Adaptive Mean Filter Research Articles

Shape-Preserving Preprocessing for Human Pulse Signals Based on Adaptive Parameter Determination

The use of the human pulse signal for medical diagnosis is a mainstay in the practice of traditional Chinese medicine. Computer processing of this signal may be used to automate diagnostic procedures and to reveal sources of information in the waveform that have been used by both eastern and western physicians for more than two millennia. A new method for preprocessing of the human pulse signal significantly improves feature extraction and classification of the waveform. Baseline distortion is first removed using the dual-tree complex wavelet transform (DT-CWT) and cubic spline interpolation, then a novel filtering method removes the residual background noise. Filtering is implemented in two stages. In the initial pass, a majority of the noise is eliminated by an adaptive mean filter whose sliding window duration is selected automatically based on a chain code and the DT-CWT. In the second pass, residual high frequency noise is removed using the DT-CWT with a new threshold determination. Experimental results demonstrate effective removal of background disturbances with excellent preservation of pulse peak information essential for proper parametric representation and classification of the waveform.

The Pre-Processing Techniques for Breast Cancer Detection in Mammography Images

Presently breast cancer detection is a very important role for worldwide women to save the life. Doctors and radio logistic can miss the abnormality due to inexperience in the field of cancer detection. The pre- processing is the most important step in the mammogram analysis due to poor captured mammogram image quality. Pre-processing is very important to correct and adjust the mammogram image for further study and processing. There are Different types of filtering techniques are available for pre- processing. This filters used to improve image quality, remove the noise, preserves the edges within an image, enhance and smoothen the image. In this paper, we have performed various filters namely, average filter, adaptive median filter, average or mean filter, and wiener filter.

Fast restoration of natural images corrupted by high-density impulse noise

In this paper, we suggest a general model for the fixed-valued impulse noise and propose a two-stage method for high density noise suppression while preserving the image details. In the first stage, we apply an iterative impulse detector, exploiting the image entropy, to identify the corrupted pixels and then employ an Adaptive Iterative Mean filter to restore them. The filter is adaptive in terms of the number of iterations, which is different for each noisy pixel, according to the Euclidean distance from the nearest uncorrupted pixel. Experimental results show that the proposed filter is fast and outperforms the best existing techniques in both objective and subjective performance measures.

Improving lidar-based forest structure mapping with crown-level pit removal

Lidar canopy height models (CHMs) often appear noisy usually due to large height variations within crowns called pitted pixels (or intra-canopy gaps). Forest structure mapping derived from these data may be erroneous since the presence of pitted pixels can distort the different statistical attributes calculated from the noisy CHM. In this paper, a robust algorithm called adaptive mean filter (AMF) has been developed to deal with the pitted pixels over CHMs. The algorithm adaptively detects pitted pixels within a 3 × 3 window by calculating a similarity index (SI) and comparing this value with the normalised value of the central pixel. Then a preset window size is used to fill the pitted pixels. The results of this filter using four preset window sizes were compared with other published filters for both individual tree and plot-level data in a pine plantation. The results show AMF can perform more efficiently than other filters for pitted pixel removal since it can fill up to 97 percent of pitted pixels while simultaneously preserving the maximum height values. Moreover, applying this pit filling method on the pitted CHM can significantly decrease the error of estimation of basal area and stand volume estimates by 10 percent. Finally, the overall accuracy of pine plantation structure estimates can be improved up to 17.7 percent using the CHM filtered by the AMF method compared to that derived from a pitted CHM.

A Highly Effective Adaptive Switching Mean Filter Algorithm for Salt and Pepper Noise Removal

This paper presents an efficient three-stage adaptive switching mean filter to remove salt-and-pepper impulse noise from highly corrupted images. Firstly, the noise detection stage is to detect pixels as “noise pixels” and “noise-free pixels”. The detected “noise pixels” will then be subjected to the second stage which is the noise cancellation, while “noise-free pixels” are retained and left unchanged. The method adaptively changes the size of the filtering window based on the number of the “noise-free pixels” in the neighborhood. For the filtering, only “noise-free pixels” in the window are considered to find the mean value. If this value is not available in the maximum window size, the last processed pixel value is used as the replacement. In the third stage, this algorithm utilizes previously processed neighboring pixel values to get better image quality as the really processed noise pixel used as a noise –free pixel for the next noisy pixel processing. Experimental results clearly show that the proposed algorithm outperforms many of the existing methods in terms of visual quality and quantitative measures. The advantage of the proposed method is that it works well for high-density salt & pepper noise even up to a noise percentage of 95%. General Terms Image processing, Image filtering, Noise reduction.

食品画像を用いた好み評価時の視線停留時間に関する実験的検討

As a fundamental study on quantifying subjective evaluation by eye movement information, a relationship between eye fixation time and subjective evaluation was discussed. Firstly, a view point measurement system to extract eye fixation in opinion tests with pictures was developed. In the system, linear projection transform and adaptive α-Trimmed mean filters were used to detect eye positions on a picture and reduce noise in view point measurements, respectively. The eye fixation was detected when the change of the eye position in less than 2° was kept for more than 150ms. Secondly, eye fixation and its time were measured when pairs of food pictures were displayed and one picture corresponding to an opinion “Like” or “Dislike” was subjectively selected. Results show the eye fixation time for a picture evaluated as the opinion “Like” or “Dislike” becomes longer than that for the other picture (not evaluated), which indicates a feasibility of detecting a difference of subjective evaluation by the eye fixation time.

Research on the White Noise Suppression by Adaptive Filtering of Genetic Algorithm

Signal transmission is often subject to the disturbance of white noise. Owing to the spectrum of white noise can be found in the real number field, it is often difficult to filter out with the traditional filter. This article describes the methods of white noise suppression using adaptive filter and mean filter. First, using the genetic algorithm to optimize the weight vector of the adaptive filter, and then using the method of the mean filter to further filter, Simulation results show that the filter can effectively suppress white noise.

An adaptive filter based on images' entropy

An adaptive filter to remove mixed Gaussian noise and impulse noise of the corrupted images based on their minimum entropies was proposed. The proposed filter calculated the noise rate of the corrupted image based on the minimum local entropy, and utilized the noise rate to determine the size of filter windows and the number of pixels to be brushed off in the filter windows of Alpha-trimmed mean filter, and then used the average of leaved pixels' gray as the output of the filter. The performance of the proposed filter was evaluated by experiments. Experimental results show that the superiority of this filter including the ability of removing noises and that of preserving the partial details of images in comparison with some existing methods (median filter and mean filter), and this filter is an adaptive Alpha-trimmed mean filter.

Edge and fine detail preservation in SAR images through speckle reduction with an adaptive mean filter

A speckle reduction algorithm, the Edge Map‐Directed Adaptive Mean (EMDAM) filter, is studied in this paper. It adapts the ordinary mean filter according to the scene heterogeneity. Edge‐crossing maps determined by an edge detector are used to find the largest homogeneous subregion in the moving filter window. Then, the mean filter is adapted only to this homogeneous part of the moving filter window and applied if no edge crossing is found. We compared some filters in the literature to the EMDAM filter using two examples: a 1997 JERS‐1 synthetic aperture radar (SAR) image of Tuzla, Istanbul and a computer‐simulated SAR image. The filter performance was assessed both quantitatively and qualitatively. We found that the EMDAM filter preserves textures and details while reducing speckle to a desired level. A new testing quantity, the Quality Factor (Q), is also introduced.

Adaptive switching mean filter using conditional morphological noise detector

The adaptive switching mean (ASM) filter is proposed to remove impulse noise. The filter first identifies the corrupted pixels using conditional morphological noise detection and then removes the detected impulses using the adaptive mean filter. Simulation results indicate that the ASM filter can suppress impulse noise effectively while preserving the details in the image very well, thus providing better restoration performance than many other switching-based filters.

Adaptive mean filtering for noise reduction in CT polymer gel dosimetry

X-ray computed tomography (CT) as a method of extracting 3D dose information from irradiated polymer gel dosimeters is showing potential as a practical means to implement gel dosimetry in a radiation therapy clinic. However, the response of CT contrast to dose is weak and noise reduction is critical in order to achieve adequate dose resolutions with this method. Phantom design and CT imaging technique have both been shown to decrease image noise. In addition, image postprocessing using noise reduction filtering techniques have been proposed. This work evaluates in detail the use of the adaptive mean filter for reducing noise in CT gel dosimetry. Filter performance is systematically tested using both synthetic patterns mimicking a range of clinical dose distribution features as well as actual clinical dose distributions. Both low and high signal-to-noise ratio (SNR) situations are examined. For all cases, the effects of filter kernel size and the number of iterations are investigated. Results indicate that adaptive mean filtering is a highly effective tool for noise reduction CT gel dosimetry. The optimum filtering strategy depends on characteristics of the dose distributions and image noise level. For low noise images (SNR approximately 20), the filtered results are excellent and use of adaptive mean filtering is recommended as a standard processing tool. For high noise images (SNR approximately 5) adaptive mean filtering can also produce excellent results, but filtering must be approached with more caution as spatial and dose distortions of the original dose distribution can occur.

An Algorithm for Adaptive Mean Filtering and Its Hardware Implementation

Noise due to the sensor and the electronics of a camera is an undesirable issue in any machine vision application. Such noise tends to corrupt images and to obstruct any further analysis. An algorithm to detect and cancel such noise, using statistical methods, is presented in this paper. The proposed algorithm is an adaptive mean filter, which filters out image regions that are found to be noise corrupted. The efficiency of the proposed filter was examined both qualitatively and quantitatively, by software simulation in several noisy conditions. The main advantage of the filter in hand is that it is appropriate for hardware implementation and can be easily incorporated to smart cameras. The hardware implementation of the filter is also presented in this paper. This implementation aims at time critical applications such as machine vision, inspection and visual surveillance.

Adaptive hybrid mean and median filtering of high-ISO long-exposure sensor noise for digital photography

This paper presents a new methodology for the reduction of sensor noise from images acquired using digital cameras at high- International Organization for Standardization (ISO) and long- exposure settings. The problem lies in the fact that the algorithm must deal with hardware-related noise that affects certain color channels more than others and is thus nonuniform over all color channels. A new adaptive center-weighted hybrid mean and median filter is formulated and used within a novel optimal-size windowing framework to reduce the effects of two types of sensor noise, namely blue-channel noise and JPEG blocking artifacts, common in high-ISO digital camera images. A third type of digital camera noise that affects long-exposure images and causes a type of sensor noise commonly known as ''stuck-pixel'' noise is dealt with by pre- processing the image with a new stuck-pixel prefilter formulation. Experimental results are presented with an analysis of the perfor- mance of the various filters in comparison with other standard noise reduction filters. © 2004 SPIE and IS&T. (DOI: 10.1117/1.1668279)

Image filtering for improved dose resolution in CT polymer gel dosimetry

X-ray computed tomography (CT) has been established as a feasible method of performing dosimetry using polyacrylamide gels (PAGs). A small density change occurs in PAG upon irradiation that provides contrast in PAG CT images. However, low dose resolution limits the clinical usefulness of the technique. This work investigates the potential of using image filtering techniques on PAG CT images in order to reduce image noise and improve dose resolution. CT image noise for the scanner and protocol used for the gel images is analyzed and found to be Gaussian distributed and independent of the contrast level in the images. As a result, several filters for reducing spatially invariant noise are investigated: mean, median, midpoint, adaptive mean, alpha-trimmed mean, sigma mean, and a relatively new filter called SUSAN (smallest univalue segment assimilating nucleus). All filters are applied, using 3x3, 5x5, and 7x7 pixel masks, to a CT image of a PAG irradiated with a stereotactic radiosurgery dose distribution. The dose resolution within 95% confidence (D(delta)95%) is calculated and compared for each filtered image, as well the unfiltered image. In addition, the ability of the filters to maintain the spatial integrity of the dose distribution is evaluated and compared. Results clearly indicate that the filters are not equal in their ability to improve D(delta)95% or in their effect on the spatial integrity of the dose distribution. In general, increasing mask size improves D(delta)95% but simultaneously degrades spatial dose information. The mean filter provides the greatest improvement in D(delta)95%, but also the greatest loss of spatial dose information. The SUSAN, mean adaptive, and alpha-trimmed mean filters all provide comparable, but slightly poorer dose resolution. In addition, the SUSAN and adaptive filters both excel at maintaining the spatial distribution of dose and overall are the best performing filters for this application. The midpoint filter, normally useful for Gaussian noise, is poor all-round, dramatically distorting the dose distribution for masks greater than 3x3. The median filter, a common edge preserving noise reduction filter, performs moderately well, but artificially increases high dose gradients. The sigma filter preserves the spatial distribution of dose very well but is least effective at improving dose resolution. In summary, dose resolution can be significantly improved in CT PAG dosimetry through postprocessing of CT images using spatial noise reduction filters. However, such filters are not equal in their ability to improve dose resolution or to maintain the spatial integrity of the dose distribution and an appropriate filter must be chosen depending on clinical demands of the application.

Image restoration techniques based on fuzzy neural networks

By establishing some suitable partitions of input and output spaces, a novel fuzzy neural network (FNN) which is called selection type FNN is developed. Such a system is a multilayer feedforward neural network, which can be a universal approximator with maximum norm. Based on a family of fuzzy inference rules that are of real senses, a simple and useful inference type FNN is constructed. As a result, the fusion of selection type FNN and inference type FNN results in a novel filter---FNN filter. It is simple in structure. And also it is convenient to design the learning algorithm for structural parameters. Further, FNN filter can efficiently suppress impulse noise superimposed on image and preserve fine image structure, simultaneously. Some examples are simulated to confirm the advantages of FNN filter over other filters, such as median filter and adaptive weighted fuzzy mean (AWFM) filter and so on, in suppression of noises and preservation of image structure.

Adaptive symmetric mean filter: a new noise-reduction approach based on the slope facet model

Two new noise-reduction algorithms, namely, the adaptive symmetric mean filter (ASMF) and the hybrid filter, are presented in this paper. The idea of the ASMF is to find the largest symmetric region on a slope facet by incorporation of the gradient similarity criterion and the symmetry constraint into region growing. The gradient similarity criterion allows more pixels to be included for a statistically better estimation, whereas the symmetry constraint promises an unbiased estimate if the noise is completely removed. The hybrid filter combines the advantages of the ASMF, the double-window modified-trimmed mean filter, and the adaptive mean filter to optimize noise reduction on the step and the ramp edges. The experimental results have shown the ASMF and the hybrid filter are superior to three conventional filters for the synthetic and the natural images in terms of the root-mean-squared error, the root-mean-squared difference of gradient, and the visual presentation.

Toward edge sharpening: a SAR speckle filtering algorithm

This paper makes two contributions. It first introduces an algorithm for synthetic aperture radar (SAR) speckle reduction and edge sharpening. Existing speckle filtering algorithms can effectively reduce the speckle effect but unfortunately also, to some degree, smear edges and blur images. Even for unfiltered images, there is still a need for edge sharpening; since SAR sensors have limited bandwidths, leading to slow responses to sudden changes (smearing sharp edges). The proposed algorithm functions as an adaptive-mean filter. Edge crossing points are detected by using the second-order derivative of the Gaussian function as a wavelet transform function. A proper dilation scale factor enables the wavelet transform function to detect only edge crossings and ignore the local oscillations. Then in a moving window the mean filter is applied if there is no edge crossing point. Otherwise, averaging is only applied to the part of the window separated by edge crossing points. Consequently, the algorithm smooths uniform areas while it sharpens and enhances edges. Edges of the filtered images are generally sharper than the original. Similarities between the proposed filter and other popular speckle filters, such as the Lee, Kuan, and Frost filters, designated for SAR multiplicative noise, are analyzed. Another contribution of the paper is the evaluation of popular filters from the viewpoint of texture preservation. The evaluation is carried out using the first and second-order histograms. Possible distortions caused by filters are explained.

Narrow-band interference suppression in spread-spectrum CDMA communications using pipelined recurrent neural networks

This paper investigates the application of pipelined recurrent neural networks (PRNN's) to the narrow-band interference (NBI) suppression over spread-spectrum (SS) code-division multiple-access (CDMA) channels in the presence of additive white Gaussian noise (AWGN) plus non-Gaussian observation noise. Optimal detectors and receivers for such channels are no longer linear. A PRNN that consists of a number of simpler small-scale recurrent neural network (RNN) modules with less computational complexity is conducted to introduce best nonlinear approximation capability into the minimum mean-squared error nonlinear predictor model in order to accurately predict the NBI signal based on adaptive learning for each module from previous non-Gaussian observations. Once the prediction of the NBI signal is obtained, a resulting signal is computed by subtracting the estimate from the received signal. Thus, the effect of the NBI can be reduced. Moreover, since those modules of a PRNN can be performed simultaneously in a pipelined parallelism fashion, this would lead to a significant improvement in its total computational efficiency. Simulation results show that PRNN-based NBI rejection provides a superior signal-to-noise ratio (SNR) improvement relative to the conventional adaptive nonlinear approximate conditional mean (ACM) filters, especially when the channel statistics and exact number of CDMA users are not known to those receivers.

Adaptive neighborhood mean and median image filtering

A new approach to adaptive-neighborhood or region-based noise filtering is presented. The basic idea in this technique is to identify contextually related features in the image and to carry out statistical filtering operations using the pixels in these areas. Neighborhoods in the image are identified as sets of pixels that are 8-connected to the reference or seed pixel and are within a specified gray-level tolerance of the seed pixel. Thus, operations are based on contextual details in the image rather than an arbitrary grouping ofpixels (as in 3 x 3 filtering). These methods are applied to synthesized and natural images, and it is shown both quantitatively and qualitatively that adaptive-neighborhood filtering techniques are superior to analogous fixed-neighborhood filtering techniques.