during the center frequency adjustment process, with flexible tuning methods. Through simulation and experiments, it has been proven that this reconfigurable filter has a tunable frequency coverage range of 0.4GHz to 2GHz, with a large tunable range.

Encryption blind spots: shining a light on cloud network traffic

Monitoring for and detecting malicious activity on your networks is being made harder by the increasing volume of encrypted traffic. This encryption exists for good reasons, particularly when cloud services are in play, but threat actors are exploiting its inscrutability to gain access to networks and loiter for extended periods. However, extended Berkeley Packet Filter (eBPF) network analysis technology offers the possibility to gain full visibility into encrypted traffic across VM or container workloads.