Morphological Filtering Research Articles

An Improved Time-Varying Morphological Filtering and Its Application to Bearing Fault Diagnosis

Morphological filtering (MF) is a typical fault feature extraction technology, which has been widely used in rolling bearing fault diagnostics. The structural element (SE) length has an important influence on feature extraction and noise removal, and the filtering result obtained by the filtered signal under a single SE length or the weighted average of different filtered signals under partial SE lengths is limited and insufficient. To address this problem, a transient feature extraction method called improved time-varying morphological filtering (ITVMF) is proposed for bearing fault diagnosis, wherein the SE length is adaptively determined according to the inherent characteristics of vibration signals. Moreover, the autocorrelation envelop spectrum (AES) is employed to further eliminate fault-unrelated components. The main innovation of this method is to automatically extract bearing fault-related transient features using a novel time-varying SE (TVSE) based on autocorrelation while effectively eliminating the interference of random impulses and strong background noise on bearing fault features. The developed ITVMF-AES is investigated and validated on the simulation signals, accelerated bearing degradation data, and artificial experimental test data, and compared with three existing feature extraction methods. The results demonstrate that the developed methodology is more efficient in excavating transient impulse features and identifying bearing faults than the comparison methods, and has potential engineering application value.

A machine learning approach for skin disease detection and classification using image segmentation

Skin diseases are common health problems around the world. The perils of the infections are invisible, which cause physical health distress as well as initiate mental depression. In addition, it sometimes leads to skin cancer in severe cases. Subsequently, diagnosing skin diseases from clinical images is one of the foremost challenging tasks in medical image analysis. Moreover, when performed manually by medical experts, diagnosing skin diseases is time-intensive and subjective. As a result, both patients and dermatologists require automatic skin disease prediction, which makes the treatments plan faster. In this work, we introduce a digital hair removal technique based on morphological filtering such as Black-Hat transformation and inpainting algorithm and then apply Gaussian filtering to de-blur or denoise the images. In addition, we apply the automatic Grabcut segmentation technique to segment out the affected lesions. For extracting underlying input patterns from the skin images, we apply the Gray Level Co-occurrence Matrix (GLCM) and statistical features techniques. Three computationally efficient machine learning techniques, Decision Tree (DT), Support Vector Machine (SVM), and K-Nearest Neighbor (KNN) classifiers are applied using the extracted features for effectively classifying the skin images as melanoma (MEL), melanocytic nevus (NV), basal cell carcinoma (BCC), actinic keratosis (AK), benign keratosis (BKL), dermatofibroma (DF), vascular lesion (VASC), and Squamous cell carcinoma (SCC). The models are validated using two standard datasets ISIC 2019 challenge and HAM10000. SVM performs slightly better than the other two classifiers. We have also compared our work with state-of-the-art methods.

Small Low-Contrast Target Detection: Data-Driven Spatiotemporal Feature Fusion and Implementation.

Detecting small low-contrast targets in the airspace is an essential and challenging task. This article proposes a simple and effective data-driven support vector machine (SVM)-based spatiotemporal feature fusion detection method for small low-contrast targets. We design a novel pixel-level feature, called a spatiotemporal profile, to depict the discontinuity of each pixel in the spatial and temporal domains The spatiotemporal profile is a local patch of the spatiotemporal feature maps concatenated by the spatial feature maps and temporal feature maps in channelwise, which are generated by the morphological black-hat filter and a ghost-free dark-focusing frame difference methods, respectively. Instead of the handcrafted feature fusion mechanisms in previous works, we use the labeled spatiotemporal profiles to train an SVM classifier to learn the spatiotemporal feature fusion mechanism automatically. To speed up detection for high-resolution videos, the serial SVM classification process on central processing units (CPUs) is reformed as parallel convolution operations on graphics processing unit (GPUs), which exhibits over 1000+ times speedup in our real experiments. Finally, blob analysis is applied to generate final detection results. Elaborate experiments are conducted, and experimental results demonstrate that the proposed method performs better than 12 baseline methods for the small low-contrast target detection. The field tests manifest that the parallel implementation of the proposed method can realize real-time detection at 15.3 FPS for videos at a resolution of 2048×1536 and the maximum detection distance can reach 1 km for drones in sunny weather.

Provide a Deep Convolutional Neural Network Optimized with Morphological Filters to Map Trees in Urban Environments Using Aerial Imagery

Provide a Deep Convolutional Neural Network Optimized with Morphological Filters to Map Trees in Urban Environments Using Aerial Imagery

Combination of morphological and distributional filtering for UAV - LiDAR point cloud density to establish the Digital Terrain Model

Filtering the LiDAR point cloud based the Unmaned Aerial Vehilce (UAV - LiDAR) in the dense land cover areas to build a Digital Terrain Model (DTM) is a basic requirement of large-scale topographic mapping. The aim of this paper is to study the use of the Simple Morphological Filter (SMRF) with suitable parameters to separate the non-terrain points (trees, noise points, etc.) and the topographical points. The methods of this article are algorithmic programming and combining the two filtering algorithms including SMRF and distributed filtering. The various data input was studied in the Ba Be case study. These parameters include the grid width called Gcell (m), the radius of filters called nwd and the threshold of the feature elevation called Eth (m). The point cloud of the terrain obtained after applying the SMRF continues to be filtered using distributional filter with the algorithm keeping only minimum elevation in the filtering window in order to remove the locations of high density of points. Then, it will contribute to lighten the point capacity to build DTM, to accurately interpolate the contour lines and to ensure the aesthetics of large-scale topographic maps. The results of the study are the fomulas to estimate reasonable input parameters (Gcell = 3 m, nwd = 3, Eth = 0.2 m) of the two filters for the establishment of a topographic map of 1:2000 scale, 1 m level in the Ba Be national forest, Bac Kan province, Vietnam.

Impact of Eastern Redcedar encroachment on water resources in the Nebraska Sandhills

Worldwide, tree or shrub dominated woodlands have encroached into herbaceous dominated grasslands. While very few studies have evaluated the impact of Eastern Redcedar (redcedar) encroachment on the water budget, none have analyzed the impact on water quality. In this study, we evaluated the impact of redcedar encroachment on the water budget in the Nebraska Sand Hills and how the decreased streamflow would increase nitrate and atrazine concentrations in the Platte River. We calibrated a Soil and Water Assessment Tool (SWAT model) for streamflow, recharge, and evapotranspiration. Using a moving window with a dilate morphological filter, encroachment scenarios of 11.9 %, 16.1 %, 28.0 %, 40.6 %, 57.5 %, 72.5 % and 100 % were developed and simulated by the calibrated model. At 11.9 % and 100 % encroachment, streamflow was reduced by 4.6 % and 45.5 %, respectively in the Upper Middle Loup River, a tributary to the Platte River. Percolation and deep aquifer recharge increased by 27 % and 26 % at 100 % encroachment. Streamflow in the Platte River, a major water source for Omaha and Lincoln, would decrease by 2.6 %, 5.5 % and 10.5 % for 28 %, 57.5 %, and 100 % encroachment of the Loup River watershed, respectively. This reduction in streamflow could increase nitrate and atrazine concentrations in the Platte River by 4 to 15 % and 4 to 30 %, respectively. While the density of redcedar is minimal, it is important to manage their encroachment to prevent reductions in streamflow and potential increases in pollutant concentrations.

Automatic detection of forest trees from digital surface models derived by aerial images

For the sustainable management of forests, obtaining the spatial information of the tree existence (location, number, height, and crown diameter of trees, etc.) with high accuracy and quickly is very important. In this context, the study aims to detect forest trees automatically through flow analysis applied to a 5 m resolution digital surface model by geospatial analysis. The study was carried out in five sample areas with different physical and topographic characteristics in the Antalya province of Turkey. The method consists of two steps which are identifying tree populations and determining tree peaks by applying flow analysis on the surface model. First, the canopy height model was extracted by applying a morphological filter to the image-based digital surface model. Then, the tree peak points are considered sink points, and these sink points were determined on the inverted surface model by the flow analysis approach which is frequently used in hydrological studies. The results showed that the applied method gives approximately 70% accuracy depending on the terrain conditions. Tree crown diameter, distance between trees, slope of the land, and digital surface model resolution significantly affect the accuracy of the results. It is predicted that this study will be an important guide for decision-makers in the preparation of forest plans.

Implementation of a Morphological Filter for Removing Spikes from the Epileptic Brain Signals to Improve Identification Ripples.

Epilepsy is a very common disease affecting at least 1% of the population, comprising a number of over 50 million people. As many patients suffer from the drug-resistant version, the number of potential treatment methods is very small. However, since not only the treatment of epilepsy, but also its proper diagnosis or observation of brain signals from recordings are important research areas, in this paper, we address this very problem by developing a reliable technique for removing spikes and sharp transients from the baseline of the brain signal using a morphological filter. This allows much more precise identification of the so-called epileptic zone, which can then be resected, which is one of the methods of epilepsy treatment. We used eight patients with 5 KHz data set and depended upon the Staba 2002 algorithm as a reference to detect the ripples. We found that the average sensitivity and false detection rate of our technique are significant, and they are ∼94% and ∼14%, respectively.

Fluoroscopic Image-Based 3-D Environment Reconstruction and Automated Path Planning for a Robotically Steerable Guidewire.

Cardiovascular diseases are the leading cause of death globally and surgical treatments for these often begin with the manual placement of a long compliant wire, called a guidewire, through different vasculature. To improve procedure outcomes and reduce radiation exposure, we propose steps towards a fully automated approach for steerable guidewire navigation within vessels. In this paper, we utilize fluoroscopic images to fully reconstruct 3-D printed phantom vasculature models by using a shape-from-silhouette algorithm. The reconstruction is subsequently de-noised using a deep learning-based encoder-decoder network and morphological filtering. This volume is used to model the environment for guidewire traversal. Following this, we present a novel method to plan an optimal path for guidewire traversal in three-dimensional vascular models through the use of slice planes and a modified hybrid A-star algorithm. Finally, the developed reconstruction and planning approaches are applied to an ex vivo porcine aorta, and navigation is demonstrated through the use of a tendon-actuated COaxially Aligned STeerable guidewire (COAST).

A Combined Approach to Infrared Small-Target Detection with the Alternating Direction Method of Multipliers and an Improved Top-Hat Transformation.

In infrared small target detection, the infrared patch image (IPI)-model-based methods produce better results than other popular approaches (such as max-mean, top-hat, and human visual system) but in some extreme cases it suffers from long processing times and inconsistent performance. In order to overcome these issues, we propose a novel approach of dividing the traditional target detection process into two steps: suppression of background noise and elimination of clutter. The workflow consists of four steps: after importing the images, the second step applies the alternating direction multiplier method to preliminarily remove the background. Comparatively to the IPI model, this step does not require sliding patches, resulting in a significant reduction in processing time. To eliminate residual noise and clutter, the interim results from morphological filtering are then processed in step 3 through an improved new top-hat transformation, using a threefold structuring element. The final step is thresholding segmentation, which uses an adaptive threshold algorithm. Compared with IPI and the new top-hat methods, as well as some other widely used methods, our approach was able to detect infrared targets more efficiently (90% less computational time) and consistently (no sudden performance drop).

Small and dim infrared moving target detection based on spatial–temporal saliency

Infrared target detection is a significant work in military, high accuracy and high speed are the essential capabilities for precise guidance. With limited conditions, making full use of spatial–temporal information can really improve performance. This article studied small and dim infrared moving target detection based on sequence images. Quaternion Fourier Transform is an efficient method of visual saliency. To make this algorithm more accurate, we introduced four preprocessing feature map as input channels for quaternion. They are selective inter frame difference, fuzzy-based morphology filter, joint local gradient and gray contrast measure and temporal-constrained Gaussian curvature filter. All of them aim to highlight small-dim target and help to construct saliency map. Simultaneously, a trajectory constraint is introduced to reduce false alarm rate. Compared with single-frame based methods and multi-frames based methods, our novel algorithm performed better on balancing time consuming and detection precision.

Blind Fault Extraction of Rolling-Bearing Compound Fault Based on Improved Morphological Filtering and Sparse Component Analysis.

In order to effectively separate and extract bearing composite faults, in view of the non-linearity, strong interference and unknown number of fault source signals of the measured fault signals, a composite fault-diagnosis blind extraction method based on improved morphological filtering of function (SMF), density peak clustering (DPC) and orthogonal matching pursuit (OMP) is proposed. In this method, the function is used as the structural element of the morphological filter for the first time to improve the traditional morphological filter. After the observation signal is processed by the improved morphological filter, the impact characteristics of the signal are improved, and the signal meets the sparsity. Then, on the premise that the number of clustering is unknown, the density peak algorithm is used to cluster sparse signals to obtain the clustering center, which is equivalent to the hybrid matrix. Finally, the hybrid matrix is transformed into a sensing matrix, and the signal is transformed into the frequency domain to complete the compressive sensing and reconstruction of the signal in the frequency domain. Both simulation and measured signal results show that this algorithm can effectively complete the blind separation of rolling bearing faults when the number of fault sources is unknown, and the time cost can be reduced by about 75%.

Complex evaluation of Raman spectra using morphological filtering: Algorithms, software implementation, and experimental verification of baseline correction, peak recognition, and cosmic ray removal in SERS spectra of designer drugs

AbstractWe discuss a full‐scale treatment of real‐life Raman spectra with pronounced artifacts via mathematical morphology: Baseline correction, peak recognition, and cosmic ray removal are provided using a single kernel. The suggested treatment is peak‐oriented, and its novelty lies in utilization of contact points between a spectrum and its morphological mapping to provide robust peak recognition via low‐order morphological operators. Reaching a compromise between complicated iterative processing and requiring tedious input from the user, the provided algorithms permit effective unsupervised evaluation of individual spectra as well as Raman maps and time or depth series. The method is demonstrated on SERS identification of selected designer drugs.

Materials property mapping from atomic scale imaging via machine learning based sub-pixel processing

Direct visualization of the atomic structure in scanning transmission electron microscopy has led to a comprehensive understanding of the structure-property relationship. However, a reliable characterization of the structural transition on a picometric scale is still challenging because of the limited spatial resolution and noise. Here, we demonstrate that the primary segmentation of atomic signals from background, succeeded by a denoising process, enables structural analysis in a sub-pixel accuracy. Poisson noise is eliminated using the block matching and three-dimensional filtering with Anscombe transformation, and remnant noise is removed via morphological filtering, which results in an increase of peak signal-to-noise ratio from 7 to 11 dB. Extracting the centroids of atomic columns segmented via K-means clustering, an unsupervised method for robust thresholding, achieves an average error of less than 0.7 pixel, which corresponds to 4.6 pm. This study will contribute to a profound understanding of the local structural dynamics in crystal structures.

Retinal Vessel Segmentation Based on B-COSFIRE Filters in Fundus Images.

Retinal vessel extraction plays an important role in the diagnosis of several medical pathologies, such as diabetic retinopathy and glaucoma. In this article, we propose an efficient method based on a B-COSFIRE filter to tackle two challenging problems in fundus vessel segmentation: (i) difficulties in improving segmentation performance and time efficiency together and (ii) difficulties in distinguishing the thin vessel from the vessel-like noise. In the proposed method, first, we used contrast limited adaptive histogram equalization (CLAHE) for contrast enhancement, then excerpted region of interest (ROI) by thresholding the luminosity plane of the CIELab version of the original RGB image. We employed a set of B-COSFIRE filters to detect vessels and morphological filters to remove noise. Binary thresholding was used for vessel segmentation. Finally, a post-processing method based on connected domains was used to eliminate unconnected non-vessel pixels and to obtain the final vessel image. Based on the binary vessel map obtained, we attempt to evaluate the performance of the proposed algorithm on three publicly available databases (DRIVE, STARE, and CHASEDB1) of manually labeled images. The proposed method requires little processing time (around 12 s for each image) and results in the average accuracy, sensitivity, and specificity of 0.9604, 0.7339, and 0.9847 for the DRIVE database, and 0.9558, 0.8003, and 0.9705 for the STARE database, respectively. The results demonstrate that the proposed method has potential for use in computer-aided diagnosis.

IMAGE ENHANCEMENT AND DE-NOISING TECHNIQUES OF MAGNETIC RESONANCE IMAGES

Medical images have noise, such as salt and pepper, speckle, and gaussian noise. So, getting the accuracy of magnetic resonance images is challenging always. The accuracy of brain images is essential for the clinical diagnosis process. The nonlinear method of median and morphological filters is used for enhancement, contrast adjustment, and histogram equalization for medical images and speech processing, to preserve them from noise and edge features. Because of its edge-preserving properties, this median spatial filter window is critically important and better for removing impulsive noise and speckle noise. Nine different shapes of windows and five different sizes of windows and five noise density injections are used for evaluating the performance of the median filter. The first person J.W.Tukey, did the smoothing process by median rank selection filter. The denoising using morphological process is also considered for analysis. These two nonlinear methods like median filtering and morphological methods are used for the restoration of noise from the original image. These two methods are applied in normal and seizure-affected images to calculate quality assessment metrics and also the performance evaluation metrics from magnetic resonance images. Any image needs to be denoised to get the accuracy, precision, and F-measure kind of common feature evaluation metrics. This paper evaluates two nonlinear filtering methods and their performance evaluated based on shapes and size of windows and for suppressing or de-noising to produce original noise-free images to assist clinically.

Quantification of worn surface using digital image processing

Digital Image processing can be used in many areas in some way or the other from medical to satellite applications. The efficiency of the machine is reduced and substantial energy loss was observed by mechanical wear of the contacting surfaces in materials rubbing against each other. The most common test method reported in the literature is pin-on-disc wear testing. The brass material is used for this study. The images of the worn material surface were obtained from the metallurgical inverted microscope. Adaptive Otsu’s global thresholding method is used as the first step of segmentation. The granulometry by opening and closing on binary image detects the wear shapes such as scars, grooves, and pits. Shape descriptors quantify the shape of the crater which is an irregular circular-shaped patch. Craters are detected by morphological filtering using standard deviation. The boundary extraction tool extracts the overall worn area and its distribution across the material surface. The results reveal that it is possible to quantify and identify the mechanism of wear by obtaining the images using a CCD camera instead of highly sophisticated instruments like scanning electron microscope or atomic field microscope (AFM) etc., by using digital image processing.

Full-circle localization diagnosis of rolling bearing outer ring defects based on double-vibration signal fusion

Localization diagnosis (LD) of rolling bearing outer ring defects plays an important role in failure cause analysis and residual life prediction. This study aims to solve three main issues in the existing LD methods. Firstly, a full-circle functional relationship between outer ring defect and localization index is established based on the dynamics simulation analysis of bearing system. Secondly, the generation mechanism and solution method of the positioning datum are determined using a novel finite element transient dynamic model. Furthermore, variation law of the initial direction of the fault impact under different angular positions is investigated. Lastly, a general localization rule applicable to outer ring defects in the full-circle range is proposed. Therefore, a full-circle localization formula for outer ring defects based on the characteristic fusion of horizontal–vertical signals is proposed. Simulation results verify the effectiveness of the proposed localization formula. Additionally, a morphological filter for LD is investigated to resolve the diagnostic error caused by noise interference in the actual signal. Experimental results show that the presented method can considerably improve diagnostic accuracy.

Automatic pore size measurements from scanning electron microscopy images of porous scaffolds

Pore sizes and distribution are amongst the main morphological characteristics of porous scaffolds which indicate the suitability of scaffolds for many biological applications. Scaffolds usually have complex structures and are designed to have a specific range of pore sizes appropriate for target cells. Pore sizes are commonly estimated manually or based on semi-automatic techniques requiring high level of human intervention. Such methods are time consuming and subject to error, mainly due to lack of consistency in the process and subjective nature of the results following operator involvement. In this work, we present a novel image processing method for the measurement pore size distribution (the main morphological characteristics of scaffolds) independent from their complexity. We use thresholding, based on the histogram analysis, to segment pore areas from scaffold, followed by morphological filters to separate pores from each other. This algorithm provides robust detection and measurement of pore sizes and the distribution. The performance of the algorithm is assessed using standard calibration kit which is used for calibration of Scanning Electron Microscopy (SEM) imaging systems. The results showed consistent output with 1.3% average error as compared against their true size.The algorithm was applied to 3D Apatite-Wollastonite scaffolds manufactured using the Thermally Induced Phase Separation technique. The results were robust and consistent with visual evaluation of SEM images. The algorithm also provides the morphology of each pore and, subsequently, offering further comprehension of the influence of microstructures across a range of fields, such as tissue engineering processes.

Cancer Identification in Walker 256 Tumor Model Exploring Texture Properties Taken from Microphotograph of Rats Liver

Recent studies have been evaluating the presence of patterns associated with the occurrence of cancer in different types of tissue present in the individual affected by the disease. In this article, we describe preliminary results for the automatic detection of cancer (Walker 256 tumor) in laboratory animals using preclinical microphotograph images of the subject’s liver tissue. In the proposed approach, two different types of descriptors were explored to capture texture properties from the images, and we also evaluated the complementarity between them. The first texture descriptor experimented is the widely known Local Phase Quantization (LPQ), which is a descriptor based on spectral information. The second one is built by the application of a granulometry given by a family of morphological filters. For classification, we have evaluated the algorithms Support Vector Machine (SVM), k-Nearest Neighbor (k-NN) and Logistic Regression. Experiments carried out on a carefully curated dataset developed by the Enteric Neural Plasticity Laboratory of the State University of Maringá showed that both texture descriptors provide good results in this scenario. The accuracy rates obtained using the SVM classifier were 96.67% for the texture operator based on granulometry and 91.16% for the LPQ operator. The dataset was made available also as a contribution of this work. In addition, it is important to remark that the best overall result was obtained by combining classifiers created using both descriptors in a late fusion strategy, achieving an accuracy of 99.16%. The results obtained show that it is possible to automatically perform the identification of cancer in laboratory animals by exploring texture properties found on the tissue taken from the liver. Moreover, we observed a high level of complementarity between the classifiers created using LPQ and granulometry properties in the application addressed here.