Iterative Filtering Research Articles

Magnitude and Similarity Based Variable Rate Filter Pruning for Efficient Convolution Neural Networks

The superior performance of the recent deep learning models comes at the cost of a significant increase in computational complexity, memory use, and power consumption. Filter pruning is one of the effective neural network compression techniques suitable for model deployment in modern low-power edge devices. In this paper, we propose a loss-aware filter Magnitude and Similarity based Variable rate Filter Pruning (MSVFP) technique. We studied several filter selection criteria based on filter magnitude and similarity among filters within a convolution layer, and based on the assumption that the sensitivity of each layer throughout the network is different, unlike conventional fixed rate pruning methods, our algorithm using loss-aware filter selection criteria automatically finds the suitable pruning rate for each layer throughout the network. In addition, the proposed algorithm adapts two different filter selection criteria to remove weak filters as well as redundant filters based on filter magnitude and filter similarity score respectively. Finally, the iterative filter pruning and retraining approach are used to maintain the accuracy of the network during pruning to its target float point operations (FLOPs) reduction rate. In the proposed algorithm, a small number of retraining steps are sufficient during iterative pruning to prevent an abrupt drop in the accuracy of the network. Experiments with commonly used VGGNet and ResNet models on CIFAR-10 and ImageNet benchmark show the superiority of the proposed method over the existing methods in the literature. Notably, VGG-16, ResNet-56, and ResNet-110 models on the CIFAR-10 dataset even improved the original accuracy with more than 50% reduction in network FLOPs. Additionally, the ResNet-50 model on the ImageNet dataset reduces model FLOPs by more than 42% with a negligible drop in the original accuracy.

OMC-SLIO: Online Multiple Calibrations Spinning LiDAR Inertial Odometry.

Light detection and ranging (LiDAR) is often combined with an inertial measurement unit (IMU) to get the LiDAR inertial odometry (LIO) for robot localization and mapping. In order to apply LIO efficiently and non-specialistically, self-calibration LIO is a hot research topic in the related community. Spinning LiDAR (SLiDAR), which uses an additional rotating mechanism to spin a common LiDAR and scan the surrounding environment, achieves a large field of view (FoV) with low cost. Unlike common LiDAR, in addition to the calibration between the IMU and the LiDAR, the self-calibration odometer for SLiDAR must also consider the mechanism calibration between the rotating mechanism and the LiDAR. However, existing self-calibration LIO methods require the LiDAR to be rigidly attached to the IMU and do not take the mechanism calibration into account, which cannot be applied to the SLiDAR. In this paper, we propose firstly a novel self-calibration odometry scheme for SLiDAR, named the online multiple calibration inertial odometer (OMC-SLIO) method, which allows online estimation of multiple extrinsic parameters among the LiDAR, rotating mechanism and IMU, as well as the odometer state. Specially, considering that the rotating and static parts of the motor encoder inside the SLiDAR are rigidly connected to the LiDAR and IMU respectively, we formulate the calibration within the SLiDAR as two separate sets of calibrations: the mechanism calibration between the LiDAR and the rotating part of the motor encoder and the sensor calibration between the static part of the motor encoder and the IMU. Based on such a SLiDAR calibration formulation, we can construct a well-defined kinematic model from the LiDAR to the IMU with the angular information from the motor encoder. Based on the kinematic model, a two-stage motion compensation method is presented to eliminate the point cloud distortion resulting from LiDAR spinning and platform motion. Furthermore, the mechanism and sensor calibration as well as the odometer state are wrapped in a measurement model and estimated via an error-state iterative extended Kalman filter (ESIEKF). Experimental results show that our OMC-SLIO is effective and attains excellent performance.

Quaternion-Based Iterative Extended Kalman Filter for Sensor Fusion of Vision Sensor and IMU in 6-DOF Displacement Monitoring

As civil structures are exposed to various external loads, their periodic evaluation is paramount to ensure their safety. By estimating the 6-degree-of-freedom (DOF) displacement of structures, structural behavior can be monitored directly. Therefore, this study aims to develop a translational and rotational displacement estimation method by fusing a vision sensor and inertial measurement unit (IMU) using a quaternion-based iterative extended Kalman filter (QIEKF). The QIEKF algorithm was applied to reduce the nonlinear influence on the measurement model. The 6-DOF displacement is predicted using the integral of the gyroscope output and updated via a combination of an accelerometer and a magnetometer through a vector matching process in the Kalman filter framework. Subsequently, the 6-DOF displacement estimation result is updated through a vision sensor using a 2-D planar marker and homography transformation in the Kalman filter framework. The performance of the proposed sensor fusion method was verified with experiments using a motorized motion stage, and the results show that the displacements can be estimated with high accuracy regardless of measurement noise and slowly varying signal drift.

State of Charge Estimation of Lithium-Ion Batteries Based on an Adaptive Iterative Extended Kalman Filter for AUVs

As a power source for autonomous underwater vehicles (AUVs), lithium-ion batteries play an important role in ensuring AUVs' electric power propulsion performance. An accurate state of charge (SOC) estimation method is the key to achieving energy optimization for lithium-ion batteries. Due to the complicated ocean environments, traditional filtering methods cannot effectively estimate the SOC of lithium-ion batteries in an AUV. Based on the standard extended Kalman filter (EKF), an adaptive iterative extended Kalman filter (AIEKF) method for the SOC in an AUV is proposed to address the traditional filter's problems, such as low accuracy and large errors. In this method, the adaptive update is introduced to deal with the uncertain noise from the lithium-ion battery. The iteration is used to improve the convergence speed and to reduce the computational burden. Compared with the EKF, iterative extended Kalman filter (IEKF) and adaptive extended Kalman filter (AEKF), the proposed AIEKF has a higher estimation accuracy and anti-interference capability, which is suitable for the AUV's SOC estimation. In addition, based on the second-order equivalent circuit model of the lithium-ion battery, a forgetting factor recursive least squares (FFRLS) method is proposed to deal with the multi-variability problem. In the end, four different methods, including EKF, IEKF, AEKF, and the proposed AIEKF, are compared in computational time. The experiment results show that the proposed method has high accuracy and fast estimation speed, meaning that it has good application potential in AUVs.

Impulse feature extraction via combining a novel voting index and a variational model penalized by center frequency constraint

A key step in rotating machinery fault diagnosis is extracting the subcomponent that best characterizes the fault features. To this end, a fault feature extraction methodology based on an adaptive spectrum segmentation method, a new voting index, and a new variational model is constructed in this paper. More specifically, the spectrum is divided into a series of sub-bands through a fast iteration filter by changing the width of the window. Then, by considering the amplitudes and locations of the fault feature spectral lines in squared envelope spectrum, a novel index based on voting mechanism is constructed to evaluate the periodic impulses in each sub-band. Finally, a new variational model penalized by a center frequency penalty item is proposed to extract the fault characteristics from the raw vibration signal by maximining the proposed index iteratively. The proposed methodology is applied to extract the fault characteristics from rolling bearing and planetary gearbox signals. The comparative results indicate the proposed method is superior to fast kurtogram, fast entrogram, and the proposed fault feature extraction methods based on other classical indices.

CONFIRMATORY FACTOR ANALYSIS OF ORGANIZATIONAL EFFICINECY IN PUBLIC SECTOR: STRUCTURAL EQUATION MODEL FOR FIVE LOCAL GOVERNMENTS IN REPUBLIC OF SRPSKA

In this paper idea is to use the Structure Equation Model approach to the model organizational structure of local governments. The hypothesis is that a complex hierarchical model can be used for improving efficiency in the public sector. The final model with satisfactory parameters showed that this is possible. Process assumed identification of an optimal set of questions as directly measured variables. Latent variables are indirectly measured by a certain set of directly measured variables. The iterative approach filters the optimal set of the directly measured variable and the optimal setting of relations between latent variables. Latent variables are organized in two levels. Certain hierarchical seating is assumed according to the relevant literature. This is confirmed with the applied structural equation model that is further used for the confirmatory factor analysis. Measured variables that are determined according to the questioner are used to extract those factors, where relations between those factors have been analyzed. Fact that organizational cohesiveness has the highest loading of 0.881, gives a conditional conclusion that for group management as whole, this is the key factor.

Joint application of hybrid iterative reconstruction and adaptive filters on neck-and-shoulder CT imaging: A clinical evaluation.

To assess whether the joint application of hybrid iterative reconstruction (HIR) and an adaptive filter (AF) could reduce streak artifacts and improve image quality of neck-and-shoulder computed tomography (CT). This study included 96 patients with suspicious neck lesions who underwent a routine nonenhanced scan on a 64-slice CT scanner. The raw data were reconstructed using four different settings: filtered back projection (FBP), HIR, FBP+AF, and HIR+AF. Regions of interest were manually drawn in erector spine, axillary fat, latissimus dorsi, and dorsal cervical fat. Mean and standard deviation (SD) of the CT number, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were obtained and compared using Wilcoxon signed-rank tests. The qualitative assessments of five factors were compared by two independent investigators. Compared to the other three settings, HIR+AF reduced noise in the area where the streak artifact of the lower neck were most serious (SD; all p≤0.001). The SNR and CNR were improved significantly (all p≤0.001). Compared to the other three settings, HIR+AF showed a significant improvement in CT image quality regarding the visibility of suspicious lesions, the extent of streaking artifacts, noise, soft-tissue contrast, and visualization of small structures (all p≤0.02). The combination of HIR and AF can significantly reduce streaking artifacts and improve image quality in neck-and-shoulder CT imaging.

Impact of Iterative Bilateral Filtering on the Noise Power Spectrum of Computed Tomography Images

A bilateral filter is a non-linear denoising algorithm that can reduce noise while preserving the edges. This study explores the characteristics of a bilateral filter in changing the noise and texture within computed tomography (CT) images in an iterative implementation. We collected images of a homogeneous Neusoft phantom scanned with tube currents of 77, 154, and 231 mAs. The images for each tube current were filtered five times with a configuration of sigma space (σd) = 2 pixels, sigma intensity (σr) = noise level, and a kernel of 5 × 5 pixels. To observe the noise texture in each filter iteration, the noise power spectrum (NPS) was obtained for the five slices of each dataset and averaged to generate a stable curve. The modulation-transfer function (MTF) was also measured from the original and the filtered images. Tests on an anthropomorphic phantom image were carried out to observe their impact on clinical scenarios. Noise measurements and visual observations of edge sharpness were performed on this image. Our results showed that the bilateral filter was effective in suppressing noise at high frequencies, which is confirmed by the sloping NPS curve for different tube currents. The peak frequency was shifted from about 0.2 to about 0.1 mm−1 for all tube currents, and the noise magnitude was reduced by more than 50% compared to the original images. The spatial resolution does not change with the number of iterations of the filter, which is confirmed by the constant values of MTF50 and MTF10. The test results on the anthropomorphic phantom image show a similar pattern, with noise reduced by up to 60% and object edges remaining sharp.

Autonomous navigation for lunar final approach based on gravity gradient measurements

Lunar final approach navigation is critical for pin-point lunar landing in future missions. This study investigates the use of lunar gravity gradient measurements for autonomous navigation of a lunar probe during the final approach phase. As the spacecraft approaches the Moon, the strength of gravity gradient signals improves. A spaceborne gravity gradiometer can precisely measure local gravity gradients, and the latest lunar gravity model GL1500E is used to provide reference values. The employed truncation degree and order of the gravity model are increased stepwise considering the decreasing altitude of the spacecraft in order to reach a compromise between computational costs and model accuracy. An iterative Kalman filter is developed for coupled orbit and attitude estimation using gravity gradient measurements and attitude quaternions obtained from star sensors. A simulated spacecraft with a gradiometer noise level of 0.01 E is considered. Simulation results show that the spacecraft’s position converges rapidly and achieves an accuracy of less than 100 m at the last epoch.

Anderson Acceleration for on-Manifold Iterated Error State Kalman Filters

Iterated Extended Kalman Filter is a promising and widely-used estimator for real-time localization applications. It iterates the observation equation to find a better linearization point and, simultaneously, only maintains the state estimation in a single time to save the computation resources. Inspired by the recent development of the iterative closest point algorithm, this letter investigates an acceleration approach to the iterations in iterative error state Kalman filters (IESKFs). We show that the IESKF can be seen as a fixed point problem, and the Anderson acceleration (AA) can be elegantly applied to the iterations of IESKF since the error state naturally lies in the tangent space and does not require additional transforms. However, the tangent space of the current estimation may change during the iterations, so we should switch the tangent space to the starting point to perform the Anderson acceleration. We propose the AA-IEKF and apply it to the lidar-inertial odometry (LIO) systems to estimate the ego motion of a lidar. The experiments show that the Anderson acceleration can efficiently reduce the number of iterations in ESKF and achieve a lower computational cost.

UWB/INS-based robust anchor-free relative positioning scheme for UGVs

With the development of new distributed cluster applications such as distributed radar and UGV cooperation, the concept of relative positioning springs up—different from the classic positioning application, relative positioning focuses on the real-time determination of relative position relationships in multi-node networks without pre-deployment anchor, prior map and GNSS. Under such application conditions, relative localization also requires high real-time performance, and high accuracy in short baseline scenes, and can solve the positioning failure caused by occlusion and other factors. A high-precision relative positioning scheme based on UWB-INS fusion is proposed in this paper: a robust relative positioning algorithm based on iterative extended Kalman filter is optimized by modeling the system error state, and the adaptive residual elimination algorithm is introduced to ensure the robustness of positioning service. A series of experiments show that the scheme proposed in this paper can provide centimeter-level real-time positioning service for multi-node dynamic networks without anchors and maps, and is expected to become a potential effective relative positioning technology scheme.

Estimating and forecasting the burden and spread of Colombia’s SARS-CoV2 first wave

Following the rapid dissemination of COVID-19 cases in Colombia in 2020, large-scale non-pharmaceutical interventions (NPIs) were implemented as national emergencies in most of the country’s municipalities, starting with a lockdown on March 20th, 2020. Recently, approaches that combine movement data (measured as the number of commuters between units), metapopulation models to describe disease dynamics subdividing the population into Susceptible-Exposed-Asymptomatic-Infected-Recovered-Diseased and statistical inference algorithms have been pointed as a practical approach to both nowcast and forecast the number of cases and deaths. We used an iterated filtering (IF) framework to estimate the model transmission parameters using the reported data across 281 municipalities from March to late October in locations with more than 50 reported deaths and cases in Colombia. Since the model is high dimensional (6 state variables in every municipality), inference on those parameters is highly non-trivial, so we used an Ensemble-Adjustment-Kalman-Filter (EAKF) to estimate time variable system states and parameters. Our results show the model’s ability to capture the characteristics of the outbreak in the country and provide estimates of the epidemiological parameters in time at the national level. Importantly, these estimates could become the base for planning future interventions as well as evaluating the impact of NPIs on the effective reproduction number (mathscr {R}_{eff}) and the critical epidemiological parameters, such as the contact rate or the reporting rate. However, our forecast presents some inconsistency as it overestimates the deaths for some locations as Medellín. Nevertheless, our approach demonstrates that real-time, publicly available ensemble forecasts can provide short-term predictions of reported COVID-19 deaths in Colombia. Therefore, this model can be used as a forecasting tool to evaluate disease dynamics and aid policymakers in infectious outbreak management and control.

Identifying spurious cycle slips based on iterative filtering under disturbed ionospheric conditions for undifferenced GNSS observations

The TurboEdit method is widely used to detect cycle slips on the global navigation satellite system (GNSS) carrier-phase measurements. However, it leads to an increasing number of false alarms in detecting cycle slips under disturbed ionospheric conditions. Besides, once the method detects a cycle slip at one satellite, it treats dual frequencies with cycle slips rather than at one frequency. Considering these two challenges, we developed a solution-based iterative filter detection method to reduce the number of spurious cycle slip detection under disturbed ionospheric conditions. The method initially assumes that there is no cycle slip at each frequency. We then estimate the solutions without cycle slips. A decision of exiting cycle slips is made by examining and comparing the two results solutions with or without cycle slips in terms of usable satellites, ambiguities, and residuals. The uncombined precise point positioning (PPP) during disturbed ionospheric conditions on 17 March 2015 at high latitude was studied to validate the proposed method. Results showed that the detected number of spurious cycle slips was reduced significantly. With fewer marked cycle slips, more stable and smoother positioning performance was achieved when fewer ambiguity parameters were reinitialized.

Application of Mediterranean Diet in Cardiovascular Diseases and Type 2 Diabetes Mellitus: Motivations and Challenges.

Objective: Cardiovascular disease (CVD) is the leading cause of disability and death in many countries. Together with CVD, Type 2 diabetes mellitus (T2DM) accounts for more than 80% of all premature non-communicable disease deaths. The protective effect of the Mediterranean diet (MedDiet) on CVD and its risk factors, including T2DM, has been a constant topic of interest. Notwithstanding, despite the large body of evidence, scientists are concerned about the challenges and difficulties of the application of MedDiet. This review aims to explore the motivations and challenges for using MedDiet in patients with CVD and T2DM. Design: An electronic search was conducted for articles about MedDiet published in PubMed, ScienceDirect, Scopus, and Web of Science up to December 2021, particularly on CVD and T2DM patients. From a total of 1536 studies, the final eligible set of 108 studies was selected. Study selection involved three iterations of filtering. Results: Motivation to apply MedDiet was driven by the importance of studying the entire food pattern rather than just one nutrient, the health benefits, and the distinct characteristics of MedDiet. Challenges of the application of MedDiet include lacking universal definition and scoring of MedDiet. Influences of nutritional transition that promote shifting of traditional diets to Westernized diets further complicate the adherence of MedDiet. The challenges also cover the research aspects, including ambiguous and inconsistent findings, the inexistence of positive results, limited evidence, and generalization in previous studies. The review revealed that most of the studies recommended that future studies are needed in terms of health benefits, describing the potential benefits of MedDiet, identifying the barriers, and mainly discussing the effect of MedDiet in different populations. Conclusions: In general, there is consistent and strong evidence that MedDiet is associated inversely with CVD risk factors and directly with glycemic control. MedDiet is the subject of active and diverse research despite the existing challenges. This review informs the health benefits conferred by this centuries-old dietary pattern and highlights MedDiet could possibly be revolutionary, practical, and non-invasive approach for the prevention and treatment CVD and T2DM.

Deep convolutional correlation iterative particle filter for visual tracking

This work proposes a novel framework for visual tracking based on the integration of an iterative particle filter, a deep convolutional neural network, and a correlation filter. The iterative particle filter enables the particles to correct themselves and converge to the correct target position. We employ a novel strategy to assess the likelihood of the particles after the iterations by applying K-means clustering. Our approach ensures a consistent support for the posterior distribution. Thus, we do not need to perform resampling at every video frame, improving the utilization of prior distribution information. Experimental results on three different benchmark datasets show that our tracker performs favorably against state-of-the-art methods.

Novel Brain Tumor Classification Model with MLPNN Using UNET

The detection and assessment of brain tumor are crucial for medical diagnosis. The MRI study supports radiologists and surgeons in the development of strategies for patient care in medical imaging. Accurate classification and segmentation of abnormal tissue over normal tissue are a key element. The proposed research work focused on an efficient Brain Tumor Segmentation and Classification approach named Multi-Feature Frequency Similarity Multi-Layer Perceptron Neural Network (MFFS-MLPNN). The proposed model considers both low- and high-grade features of glioblastoma present in MR images and varying features of tumors according to their size, contrast and texture. The MRI brain image is preprocessed by applying contrast curvature-based iterative shearlet filter and histogram equalization technique for enhancement. Followed by preprocessing, the denoised image has been segmented using the U-NET segmentation methodology. After segmentation, the tumors are extracted by using the cross-multi-linear local embedding method. The extracted features including Gray–White (GW) features were trained with multi-layer perception neural network for classification. The tumor has been classified into malignant and benign based on the result of multi-layer perceptron neural network. Based on the value of MFFS value, the algorithm performs classification. The effectiveness of the proposed MFFS-MPLNN algorithms has been verified over BRATS 2018 and 2019 datasets and compared with other state-of-the-art techniques. The experimental results show that average dice score is 85.47% for BRATS 2018 dataset whereas average dice score is 75.4% for BRATS 2019. It is observed that the dice score value is increased from 73.3% to 85.47% for BRATS 2018 and 64.3% to 75.4% for BRATS 2019 dataset. The average Hausdorff distance value has been decreased from 9.6% to 4.98% for BRATS 2018 dataset and 14.3% to 6.24% for BRATS 2019 dataset. Similarly, the sensitivity of the proposed algorithm has been increased from 70.7% to 89% and 61% to 82.1% for both BRATS 2018 and BRATS 2019 datasets, respectively.

An iterative frequency-domain envelope-tracking filter for dispersive signal decomposition in structural health monitoring

Dispersive signals are omnipresent in structural health monitoring and nondestructive testing. Due to the strong dispersion and multimodal characteristics, such signals often contain multiple components with frequency-varying group delays (GDs) overlapping in time–frequency (TF) domain, bringing great challenges to the existing signal processing methods. This paper presents a novel dispersive signal decomposition method named iterative frequency-domain envelope-tracking filter (IFETF) to accurately estimate the dispersion curves and separate overlapped components. Specifically, an envelope tracking issue is introduced in frequency domain based on dispersive signal model. Then, GDs are refined by recovered frequency envelope, and an iterative processing procedure is designed to ameliorate the TF resolution. To initialize the IFETF, an adaptive ridge detection technique is introduced to automatically estimate the number of components and their GDs with little prior knowledge. In addition, a high-quality TF representation can be constructed according to the results of IFETF, unambiguously revealing TF patterns of a multi-component dispersive signal. The proposed method was demonstrated by numerical analysis and engineering applications on broadband Lamb wave inspection and rail damage detection. Compared with traditional TF analysis techniques, the results show that the IFETF provides higher accuracy and efficiency in signal reconstruction and GD estimation.

Application of adaptive local iterative filtering in axis trace purification of turbine generator rotor

Application of adaptive local iterative filtering in axis trace purification of turbine generator rotor

Robust stable iterated unscented Kalman filter based on maximum correntropy criterion

The Unscented Kalman filter (UKF) based on maximum correntropy criterion (MCC) is robust to heavy-tailed non-Gaussian noise. However, the approximate linear measurement equation obtained by statistical linearization technique may be not accurate enough since it only uses prior information. In this paper, a robust stable iterative maximum correntropy criterion UKF (RS-IMCC-UKF) is proposed by using nonlinear measurement function directly and numerical stability methods. Different from the existing UKF algorithms, we only need to perform one-time unscented transformation in each filtering cycle, reducing the execution time of algorithm. Then a nonlinear enhancement model is constructed to handle predictions and observations simultaneously, which will be included in the cost function of robust MCC. In the process of iterative solution, thanks to the latest iteration values which are used to update the measurement information, RS-IMCC-UKF has more accurate results compared with traditional filters. In order to avoid non-positive definite characteristic in covariance matrix and the inverse operation in ill-conditioned matrix, the hyperbolic QR decomposition and Moore–Penrose pseudo-inversion are introduced to improve the numerical stability of the algorithm. Finally, a target tracking is modeled to verify the effectiveness of the algorithm.

Risk Assessment of Liver Metastasis in Pancreatic Cancer Patients Using Multiple Models Based on Machine Learning: A Large Population-Based Study.

Background A more accurate prediction of liver metastasis (LM) in pancreatic cancer (PC) would help improve clinical therapeutic effects and follow-up strategies for the management of this disease. This study was to assess various prediction models to evaluate the risk of LM based on machine learning algorithms. Methods We retrospectively reviewed clinicopathological characteristics of PC patients from the Surveillance, Epidemiology, and End Results database from 2010 to 2018. The logistic regression, extreme gradient boosting, support vector, random forest (RF), and deep neural network machine algorithms were used to establish models to predict the risk of LM in PC patients. Specificity, sensitivity, and receiver operating characteristic (ROC) curves were used to determine the discriminatory capacity of the prediction models. Results A total of 47,919 PC patients were identified; 15,909 (33.2%) of which developed LM. After iterative filtering, a total of nine features were included to establish the risk model for LM based on machine learning. The RF showed the most promising results in the prediction of complications among the models (ROC 0.871 for training and 0.832 for test sets). In risk stratification analysis, the LM rate and 5-year cancer-specific survival (CSS) in the high-risk group were worse than those in the intermediate- and low-risk groups. Surgery, radiotherapy, and chemotherapy were found to significantly improve the CSS in the high- and intermediate-risk groups. Conclusion In this study, the RF model constructed could accurately predict the risk of LM in PC patients, which has the potential to provide clinicians with more personalized clinical decision-making recommendations.