Traditional Filtering Algorithm Research Articles

A dwindling filter line search algorithm for nonlinear equality constrained optimization

This paper proposes a dwindling filter line search algorithm for nonlinear equality constrained optimization. A dwindling filter, which is a modification of the traditional filter, is employed in the algorithm. The envelope of the dwindling filter becomes thinner and thinner as the step size approaches zero. This new algorithm has more flexibility for the acceptance of the trial step and requires less computational costs compared with traditional filter algorithm. The global and local convergence of the proposed algorithm are given under some reasonable conditions. The numerical experiments are reported to show the effectiveness of the dwindling filter algorithm.

Pedestrian Motion Tracking and Crowd Abnormal Behavior Detection Based on Intelligent Video Surveillance

Pedestrian tracking and detection of crowd abnormal activity under dynamic and complex background us ing I ntelligent V ideo S urveillance (IVS) system are beneficial for security in public places. T his paper present s a pedestrian tracking method combing Histogram of Oriented Gradients (HOG) detection and particle filter . This method regards the particle filter as the tracking framework, identifies the target area according to the result of HOG detection and modifies particle sampling constantly. Our method can track pedestrians in dynamic backgrounds more accurately compared with the traditional particle filter algorithms. Meanwhile, a method to detect crowd abnormal activity is also proposed based on a model of crowd features using Mixture of Gaussian (MOG). This method calculates features of crowd-interest points, then establishes the crowd features model using MOG, conducts self-adaptive updating and detect s a bnormal activity by matching the input feature with model distribution. Experiments show o ur algorithm can efficiently detect abnormal velocity and escape panic in crowds with a high detection rate and a relatively low false alarm rate

Color Remote-sensing Image Segmentation Based on Improved Region Filter

High resolution remote-sensing images provide abundant color, shape structure and texture information. However, region-based segmentations do not allow to fully exploit the richness of this kind of images. Despite the enormous progress in the analysis of remote sensing imagery over the past three decades, there is a lack of guidance on how to select an image segmentation algorithm suitable for the image type and size. In accordance with the characteristics of color high-resolution remote sensing images, we put forward an improved region filter remote-sensing color-image segmentation algorithm based on traditional JSEG segmentation. In order to well describe the color homogeneity in region, the local homogeneity matrix can be adopted to correct the local value in traditional JSEG segmentation algorithm so as to accurately reflect the boundary region and enhance the accuracy of image segmentation. Simulation experiment results demonstrate that our proposed algorithm can efficiently overcome the inaccurate segmentation of boundary region. In addition, compared with the traditional filter algorithm, our proposed algorithm can extract weak edge more accurately so as to achieve the better segmentation result in high resolution remote-sensing image segmentation

A Blind Adaptive Kalman-PIC MUD Algorithm for the Multiple Access Communication Network

In the parallel signal processing, the multi-user detection precision of parallel interference cancellation (PIC) detector is always easily affected by the decision error diffusion. Based on the fast convergence and low complexity of blind adaptive Kalman algorithm, a new blind adaptive Kalman-PIC (KPIC) multi-user detection (MUD) algorithm is proposed for the direct sequence spread spectrum code division multiple access (DS-CDMA) system network with strong multiple access interference (MAI). Compared with traditional standard Kalman filter and PIC algorithm, the proposed combined program can totally track the time-varying channel, effectively estimate unknown noise statistics characteristics on-line while conducting state filtering, possibly minimize the detection error diffusion in the interference cancellation processing of single PIC algorithm, thus effectively suppress MAI. Simulation results show that the KPIC algorithm is of better convergence, dynamic tracking ability and precision

Study of Steady-State Kalman Filtering Based on Thresholding Optimal Gain

For the devices with limited computing capacity and processing time, this paper proposes a new steady-state kalman filtering method based on thresholding optimal gain. The core algorithm is about improving the updating stage of the kalman filtering with a double-threshold optimal gain, which improves the dynamic tracking performance of the system steady-state while reducing the filtering algorithm computation. The results of the simulation prove that the algorithm has a better steady-state convergence effect than traditional kalman filtering algorithm. The method proposed in this paper can be widely used in many occasions, as it simplifies the computation of the optimal kalman gain and the a priori estimation.

均值滤波算法研究

We introduce the classical non-local means filtering algorithm and the improved non-local means filtering algorithm with the weight function modified by Manjon. In this paper,we propose different weight function,and make it have rotating shift invariance for the local windows while keeping the time complexity of optimizing the visual effect and SNR. By adding noise standard deviation from Gaussian additive noise ranging from 10 to 100,we compare the improved algorithms with traditional filtering algorithms and Manjon non-mean filtering algorithm. The results show that the improved algorithm from either visual or numerical is superior to Manjon non-mean filtering algorithm.

The Research of Power plant Operating Data Based on Real-time Digital filtration technology

Real-time monitoring of the data of the thermal power plant is the basis of accurate analyzing thermal economy and accurate reconstruction of the operating state. Due to noise is inevitable, we need real-time monitoring data filtering to get accurate information of units and equipment in the operating data of thermal power plant. Real-time filtering algorithm can’t be used to correct the current data with future data. Compared with traditional filtering algorithm, there are a lot of constraints. First-order lag filtering method and weighted recursive average filtering method can be used for real-time filtering. This paper analyzes the characteristics of the two filtering methods and applications for real-time processing of the positive spin simulation data, and the thermal power plant operating data. The analysis revealed that the weighted recursive average filtering method applied to the simulation and real-time plant data filtering achieved very good results. DOI: http://dx.doi.org/10.11591/telkomnika.v11i10.3468

Distortion Parameters Analysis Method Based on Improved Filtering Algorithm

In order to realize the accurate distortion parameters test of aircraft power supply system, and satisfy the requirement of corresponding equipment in the aircraft, the novel power parameters test system based on improved filtering algorithm is introduced in this paper. The hardware of the test system has the characters of s portable and high-speed data acquisition and processing, and the software parts utilize the software Labwindows/CVI as exploitation software, and adopt the pre-processing technique and adding filtering algorithm. Compare with the traditional filtering algorithm, the test system adopted improved filtering algorithm can help to increase the test accuracy. The application shows that the test system with improved filtering algorithm can realize the accurate test results, and reach to the design requirements. DOI: http://dx.doi.org/10.11591/telkomnika.v11i10.3416

An Optimized Particle Filter for Signal De-Noising Processing

The traditional particle filter (PF) algorithm is well known for signal noise reduction processing, but it exists problems of particle impoverishment and cumulation of estimation errors. An optimized PF algorithm called RBF-PF is proposed in this paper, which uses radial basis function network for training and optimizing the process of particle filter in the sampling. Experimental analysis verifies that the new method used to gain the signal-to-noise ratio is better than traditional PF algorithm during dealing with the added noise signal.

A Mixed Noise Filtering Algorithm for High-Speed Sequence Image Processing

t is a effective means by using high-speed vision to locate mobile targets. Under the circumstance of high frame rate and high sensitivity (300Hz), in addition to the Gaussian noise and impulse noise, the image quality is also influenced by atmospheric instability, and it is mainly expressed as Gaussian noise. An improved adaptive threshold weighted mean (IATWM) de-noising algorithm is proposed in this paper. According to the characteristics of impulse noise, the algorithm is able to obtain the threshold adaptively and separate the impulse noise. Then, the weighted median filtering algorithm is used to remove the impulse noise. And the improved weighted average filter algorithm is adopted to remove the Gaussian noise for graphics with Gaussian noise. The algorithm could deal with the Gaussian noise and impulse noise separately, avoiding the weaken handling for the parts not subject to pixels pollution of the impulse noise. The experimental results show that the processing result of the algorithm is able to retain the image details, superior to the traditional filtering algorithms and MTM algorithm. In addition, the algorithm provides an effective way to eliminate the mixed noise, along with a good effect on the high-speed sequence image processing.

MULTI-OBJECTIVE EVOLUTIONARY ALGORITHMS FOR FILTER BASED FEATURE SELECTION IN CLASSIFICATION

Feature selection is a multi-objective problem with the two main conflicting objectives of minimising the number of features and maximising the classification performance. However, most existing feature selection algorithms are single objective and do not appropriately reflect the actual need. There are a small number of multi-objective feature selection algorithms, which are wrapper based and accordingly are computationally expensive and less general than filter algorithms. Evolutionary computation techniques are particularly suitable for multi-objective optimisation because they use a population of candidate solutions and are able to find multiple non-dominated solutions in a single run. However, the two well-known evolutionary multi-objective algorithms, non-dominated sorting based multi-objective genetic algorithm II (NSGAII) and strength Pareto evolutionary algorithm 2 (SPEA2) have not been applied to filter based feature selection. In this work, based on NSGAII and SPEA2, we develop two multi-objective, filter based feature selection frameworks. Four multi-objective feature selection methods are then developed by applying mutual information and entropy as two different filter evaluation criteria in each of the two proposed frameworks. The proposed multi-objective algorithms are examined and compared with a single objective method and three traditional methods (two filters and one wrapper) on eight benchmark datasets. A decision tree is employed to test the classification performance. Experimental results show that the proposed multi-objective algorithms can automatically evolve a set of non-dominated solutions that include a smaller number of features and achieve better classification performance than using all features. NSGAII and SPEA2 outperform the single objective algorithm, the two traditional filter algorithms and even the traditional wrapper algorithm in terms of both the number of features and the classification performance in most cases. NSGAII achieves similar performance to SPEA2 for the datasets that consist of a small number of features and slightly better results when the number of features is large. This work represents the first study on NSGAII and SPEA2 for filter feature selection in classification problems with both providing field leading classification performance.

An infrared image impulse noise suppression algorithm based on fuzzy logic

In this paper we propose a novel impulse noise suppression algorithm for infrared images based on fuzzy logic. The traditional impulse noise filter algorithms, such as median filter and its modified versions, will cause the loss of dim targets and detail information along with the impulse noise. The proposed algorithm includes two steps, noise detection and removal. First, we introduce fuzzy algorithm to deal with the fuzzy and imprecise problem to obtain the membership function, which describes the extent of the noise pollution and is used to detect noisy pixels; Second, consider that the gray-levels of non-noisy pixels change continuously, we use a weighted-interpolation algorithm in order to improve the preservation of the detail information and dim targets. Experiment results demonstrate that the proposed algorithm, with highly practical value, not only performs excellently on impulse noise filtering while preserves the detail information and dim targets, but also requires no prior knowledge.

Improved Based on "Self-Adaptive Turning Rate" Model Algorithm

For tracking the object with tracking nonlinear, high maneuvering target，traditional interactive multiple model self-adaptive filter algorithm was usually adopted．The turning rate estimate was very important. However，the performance of turning rate algorithm was not so satisfactory in the model．Thus，a new the average value turning rate algorithm based on self-adaptive turning model was proposed．Aiming at additional device for turning rate estimation turning model, the parameters α and β were introduced to adjust the roughness of turning rate．Aiming at target constant turning movement and orthogonal turning rate unequal, estimates, turning rate was used the average value model to reduce the noise and error influence. Simulation results showed that the proposed algorithm was more suitable for the objects with Nonlinear, high maneuvering target tracking and could remarkably reduce the sample，and thus achieve much better tracking performance. DOI: http://dx.doi.org/10.11591/telkomnika.v11i6.2602

Urban Expressway Travel Time Prediction Method Based on Fuzzy Adaptive Kalman Filter

According to the poor adaptive ability of traditional filter algorithm in the estimatio n for traffic state and travel time with Kalman filter, an improved fuzzy adaptive Kalman filtering method was pro posed. The new interest of observation noise was defined, and the fuzzy logic was used to adjust the importance weights of system noise and observation noise through on-line monitoring the interest changes, which changed the trust and utilization degree of the model for the observation information, and this made the filter eventually tend to be stable. To guarantee the real-time performance of system, a direct input - output fuzzy membership function matching method was put forward to take the place of fuzzy reasoning. The method was tested on the urban expressway in Guangzhou by using real-time detection data, and the result show that the traffic state es timation model had better tracking ability than conventional Kalman filter, and results of travel time prediction show that there was a sligh t difference between the prediction value and that of actual observation in free traffic flow state, and the relative error was under 1 5% in traffic congested state. The precision and applicability of this method were acceptable, and it can be used to provide a basis for travel time of urban expressway in traffic control and guidance system.

Study on Underwater Terrain Matching Techniques Based on an Improved PMF Algorithm

Due to the existence of the accumulation errors in inertial navigation system(INS) and traditional point-mass filter (PMF) algorithm has problems of navigation precision and reliability when it is used for underwater terrain matching navigation, an improved PMF algorithm is proposed. Because the PMF algorithm is sensitive to initial errors and degeneration phenomenon usually appears with the increasing of matching number, the improved algorithm adopts cubic spline interpolation method to deal with the posterior probability density to prevent degeneration phenomenon. Then the TERCOM algorithm is introduced for initial alignment to reduce the influence of the initial errors. Finally the improved algorithm is simulated. The simulation results show the algorithm has a good robustness and accuracy.

The Realization of Rapid Median Filter Algorithm on FPGA

Traditional median filter algorithm has the long processing time, which goes against the real-time image processing. According to its shortcomings, this paper puts forward the rapid median filter algorithm, and uses DE2 board of the company called Altera to do the realization on FPGA (CycloneII 2C35). The experimental results show that the image pre-processing system is able to complete a variety of high-level image algorithms in milliseconds, and FPGA's parallel processing capability and pipeline operations can dramatically improve the speed of image processing, so the FPGA-based image processing system has broad prospects for development．

Research about Separation of the Surface Geometry Error for Large-Scale Holes Based on Wavelet Transform

In order to meet high accurate measuring of geometry error for large-scale holes, an non-contact measuring method used laser probe is presented. According to the characteristics of holes’s actual surface composited by multiplefrequency component. in order to accurately extract the various types of error signal，in the analysis of the traditional digital filter algorithm and wavelet transform through the experimental analysis of the steps of wavelet filter decomposition, and extracte integratedly the surface form error, surface waviness and surface roughness, to meet the measurement requirements of large-scale holes. The measuring system has the characteristics of simple, practical, reliable, and anti-interference.

State-of-Charge Estimation of the Lithium-Ion Battery Using an Adaptive Extended Kalman Filter Based on an Improved Thevenin Model

An adaptive Kalman filter algorithm is adopted to estimate the state of charge (SOC) of a lithium-ion battery for application in electric vehicles (EVs). Generally, the Kalman filter algorithm is selected to dynamically estimate the SOC. However, it easily causes divergence due to the uncertainty of the battery model and system noise. To obtain a better convergent and robust result, an adaptive Kalman filter algorithm that can greatly improve the dependence of the traditional filter algorithm on the battery model is employed. In this paper, the typical characteristics of the lithium-ion battery are analyzed by experiment, such as hysteresis, polarization, Coulomb efficiency, etc. In addition, an improved Thevenin battery model is achieved by adding an extra <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">RC</i> branch to the Thevenin model, and model parameters are identified by using the extended Kalman filter (EKF) algorithm. Further, an adaptive EKF (AEKF) algorithm is adopted to the SOC estimation of the lithium-ion battery. Finally, the proposed method is evaluated by experiments with federal urban driving schedules. The proposed SOC estimation using AEKF is more accurate and reliable than that using EKF. The comparison shows that the maximum SOC estimation error decreases from 14.96% to 2.54% and that the mean SOC estimation error reduces from 3.19% to 1.06%.

Automatic Estimation of Inertial Navigation System Errors for Global Positioning System Outage Recovery

This article presents an alternative approach of solving global positioning system (GPS) outages without requiring any prior information about the characteristics of the inertial navigation system (INS) and GPS sensors. INS can be used as a standalone system to bridge the outages during GPS signal loss. Kalman filter (KF) is widely used in INS and GPS integration to present a forceful navigation solution by overcoming the GPS outage problems. Unfortunately, KF is usually criticized for working under predefined models and for its observability problem of hidden state variables, sensor dependency, and linearization dependency. This approach utilizes a genetic neuro-fuzzy system (GANFIS) to predict the INS position and velocity errors during GPS signal blockages suitable for real-time application. The proposed model is able to deal with noise and disturbances in the GPS and INS output data in different dynamic environments compared to other traditional filtering algorithms such as the neural network and neuro fuzzy. Real field test results using the micro-electro-mechanical system grade inertial measurement unit with an integrated GPS shows a significant improvement obtained from the integrated GPS/INS system using the GANFIS module compared to traditional methods such as Kalman filtering, particularly during long GPS satellite signal blockage.

Support vector regression based color image restoration in YUV color space

A support vector regression (SVR) based color image restoration algorithm is proposed. The test color images are firstly mapped into the YUV color space, and then SVR is applied to build up a theoretical model between the degraded images and the original one. Performance comparisons of the proposed algorithm versus traditional filtering algorithms are given. Experimental results show that the proposed algorithm has better performance than traditional filtering algorithms and has less computation time than iterative blind deconvolution algorithm.