Two-dimensional Median Filter Research Articles

Ultrasonic computed tomography-based internal-defect detection and location of underwater concrete piers

Concrete piers are the key load-bearing components of wading bridges. Most detection equipment and methods focus on the surface defects of underwater piers, whereas the detection methods of internal defects are very deficient. This study proposes an ultrasonic computed tomography method combining linear travel time interpolation and simultaneous iterative reconstruction techniques for detecting internal defects in underwater concrete piers. The two-dimensional median filtering and self-organizing feature map are utilized to post-process tomographic images and the confidence approach is employed to detect defects. The efficacy of the proposed method is validated through numerical simulation and underwater experiments, respectively. The Comsol Multiphysics is utilized to simulate the two-dimensional random aggregate model of the concrete and internal defects are added to validate the method’s effectiveness from a mesoscopic perspective. Then actual scale piers containing embedded defects are cast and detected. The results demonstrate that the method can identify and locate internal defects in numerical simulations and practical tests. The numerical simulation experiments results are superior to those of scale concrete specimens. In addition, underwater concrete piers with larger internal defects, square holes, and no steel reinforcements tend to have better detection results.

Denoising of iodine-131 images using a median filter.

An iodine-131 (I) image visually appears to be contaminated with impulse noise. The two-dimensional median filter removes noise without sacrificing the edge information. Its performance depends on the shape and size of the mask. In this study, we have compared the performance of a plus-shape and a square-shape median filter for I whole-body images and found the filter with optimum parameter that improves I image quality acceptable to nuclear medicine physicians. A total of 150 whole-body I images were exported in DICOM format. These images were converted into PNG format and processed with a plus-shape and a square-shape median filter, with each shape mask having sizes of 3, 5, 7, and 9 pixels. The quality of the processed images was assessed by visual assessment by two nuclear medicine physicians and also quantitatively by evaluating metrics: mutual information, mean square error, peak signal-to-noise ratio, and difference entropy. Nuclear medicine physicians assigned a score to each image on the scale 1 (lowest) to 5 (highest) for image quality on the basis of the noise removal, smoothness, and edge information available in the image. Student's t-test was carried out to find the significant difference in the image quality (α=0.05) between the processed images with square-shape and cross-shape mask with the same pixel size. All experiments including statistical analysis were conducted using R installed on a personal computer. Both median filters improved the image quality of I images. The plus-shape median filter was found to show better performance in comparison with the square-shape median filter (P<0.001). The plus-shape median filter with a mask size of 7 pixels was found to be optimum for the processing of whole-body I images. The plus-shape median filter with a mask size of 7 pixels can be used to process whole-body I scintigraphic images without loss of clinical information.

A Low-Cost Design of 2D Median Filter

The median filter plays an important role in image processing for noise removing. To enhance the operating speed, hardware implementation of median filter on field programmable gate arrays (FPGA) or application specific IC (ASIC) are necessary and inevitable. This paper presents a low-cost and high-throughput hardware design for two-dimensional (2D) median filter. Two techniques are employed to reduce circuit area. One is the parallel three-valued sorting method for reducing the number of pipelined registers under constant speed, and the other is the functional sharing method used to implement median value sorting with the least number of comparisons. The proposed 2D median filter is implemented by two different synthesized methods. One is synthesized by Synopsys Design Compiler with the TSMC 90nm cell library (ASIC design) and the other is synthesized by Xilinx FPGA Virtex7 XC7VX690T (FPGA design). Experimental results show that the area cost of the proposed design is reduced more than 30% on average in comparison to the previous designs.

Russian scatterometer: discussion of the concept and the numerical simulation of wind field retrieval

Orbital scatterometry is briefly overviewed and its trends are indicated. Two scatterometer concepts are currently considered for trade-offs: with fixed and rotating antenna systems. The concept with a rotating antenna system was selected and SeaWinds was chosen as the prototype for the first Russian scatterometer. The scatterometer concept was then further developed and instead of two pencil beams, a fan-beam antenna was proposed (about 1° × 6°). The fan-beam antenna allows successive measurements for horizontal and vertical polarization in each wind vector cell (WVC). This increases the number of observations of the WVC at different incidence and azimuth angles during flight. The scatterometer parameters required to implement the proposed measurement geometry for an orbit altitude of 650 km and a swath width of 1525 km are discussed. A numerical scatterometer model that accounts for both the specifications and the observation geometry is developed. The scatterometer performance, with subsequent formation of a swath and splitting into WVCs, is simulated. The procedure of wind vector retrieval includes two stages: 1) determining wind speed and wind direction in a single WVC; and 2) using the information from adjacent WVCs to correct wind direction. It is shown that the accuracy of wind direction retrieval by a WVC can be increased by simultaneous radar cross-section (RCS) measurements at vertical and horizontal polarization. The basic error in determining wind direction is due to a 180° wind direction ambiguity caused by the form of RCS azimuth dependence. Two-dimensional median filtering is commonly employed in scatterometry to increase the accuracy of wind direction retrieval. In this study, two-dimensional angular median filtering was employed and it is shown that the error in wind direction retrieval significantly decreased. The results of the research indicate that wind field can be retrieved by the new scatterometer with the level of precision required.

Implementation of image median filters using programmable logic devices

A two-dimensional median filter structure, oriented at implementation in programmable logic devices and ensuring ultrafast determination of a median of data in a two-dimensional window owing to the parallel execution of comparison and ordering operations is presented. This approach ensures the cost-effective implementation is suitable for filtering both stationary and movable images, and allows one to expand the application area of the median filtering.

Detection of high‐speed and accelerated target based on the linear frequency modulation radar

The high-speed movement of a target may cause range migration and Doppler frequency migration of the radar echo, which has a serious impact on the detection performance of the radar. To resolve the problem of detecting a high-speed target in linear frequency modulation radar, this study analyses the effect on the integration gain caused by range migration and Doppler frequency migration, and proposes a corresponding compensation method according to the different input signal-to-noise ratios (SNRs) of the echo signal. To compensate for range migration in high SNRs, two-dimensional median filtering and constant false alarm rate technology are combined to estimate the speed. For low SNRs, based on coarse valuations, the authors use the discrete Fourier transformation (DFT) to realise the fractional delay cell to improve speed accuracy. Furthermore, to compensate for Doppler frequency migration, an instantaneous cross-correlation method is proposed for high SNRs, which is combined with the fractional Fourier transform method to estimate the acceleration for low SNRs. The input SNR threshold for the different algorithms is then analysed using simulation data, and the theoretical reference value is shown. Finally, the study verifies the effectiveness of the proposed methods through simulation and measured data.

Regularization of DT-MRI Using 3D Median Filtering Methods

DT-MRI (diffusion tensor magnetic resonance imaging) tractography is a method to determine the architecture of axonal fibers in the central nervous system by computing the direction of the principal eigenvectors obtained from tensor matrix, which is different from the conventional isotropic MRI. Tractography based on DT-MRI is known to need many computations and is highly sensitive to noise. Hence, adequate regularization methods, such as image processing techniques, are in demand. Among many regularization methods we are interested in the median filtering method. In this paper, we extended two-dimensional median filters already developed to three-dimensional median filters. We compared four median filtering methods which are two-dimensional simple median method (SM2D), two-dimensional successive Fermat method (SF2D), three-dimensional simple median method (SM3D), and three-dimensional successive Fermat method (SF3D). Three kinds of synthetic data with different altitude angles from axial slices and one kind of human data from MR scanner are considered for numerical implementation by the four filtering methods.

Material Classification of Underground Utilities From GPR Images Using DCT-Based SVM Approach

In this letter, we introduce the utilization of discrete cosine transform (DCT) coefficients as features supplied to the support vector machine (SVM) classifier to identify underground utility material from ground penetrating radar (GPR) imagery. Different types of features, reflected signal amplitudes, and statistical features, combined with the SVM classifier for material identification of underground utilities, are also studied and compared to the DCT-based approach. The system performance is conducted by simulation studies using generated GPR images created by a GPR finite-difference time-domain-based simulator used to develop various acquisition situations by changing the utility material type, position, and size parameters. The efficiency of the proposed technique in material identification is assessed using noisy generated GPR images degraded with speckle noise. Two-dimensional median and adaptive Wiener filters are also examined as a preprocessing step to the studied techniques. Simulation results show that the proposed technique combined with adaptive Wiener filter reveals a good performance regarding the recognition accuracy compared to the other studied techniques in noisy environment.

Signal processing in optical coherence tomography for aerospace material characterization

Based on a customized time-domain optical coherence tomography (OCT) system, a series of signal processing approaches have been designed and reviewed. To improve demodulation accuracy and image quality, demodulation approaches such as median filter, Hilbert transform, and envelope detector were investigated with simulated as well as experimental data. Without noise, the Hilbert transform has the best performance, but after considering the narrow-band noise in the modulated signal, the envelope detector was selected as the ideal demodulation technique. To reduce noise and enhance image contrast, digital signal processing techniques such as a bandpass filtering and two-dimensional median filtering were applied before and after the demodulation, respectively. Finally with integration of the customized OCT setup and designed signal processing algorithms, aerospace materials, such as polymer coatings and glass-fiber composites, were successfully characterized. The cross-sectional images obtained clearly show the microstructures of the materials.

An Implementation and Improvement of Fast Two-Dimensional Median Filtering

This paper discussed a conventional fast median filtering algorithm for FPGA implementation. An improved way -- Quasi-median filtering algorithm -- have been proposed to reduce the occupancy rate of FPGA resources on the premise of ensuring the result of median filtering. Through the detailed analysis and comparison of results of simulation and experiments, conclusions can be drawn that such improvements can achieve better filtering results, and can reduce FPGA resource utilization. It offers some value for the application of design which requires more FPGA resources.

New Multileaf Collimator with a Leaf Width of 5 mm Improves Plan Quality Compared to 10 mm in Step-and-Shoot IMRT of HNC Using Integrated Boost Procedure

To investigate whether a new multileaf collimator with a leaf width of 5 mm (MLC-5) over the entire field size of 40 x 40 cm(2) improves plan quality compared to a leaf width of 10 mm (MLC-10) in intensity-modulated radiotherapy (IMRT) with integrated boost for head and neck cancer. A plan comparison was performed for ten patients with head and neck cancer. For each patient, seven plans were calculated: one plan with MLC-10 and nine beams, four plans with MLC-5 and nine beams (with different intensity levels and two-dimensional median filter sizes [2D-MFS]), and one seven-beam plan with MLC-5 and MLC-10, respectively. Isocenter, beam angles and planning constraints were not changed. Mean values of common plan parameters over all ten patients were estimated, and plan groups of MLC-5 and MLC-10 with nine and seven beams were compared. The use of MLC-5 led to a significantly higher conformity index and an improvement of the 90% coverage of PTV1 (planning target volume) and PTV2 compared with MLC-10. This was noted in the nine- and seven-beam plans. Within the nine-beam group with MLC-5, a reduction of the segment number by up to 25% at reduced intensity levels and for increased 2D-MFS did not markedly worsen plan quality. Interestingly, a seven-beam IMRT with MLC-5 was inferior to a nine-beam IMRT with MLC-5, but superior to a nine-beam IMRT with MLC-10. The use of an MLC-5 has significant advantages over an MLC-10 with respect to target coverage and protection of normal tissues in step-and-shoot IMRT of head and neck cancer.

Dual-use of compact HF radars for the detection of mid- and large-size vessels

This paper describes the development of the SeaSonde High Frequency Radar system into a dual-use application for the mapping of ocean surface currents and detection of ships at sea. This development entailed the creation of a new radar waveform that would permit this dual-use as well as a detection algorithm to identify the ships in the radar spectra. The detection algorithm utilizes two methods for calculating a background signal level: an infinite impulse response (IIR) filter and a two-dimensional median filter. These two methods are employed simultaneously with multiple length averaging times to maximize the number of detections. The initial phase of development focused on improving the radar waveform to maximize the results for ship detection while still retaining the ability to measure surface currents. The latter phase of the development concentrated on testing the detection algorithm on a known vessel in different environmental conditions.

A high-accuracy method for the removal of point sources from maps of the cosmic microwave background

A new method for removing point radio sources and other non-Gaussian noise is proposed as a means of improving the accuracy of estimates of the angular power spectrum of the cosmic microwave background (CMB). The main idea of the method is to reconstruct fluctuations of the CMB in places contaminated by such emission, while traditional methods simply exclude these regions from consideration, leading to the appearance of “holes” in the resulting maps. The fundamental possibility of reconstructing the CMB signal in such holes follows from the analytical properties of a function with a finite spatial spectrum (the Silk damping frequency). A two-dimensional median filter is used to localize the point radio sources. Results of simulations of the method for maps of modest size are presented. The efficiency of applying the method to reconstruct the CMB from data with limited resolution and contaminated by appreciable pixel noise is investigated. The fundamental possibility of applying the method to reconstruct the CMB distribution in the region of the Galaxy is also demonstrated.

Root Images of Median Filters

Median filters are frequently used in signal analysis because of their smoothing properties and their insensitivity with respect to outliers in the data. Since median filters are nonlinear filters, the tools of linear theory are not applicable to them. One approach to deal with nonlinear filters consists in investigating their root images (fixed elements or signals transparent to the filter). Whereas for one-dimensional median filters the set of all root signals can be completely characterized, this is not true for higher dimensional filters. In 1989, Dohler stated a result on certain root images for two-dimensional median filters. Although Dohlers results are true for a wide class of median filters, his arguments were not correct and his assertions do not hold universally. In this paper we give a rigorous proof of Dohlers results. Moreover, his approach is generalized to the d-dimensional case.

PIV Technique for the Simultaneous Measurement of Dilute Two-Phase Flows

A Particle Image Velocimetry (PIV) image processing technique has been developed which can be applied to solid-liquid two-phase turbulent flows. The main principle of the technique is to utilize a two-dimensional median filter to generate separate images of the two phases, thus eliminating the errors induced by the distinct motion of the dispersed component. The accuracy and validity of the technique have been studied in the present research for different filter widths, f, and for 4 groups of different sized dispersed particles ranging from an effective image diameter of dp=2.9 pixels to 13 pixels in combination with tracer particles with an effective image size of dt∼2.4 pixels. The results have shown that the errors introduced by the filter are negligible, and mainly arise in regions of large velocity gradients that are sensitive to the slight loss of information incurred by the processing. The filter width f also affects the algorithm’s ability to correctly separate and identify the dispersed phase particles from the two-phase images, with the main result that above a critical particle image size ratio, dp/dt≈3.0, the particle size had no significant influence on the number of particles identified, or the accuracy of the displacement calculation. Sample results of particle-fluid interaction and cross-correlation terms which can be obtained from the method are also presented. [S0098-2202(00)01104-4]

Single-clock-cycle two-dimensional median filter

Median filters are of interest to image processing due to their ability to remove impulsive noise. Conventional digital implementations of the median function, however, require multiple clock cycles, a number that is proportional to the size of the 2-D data block. We present in the Letter a complete CMOS implementation, which consumes very little power and computes the median in just one clock cycle, independently from the size of the data block.

Deterministic and statistical properties of two-dimensional FIR-median hybrid filters

In this paper, we present a new class of two-dimensional FIR-median hybrid (FMH) filters, which we call separable FMH filters, for image noise smoothing. In a separable FMH filter, a one-dimensional FMH filter is applied to the rows and the columns of an image successively. The deterministic properties of separable and cross window FMH filters are discussed. Under certain assumptions, it is proved that a root of a separable FMH filter is a root of the corresponding cross window FMH filter. The noise attenuating properties of a separable FMH filter are studied and compared with those of the separable median filter, the two-dimensional median filter with a square window, and the cross window FMH filter.

Two-dimensional median filter algorithm for parallel reconfigurable computers

A fast and flexible median algorithm is presented which scales well with window size. It is based on the use of image histograming. Two accumulator arrays are used to determine the median value of a discrete sequence of numbers. Speed-up factors of 3 and 4 are achieved over conventional histograming methods (those using single accumulator arrays). Two approaches have been implemented with optimisation in mind. Worst- and best case machine performance boundaries are defined. Timings are given for both types of parallel architecture.

Efficient serial and parallel algorithms for median filtering

A serial algorithm for separable median filtering is developed that requires only two comparisons per element when the window size is three. In addition, fast parallel concurrent-read-exclusive-write parallel random-access machine (CREW PRAM) algorithms with good processor-time product are developed for separable median filtering and two-dimensional median filtering. >

Adaptive Median Filtering for Impulse Noise Elimination in Real-Time TV Signals

A two-dimensional (3 × 3) median filter with controlled turn-on is employed in an adaptive fashion for selective removal of impulse noise interference in real-time television signals. Various filters and threshold conditions are selected from programmed PROM's as a function of impulse noise detected and counted during a vertical interval window period. The adaptive median filter is used in conjunction with a satellite FM communication downlink system and can effectively improve the onset of threshold performance by about 3 dB carrier-to-noise ratio. The particular FM system discussed employs a multiplexed analog component, MAC, formatted TV signal. Use of the filter for standard NTSC composite signals is indicated.