Multiscale Filtering Research Articles

Improved detection of soma location and morphology in fluorescence microscopy images of neurons

BackgroundAutomated detection and segmentation of somas in fluorescent images of neurons is a major goal in quantitative studies of neuronal networks, including applications of high-content-screenings where it is required to quantify multiple morphological properties of neurons. Despite recent advances in image processing targeted to neurobiological applications, existing algorithms of soma detection are often unreliable, especially when processing fluorescence image stacks of neuronal cultures. New methodIn this paper, we introduce an innovative algorithm for the detection and extraction of somas in fluorescent images of networks of cultured neurons where somas and other structures exist in the same fluorescent channel. Our method relies on a new geometrical descriptor called Directional Ratio and a collection of multiscale orientable filters to quantify the level of local isotropy in an image. To optimize the application of this approach, we introduce a new construction of multiscale anisotropic filters that is implemented by separable convolution. ResultsExtensive numerical experiments using 2D and 3D confocal images show that our automated algorithm reliably detects somas, accurately segments them, and separates contiguous ones. Comparison with existing methodsWe include a detailed comparison with state-of-the-art existing methods to demonstrate that our algorithm is extremely competitive in terms of accuracy, reliability and computational efficiency. ConclusionsOur algorithm will facilitate the development of automated platforms for high content neuron image processing. A Matlab code is released open-source and freely available to the scientific community.

A Multi-Scale Line Filter for Automatic Crack Inspection Based on X-Ray Images

Crack inspection is a critical processing step in an AXI (automatic X-ray inspection) system. There are challenges for crack inspection since images are usually with serious noise, non-uniform brightness, etc. This paper proposes a novel line filter that can segment various widths of crack regions in those inhomogeneity brightness and noisy Xray images. First, the input image is smoothed by the Gaussian functions with different scales, resulting in an image pyramid. Second, the adjacent images in the pyramid are subtracted to generate a Difference of Gaussian (DoG) pyramid and the extreme points from the subtracting result are selected as the output of the method. An edge suppression function is then defined to remove the edge response of the filter. Finally, the output of the proposed filter is defined as the combination of the DoG extreme result and the suppression function. The proposed method was applied to an AXI system and it performed better than the other two exited approaches. Experimental results showed that the proposed filter is robust to the image noise, crack sizes and brightness.

Diagonal slice spectrum assisted optimal scale morphological filter for rolling element bearing fault diagnosis

This paper presents a novel signal processing scheme, diagonal slice spectrum assisted optimal scale morphological filter (DSS-OSMF), for rolling element fault diagnosis. In this scheme, the concept of quadratic frequency coupling (QFC) is firstly defined and the ability of diagonal slice spectrum (DSS) in detection QFC is derived. The DSS-OSMF possesses the merits of depressing noise and detecting QFC. It can remove fault independent frequency components and give a clear representation of fault symptoms. A simulated vibration signal and experimental vibration signals collected from a bearing test rig are employed to evaluate the effectiveness of the proposed method. Results show that the proposed method has a superior performance in extracting fault features of defective rolling element bearing. In addition, comparisons are performed between a multi-scale morphological filter (MMF) and a DSS-OSMF. DSS-OSMF outperforms MMF in detection of an outer race fault and a rolling element fault of a rolling element bearing.

Fault detection method for railway wheel flat using an adaptive multiscale morphological filter

This study explores the capacity of the morphology analysis for railway wheel flat fault detection. A dynamic model of vehicle systems with 56 degrees of freedom was set up along with a wheel flat model to calculate the dynamic responses of axle box. The vehicle axle box vibration signal is complicated because it not only contains the information of wheel defect, but also includes track condition information. Thus, how to extract the influential features of wheels from strong background noise effectively is a typical key issue for railway wheel fault detection. In this paper, an algorithm for adaptive multiscale morphological filtering (AMMF) was proposed, and its effect was evaluated by a simulated signal. And then this algorithm was employed to study the axle box vibration caused by wheel flats, as well as the influence of track irregularity and vehicle running speed on diagnosis results. Finally, the effectiveness of the proposed method was verified by bench testing. Research results demonstrate that the AMMF extracts the influential characteristic of axle box vibration signals effectively and can diagnose wheel flat faults in real time.

Robust Retinal Vessel Segmentation via Locally Adaptive Derivative Frames in Orientation Scores.

This paper presents a robust and fully automatic filter-based approach for retinal vessel segmentation. We propose new filters based on 3D rotating frames in so-called orientation scores, which are functions on the Lie-group domain of positions and orientations [Formula: see text]. By means of a wavelet-type transform, a 2D image is lifted to a 3D orientation score, where elongated structures are disentangled into their corresponding orientation planes. In the lifted domain [Formula: see text], vessels are enhanced by means of multi-scale second-order Gaussian derivatives perpendicular to the line structures. More precisely, we use a left-invariant rotating derivative (LID) frame, and a locally adaptive derivative (LAD) frame. The LAD is adaptive to the local line structures and is found by eigensystem analysis of the left-invariant Hessian matrix (computed with the LID). After multi-scale filtering via the LID or LAD in the orientation score domain, the results are projected back to the 2D image plane giving us the enhanced vessels. Then a binary segmentation is obtained through thresholding. The proposed methods are validated on six retinal image datasets with different image types, on which competitive segmentation performances are achieved. In particular, the proposed algorithm of applying the LAD filter on orientation scores (LAD-OS) outperforms most of the state-of-the-art methods. The LAD-OS is capable of dealing with typically difficult cases like crossings, central arterial reflex, closely parallel and tiny vessels. The high computational speed of the proposed methods allows processing of large datasets in a screening setting.

Multiscale anisotropic diffusion for ringing artifact suppression in geophysical deconvolution data

Ringing artifact degradations always appear in the deconvolution of geophysical data. To address this problem, we propose a postprocessing approach to suppress ringing artifacts that uses a novel anisotropic diffusion based on a stationary wavelet transform (SWT) algorithm. In this paper, we discuss the ringing artifact suppression problem and analyze the characteristics of the deconvolution ringing artifact. The deconvolution data containing ringing artifacts are decomposed into different SWT subbands for analysis, and a new multiscale adaptive anisotropic filter is developed to suppress these degradations. Finally, we demonstrate the performance of the proposed method and describe the experiments in detail.

Multiscale Analysis of Neurite Orientation and Spatial Organization in Neuronal Images.

The spatial organization of neurites, the thin processes (i.e., dendrites and axons) that stem from a neuron's soma, conveys structural information required for proper brain function. The alignment, direction and overall geometry of neurites in the brain are subject to continuous remodeling in response to healthy and noxious stimuli. In the developing brain, during neurogenesis or in neuroregeneration, these structural changes are indicators of the ability of neurons to establish axon-to-dendrite connections that can ultimately develop into functional synapses. Enabling a proper quantification of this structural remodeling would facilitate the identification of new phenotypic criteria to classify developmental stages and further our understanding of brain function. However, adequate algorithms to accurately and reliably quantify neurite orientation and alignment are still lacking. To fill this gap, we introduce a novel algorithm that relies on multiscale directional filters designed to measure local neurites orientation over multiple scales. This innovative approach allows us to discriminate the physical orientation of neurites from finer scale phenomena associated with local irregularities and noise. Building on this multiscale framework, we also introduce a notion of alignment score that we apply to quantify the degree of spatial organization of neurites in tissue and cultured neurons. Numerical codes were implemented in Python and released open source and freely available to the scientific community.

Automated removal of stripe interference in full-disk solar images

The quality of full-disk solar Hα images is significantly degraded by stripe interference. In this paper, to improve the analysis of morphological evolution, a robust solution for stripe interference removal in a partial full-disk solar Hα image is proposed. The full-disk solar image is decomposed into a set of support value images on different scales by convolving the image with a sequence of multiscale support value filters, which are calculated from the mapped least-squares support vector machines (LS-SVMs). To match the resolution of the support value images, a scale-adaptive LS-SVM regression model is used to remove stripe interference from the support value images. We have demonstrated the advantages of our method on solar Hα images taken in 2001–2002 at the Huairou Solar Observing Station. Our experimental results show that our method can remove the stripe interference well in solar Hα images and the restored image can be used in morphology researches.

Application of statistical modeling of image spatial structures to automated visual inspection of product quality

Automated visual inspection (AVI) attracts increasing interest in product quality control both academic and industrial communities, particularly on mass production processes, because product qualities of most types can be characterized with their corresponding surface visual attributes. However, many product images in AVI systems are comprised of stochastically accumulative fragmentations (particles) of local homogeneity, without distinctive foregrounds and backgrounds, which brings great challenges in computer analysis, e.g., rice images, fabric images, and consequently, in the intelligent identification of the product qualities. A method of Weibull distribution (WD)-based statistical modeling of image spatial structures (ISSs) to inspect automatically the product quality is presented. The ISS, obtained with multi-scale and omnidirectional Gaussian derivative filters (OGDFs), was demonstrated to be subject to a representative WD model of integral form based on the theory of sequential fragmentation in advance. The WD-model parameters (WD-MPs) of the ISS, with essential human perceptual significance, were extracted as the visual features for product quality identification. The classification performance of the proposed product quality inspection method, namely, the proposed WD-MP features integrated with an introduced spline regression (SR) classifier in this study, was verified on two case studies in the field of the AVI of product quality, namely, automated rice quality classification, and intelligent fabric quality assessment in the corresponding assembly lines of industrial scale. Experimental results indicate that the proposed WD-MP features can effectively characterize the statistical distribution profiles of ISS of complex grain images, piled with a large number of stochastically accumulative fragmentations. The proposed method provides an effective tool for grain image modeling and analysis and consequently lays a foundation for the intelligent perception of product qualities on assembly lines.

Virus image classification using multi-scale completed local binary pattern features extracted from filtered images by multi-scale principal component analysis

Virus image classification is an important issue in clinical virology, highly accurate algorithm of automatic virus image classification is very helpful. In this paper, instead of extracting virus feature from the original image, we propose a novel method that extracts the virus feature from the filtered images by multi-scale principal component analysis (PCA). Firstly, multi-scale PCA filters are learned from all original images in the data set. Secondly, the original images are convolved with the learned filters. Therefore, the filtered images can capture the principal texture information from different perspectives. Then, the completed local binary pattern (CLBP) descriptor is firstly utilized to depict the features of all filtered virus images. The multi-scale CLBP features extracted from filtered images by multi-scale PCA are combined as the feature MPMC (Multi-scale PCA and Multi-scale CLBP), which is proposed in this paper. Finally, support vector machine (SVM) with polynomial kernel is used for classification. Experiments show that the classification accuracy based on MPMC outperforms the previous methods in the literature for the same virus image data set.

Multi-spectral palmprint recognition based on oriented multiscale log-Gabor filters

Among several palmprint recognition methods proposed recently, coding-based approaches using multi-spectral palmprint images are attractive owing to their high recognition rates. Aiming to further improve the performance of these approaches, this paper presents a novel multi-spectral palmprint recognition approach based on oriented multiscale log-Gabor filters. The proposed method aims to enhance the recognition performances by proposing novel solutions at three stages of the recognition process. Inspired by the bitwise competitive coding, the feature extraction employs a multi-resolution log-Gabor filtering where the final feature map is composed of the winning codes of the lowest filters’ bank response. The matching process employs a bitwise Hamming distance and Kullback–Leibler divergence as novel metrics to enable an efficient capture of the intra- and inter-similarities between palmprint feature maps. Finally, the decision stage is carried pout using a fusion of the scores generated from different spectral bands to reduce overlapping. In addition, a fusion of the feature maps through two proposed novel feature fusion techniques to allow us to eliminate the inherent redundancy of the features of neighboring spectral bands is also proposed. The experimental results obtained using the multi-spectral palmprint database MS-PolyU have shown that the proposed method achieves high accuracy in mono-spectral and multi-spectral recognition performances for both verification and identification modes; and also outperforms the state-of-the-art methods.

Multiscale morphological manifold for rolling bearing fault diagnosis

The vibration signals of fault rolling bearing are high-dimensional information with complex components. In order to identify different classes of bearing fault, a new multiscale morphological manifold method based on multiscale morphology and manifold learning is proposed. The multiscale morphological manifold method consists of three main steps. Firstly, multiscale difference filter based on multiscale morphological transformation is applied to obtain multiscale observation results of each signal sample. Secondly, the nonlinear feature vectors of each signal sample are constructed according to the observation approach. Finally, manifold learning is introduced to extract the low-dimensional multiscale morphological manifold features through reducing the dimension of nonlinear features. The low-dimensional multiscale morphological manifold features can reveal the differences of signal classes, which are applicable for fault diagnosis. The performance of proposed method is tested by experimental data from bearings with different types of defects. Experimental verifications confirm that the proposed method is applicable and effective for rolling bearing fault diagnosis.

Road Centerline Extraction via Semisupervised Segmentation and Multidirection Nonmaximum Suppression

Accurate road centerline extraction from remotely sensed images plays a significant role in road map generation and updating. In the road extraction problem, the acquisition of labeled data is time consuming and costly; thus, there are only a small amount of labeled samples in reality. In the existing centerline extraction algorithms, the thinning-based algorithms always produce small spurs that reduce the smoothness and accuracy of the road centerline; the regression-based algorithms can extract a smooth road network, but they are time consuming. To solve the aforementioned problems, we propose a novel road centerline extraction method, which is constructed based on semisupervised segmentation and multiscale filtering (MF) and multidirection nonmaximum suppression (M-NMS) (MF&M-NMS). Specifically, a semisupervised method, which explores the intrinsic structures between the labeled samples and the unlabeled ones, is introduced to obtain the segmentation result. Then, a novel MF&M-NMS-based algorithm is proposed to gain a smooth and complete road centerline network. Experimental results on a public data set demonstrate that the proposed method achieves comparable or better performances by comparing with the state-of-the-art methods. In addition, our method is nearly ten times faster than the state-of-the-art methods.

Multiscale Gaussian Derivative Functions for Hyperspectral Image Feature Extraction

A method for extracting spatiospectral features from hyperspectral (HS) data is proposed. Based on 2-D Gaussian derivative (GD) functions, a bank of 2-D filters is designed. These filters are utilized with various scale parameters to form a multiscale filter bank. This filter bank is applied on a few principal components extracted from HS data, resulting in a set of features called GD features. Four different scenarios are examined to employ these features in HS image classification. Some experiments are conducted on three well-known HS data sets. Based on the results, one scenario is adopted as the proposed framework. The effect of differentiation order of GD filters on classification accuracies is investigated. The proposed method is compared with some state-of-the-art spatiospectral methods. The experimental results confirm the remarkable performance of GD features in HS image classification.

An integrated method for ancient Chinese tablet images de-noising based on assemble of multiple image smoothing filters

There are unavoidably lots of noises in tablet images due to natural or man-made decay, which have a significant affect on learning and studying of the ancient Chinese calligraphy works with Chinese tablet images. To address this problem, an integrated de-noising method, based on assemble of multiple image smoothing filters, is proposed in this paper. To avoid damaging characters and losing detail information, input Chinese tablet images are enhanced by the Guided filter and multi-scale Retinex filter firstly. Then the enhanced tablet images are converted to binary ones by the Otsu thresholding filter. Finally, most random and block noises are removed using an improved scan-length statistics filter based on connected region. The performance of the proposed method was validated on our Chinese tablet image data set, which consists of 200 Chinese tablet images with different kinds of noise. Experiments show that, the proposed method can effectively remove most image noise (including various block noise, linear noise and ant-like noise) and preserve characters better than existing methods.

Multiscale Estimation of Leaf Area Index from Satellite Observations Based on an Ensemble Multiscale Filter

Currently, multiple leaf area index (LAI) products retrieved from remote sensing data are widely used in crop growth monitoring, land-surface process simulation and studies of climate change. However, most LAI products are only retrieved from individual satellite observations, which may result in spatial-temporal discontinuities and low accuracy in these products. In this paper, a new method was developed to simultaneously retrieve multiscale LAI data from satellite observations with different spatial resolutions based on an ensemble multiscale filter (EnMsF). The LAI average values corresponding to the date of satellite observations were calculated from the multi-year Moderate Resolution Imaging Spectroradiometer (MODIS) LAI product and were used as a priori knowledge for LAI in order to construct an initial ensemble multiscale tree (EnMsT). Satellite observations obtained at different spatial resolutions were then applied to update the LAI values at each node of the EnMsT using a two-sweep filtering procedure. Next, the retrieved LAI values at the finest scale were used as a priori knowledge for LAI for the new round of construction and updating of the EnMsT, until the sum of the difference of LAI values at each node of the EnMsT between two adjacent updates is less than a given threshold. The method was tested using Thematic Mapper (TM) or Enhanced Thematic Mapper Plus (ETM+) surface reflectance data and MODIS surface reflectance data from five sites that have different vegetation types. The results demonstrate that the retrieved LAI values for each spatial resolution were in good agreement with the aggregated LAI reference map values for the corresponding spatial resolution. The retrieved LAI values at the coarsest scale provided better accuracy with the aggregated LAI reference map values (root mean square error (RMSE) = 0.45) compared with that obtained from the MODIS LAI values (RMSE = 1.30).

Curvelet initialized level set cell segmentation for touching cells in low contrast images

Cell segmentation is an important element of automatic cell analysis. This paper proposes a method to extract the cell nuclei and the cell boundaries of touching cells in low contrast images. First, the contrast of the low contrast cell images is improved by a combination of multiscale top hat filter and h-maxima. Then, a curvelet initialized level set method has been proposed to detect the cell nuclei and the boundaries. The image enhancement results have been verified using PSNR (Peak Signal to noise ratio) and the segmentation results have been verified using accuracy, sensitivity and precision metrics. The results show improved values of the performance metrics with the proposed method.

Study of Single Scale Spatial Adaptive Filters, Diffusion Filters and Multi Scale Filters for Speckle Noise Suppression

Ultrasound imaging owing to its low cost and non-invasive nature is a widely accepted imaging technique. The image quality in ultrasound images is degraded by a special type of acoustic noise known as speckle. Speckle noise is a random presence of bright and dark spots which hinders human interpretation of images and computer assisted diagnostic techniques. The success of ultrasonic examination depends upon image quality; therefore, despeckling is necessary to improve visual quality for better diagnosis. Several speckle reduction methods are applied to the ultrasound images to suppress noise and preserve the useful diagnostic information. This paper presents a review of multi-scale filters (wavelets), single scale spatial adaptive filters (viz. Median, Wiener, Lee, Frost, Kuan, and Gamma MAP filter) and two Diffusion filters (viz. Anisotropic Diffusion filter and Speckle Reduction Anisotropic Diffusion filter)that are widely used for speckle reduction in biomedical ultrasound B-scan images. General Terms Digital Image Processing.

Generalised morphological image diffusion

Relationships between linear and morphological scale-spaces have been considered by various previous works. The aim of this paper is to study how to generalise the discrete and continuous diffusion-based approaches in order to introduce nonlinear filters whose limit effects mimic the asymmetric behaviour of morphological dilation and erosion, as well as other multiscale filters, hybrid between the standard linear and morphological filters. A methodology based on the counter-harmonic mean is adopted here. Partial differential equations are formulated and details of numerical implementation are discussed to illustrate the various studied cases: isotropic, edge-preserving and coherence-enhancing diffusion. We also found a new way to derive the classical link between Gaussian scale-space and dilation/erosion scale-spaces based on quadratic structuring functions in the discrete and continuous setting. We have included some preliminary applications of the generalised morphological diffusion to solve image processing problems such as denoising and image enhancement in the case of asymmetric bright/dark image properties.

A supervised texton based approach for automatic segmentation and measurement of the fetal head and femur in 2D ultrasound images

This paper presents a supervised texton based approach for the accurate segmentation and measurement of ultrasound fetal head (BPD, OFD, HC) and femur (FL). The method consists of several steps. First, a non-linear diffusion technique is utilized to reduce the speckle noise. Then, based on the assumption that cross sectional intensity profiles of skull and femur can be approximated by Gaussian-like curves, a multi-scale and multi-orientation filter bank is designed to extract texton features specific to ultrasound fetal anatomic structure. The extracted texton cues, together with multi-scale local brightness, are then built into a unified framework for boundary detection of ultrasound fetal head and femur. Finally, for fetal head, a direct least square ellipse fitting method is used to construct a closed head contour, whilst, for fetal femur a closed contour is produced by connecting the detected femur boundaries. The presented method is demonstrated to be promising for clinical applications. Overall the evaluation results of fetal head segmentation and measurement from our method are comparable with the inter-observer difference of experts, with the best average precision of 96.85%, the maximum symmetric contour distance (MSD) of 1.46 mm, average symmetric contour distance (ASD) of 0.53 mm; while for fetal femur, the overall performance of our method is better than the inter-observer difference of experts, with the average precision of 84.37%, MSD of 2.72 mm and ASD of 0.31 mm.