Morphological Filtering Method Research Articles

Satellite road extraction method based on RFDNet neural network

<abstract><p>The road network system is the core foundation of a city. Extracting road information from remote sensing images has become an important research direction in the current traffic information industry. The efficient residual factorized convolutional neural network (ERFNet) is a residual convolutional neural network with good application value in the field of biological information, but it has a weak effect on urban road network extraction. To solve this problem, we developed a road network extraction method for remote sensing images by using an improved ERFNet network. First, the design of the network structure is based on an ERFNet; we added the DoubleConv module and increased the number of dilated convolution operations to build the road network extraction model. Second, in the training process, the strategy of dynamically setting the learning rate is adopted and combined with batch normalization and dropout methods to avoid overfitting and enhance the generalization ability of the model. Finally, the morphological filtering method is used to eliminate the image noise, and the ultimate extraction result of the road network is obtained. The experimental results show that the method proposed in this paper has an average F1 score of 93.37% for five test images, which is superior to the ERFNet (91.31%) and U-net (87.34%). The average value of IoU is 77.35%, which is also better than ERFNet (71.08%) and U-net (65.64%).</p></abstract>

AUTOMATIC SHADOW DETECTION FOR HIGH-RESOLUTION REMOTE SENSING DATA

Abstract. Shadow can be casted by daylight or any other light sources. We will not get a clear and quality image if it's hovered by the shadow. Shadows are frequently formed in high-resolution satellite imagery by the limitations of the imaging environment and the presence of high rise structures, and this scenario is true especially in metropolitan regions. Shadow is one of the noteworthy evils in remotely sensed imagery which hinders the precision of information extraction and change identification. To attenuate the effects of shadow in high resolution imagery regarding their supplemental functions, our paper suggests a novel algorithm for shadow masking built on computational methods. Firstly we transformed the images from RGB space to CIELCh space model, next we evaluated a modified Specthem ratio, and then used multilevel thresholding. We also created shadow masks for areas having vegetation, water, and soil. Shadow mask noise was decreased by morphological techniques. The ratio of lighting for the shadowed and unshadowed areas is utilized to create shadow masks, which are then used to remove shadows from the source photos. The thresholding approach creates an initial shadow mask during the shadow detection step, and the morphological filtering method is used to remove the noise and incorrect shadow regions. We also vectorized the raster data which can be further applied for various other studies.

Hyperspectral Anomaly Detection via Low-Rank Decomposition and Morphological Filtering

To effectively detect anomalies and eliminate the influence of noise on anomaly detection (AD), we propose a hyperspectral AD method based on low-rank decomposition and morphological filtering (LRDMF). For one thing, given the different ways in which anomalies and noise occur in the spectral bands, a low-rank decomposition model is proposed to decompose the original hyperspectral image (HSI) into the background, anomaly, and noise components, where a superpixel segmentation method and the sparse representation (SR) model are used to construct a robust background dictionary. For another thing, considering that the anomalies in HSI possess small area characteristics, a morphological filtering method is applied to preserve the small connected components. Finally, anomalies are detected by jointly considering the LRDMF results. The experimental results conducted on two real hyperspectral datasets demonstrate that the proposed method outperforms some of the state-of-the-art methods.

Rolling bearing diagnosis based on an unbiased-autocorrelation morphological filter method

Traditional morphological filter methods (MFs) are commonly affected by deterministic signals and random impulses, so an adaptive unbiased-autocorrelation morphological filter (UAMF) combined with an autoregressive filter (AR-UAMF) is proposed to tackle the issue. The deterministic component of the raw signal is first removed using an autoregressive filter (AR), and then the random impulse interference is attenuated using the UAMF. Additionally, unbiased-autocorrelation kurtosis (AK) is developed to ascertain the optimum SE scale. This indicator has the advantage of being sensitive to fault-related impulses under the interference of deterministic signals, random impulses and white noise. Finally, the simulation and experiment analysis results demonstrate that AR-UAMF is much more powerful and effective than the other single-scale and multiscale MFs in recovering the fault-related impulses and performing fault diagnoses.

A New Approach to the Analysis of Water Treeing Using Feature Extraction of Vented Type Water Tree Images

In this study, vented type water trees were initiated and grown in laboratory environment. A smart test platform was used to accelerate the initiation and growth of vented type water trees. 6 kV/4 kHz voltage was applied to the specimens to initiate and grow water trees. Mel-frequency cepstral coefficients of the vented type water tree images are obtained after 2 h and 10 h of aging respectively. The insignificant regions in the vented type water tree images were removed by using morphological filtering method before MFCC feature extraction. Finally, the statistical values of these features were analyzed. Scatter plots of the standard deviations and mean values of the cepstral coefficients were plotted. As expected, it has been observed that the points in the scatter plot are clustered in a certain area. MFCC is a popular and frequently used feature extraction method in speech recognition, however there are some studies which employs MFCC as a successful feature extraction method in image processing applications. This study provides a new approach to the analysis of vented water treeing using image processing techniques. The other new approach is using MFCC as a feature extraction method in microscopic water tree images.

A Morphology Filter-Assisted Extreme-Point Symmetric Mode Decomposition (MF-ESMD) Denoising Method for Bridge Dynamic Deflection Based on Ground-Based Microwave Interferometry

Bridge dynamic deflection is an important indicator of structure safety detection. Ground-based microwave interferometry is widely used in bridge dynamic deflection monitoring because it has the advantages of noncontact measurement and high precision. However, due to the influences of various factors, there are many noises in the obtained dynamic deflection of bridges obtained by ground-based microwave interferometry. To reduce the impacts of noise for bridge dynamic deflection obtained with ground-based microwave interferometry, this paper proposes a morphology filter-assisted extreme-point symmetric mode decomposition (MF-ESMD) for the signal denoising of bridge dynamic deflection obtained by ground-based microwave interferometry. First, the original bridge dynamic deflection obtained with ground-based microwave interferometry was decomposed to obtain a series of intrinsic mode functions (IMFs) with the ESMD method. Second, the noise-dominant IMFs were removed according to Spearman’s rho algorithm, and the other decomposed IMFs were reconstructed as a new signal. Finally, the residual noises in the reconstructed signal were further eliminated using the morphological filter method. The results of both the simulated and on-site experiments showed that the proposed MF-ESMD method had a powerful signal denoising ability.

Comparisons of envelope morphological filtering methods and various regular algorithms for surface texture analysis

Comparisons of envelope morphological filtering methods and various regular algorithms for surface texture analysis

Improved morphological filtering method for interferograms

Improved morphological filtering method for interferograms

Application of Mathematical Morphological Filtering to Improve the Resolution of Chang’E-3 Lunar Penetrating Radar Data

As one of the important scientific instruments of lunar exploration, the Lunar Penetrating Radar (LPR) onboard China’s Chang’E-3 (CE-3) provides a unique opportunity to image the lunar subsurface structure. Due to the low-frequency and high-frequency noises of the data, only a few geological structures are visible. In order to better improve the resolution of the data, band-pass filtering and empirical mode decomposition filtering (EMD) methods are usually used, but in this paper, we present a mathematical morphological filtering (MMF) method to reduce the noise. The MMF method uses two structural elements with different scales to extract certain scale-range information from the original signal, at the same time, the noise beyond the scale range of the two different structural elements is suppressed. The application on synthetic signals demonstrates that the morphological filtering method has a better performance in noise suppression compared with band-pass filtering and EMD methods. Then, we apply band-pass filtering, EMD, and MMF methods to the LPR data, and the MMF method also achieves a better result. Furthermore, according to the result by MMF method, three stratigraphic zones are revealed along the rover’s route.

Rolling Bearing Fault Detection Using Autocorrelation Based Morpho-logical Filtering and Empirical Mode Decomposition

In order to identify incipient rolling bearing pitting fault characteristics, an autocorrelation based multi-structure elements difference morphological filter and empirical mode decomposition method of fault diagnosis is presented in this paper. Through the experiment of rolling bearing inner and outer ring pitting failure, the fault vibration frequency is extracted to verify the feasibility of this method. The superiority of this method is verified by comparing with the empirical mode decomposition method with autocorrelation based multi-structure element mixed morphological filter and without filter. DOI: http://dx.doi.org/10.5755/j01.mech.24.6.22471

Comparison of Baseline Removal Methods in ECG and Algorithm Implementation

The removal of baseline drift is a difficult and important point in ECG preprocessing. This article discusses several commonly used methods to suppress ECG baseline drift:median filtering, high-pass filtering, curve fitting, wavelet transform, ECG-based feature points and morphological filtering methods. Under the circumstances of the baseline drift serious or weak, several methods for data are analyzed and compared. This article compares several commonly used methods as well as evaluation of application effects to provide a reference for the selection of baseline drift removal methods for ECG signals.

A Two-Step Semiglobal Filtering Approach to Extract DTM From Middle Resolution DSM

Many filtering algorithms have been developed to extract the digital terrain model (DTM) from dense urban light detection and ranging data or the high-resolution digital surface model (DSM), assuming a smooth variation of topographic relief. However, this assumption breaks for a middle-resolution DSM because of the diminished distinction between steep terrains and nonground points. This letter introduces a two-step semiglobal filtering (TSGF) workflow to separate those two components. The first SGF step uses the digital elevation model of the Shuttle Radar Topography Mission to obtain a flat-terrain mask for the input DSM; then, a segmentation-constrained SGF is used to remove the nonground points within the flat-terrain mask while maintaining the shape of the terrain. Experiments are conducted using DSMs generated from Chinese ZY3 satellite imageries, verified the effectiveness of the proposed method. Compared with the conventional progressive morphological filter method, the usage of flat-terrain mask reduced the average root-mean-square error of DTM from 9.76 to 4.03 m, which is further reduced to 2.42 m by the proposed TSGF method.

CFAR based morphological filter design to remove clutter from GB‐SAR images: An application to real data

AbstractReconstructed ground based synthetic aperture radar (GB‐SAR) data generally presents high resolution images of objects on the ground. However ground and multipath reflections occurred from depression angle appear as a clutter signs in the image. Constant false alarm rate (CFAR) detector is commonly preferred threshold algorithm for removing the clutter. But there is challenge in tuning CFAR. By enlarging or shrinking the window size, the background data taken into account respectively increases or decreases, and thus the filtered image accordingly includes much clutter or less clutter with target information loss. In this study, a novel adaptive clutter removing procedure based on morphological filter (MF) is designed to compensate this tradeoff by applying after a large windowed detector. In order to test the proposed CFAR based MF (CFAR‐MF) method, two different examples of GB‐SAR experiments in which the images are reconstructed by back projection algorithm (BPA) were performed using stepped frequency continuous waveform operating. The performance of CFAR‐MF was then examined in terms of signal to clutter ratio and integrated side lobe ratio. The results show that the clutter remains of the CFAR detector are effectively cleared from the GB‐SAR image and targets were accurately detected at their true locations without any disorder.

Automatic Diagnosis of Obstructive Sleep Apnea/Hypopnea Events Using Respiratory Signals.

Obstructive sleep apnea is a sleep disorder which may lead to various results. While some studies used real-time systems, there are also numerous studies which focus on diagnosing Obstructive Sleep Apnea via signals obtained by polysomnography from apnea patients who spend the night in sleep laboratory. The mean, frequency and power of signals obtained from patients are frequently used. Obstructive Sleep Apnea of 74 patients were scored in this study. A visual-scoring based algorithm and a morphological filter via Artificial Neural Networks were used in order to diagnose Obstructive Sleep Apnea. After total accuracy of scoring was calculated via both methods, it was compared with visual scoring performed by the doctor. The algorithm used in the diagnosis of obstructive sleep apnea reached an average accuracy of 88.33%, while Artificial Neural Networks and morphological filter method reached a success of 87.28%. Scoring success was analyzed after it was grouped based on apnea/hypopnea. It is considered that both methods enable doctors to reduce time and costs in the diagnosis of Obstructive Sleep Apnea as well as ease of use.

A high-accuracy roundness measurement for cylindrical components by a morphological filter considering eccentricity, probe offset, tip head radius and tilt error

A morphological filter is proposed to obtain a high-accuracy roundness measurement based on the four-parameter roundness measurement model, which takes into account eccentricity, probe offset, probe tip head radius and tilt error. This paper analyses the sample angle deviations caused by the four systematic errors to design a morphological filter based on the distribution of the sample angle. The effectiveness of the proposed method is verified through simulations and experiments performed with a roundness measuring machine. Compared to the morphological filter with the uniform sample angle, the accuracy of the roundness measurement can be increased by approximately 0.09 μm using the morphological filter with a non-uniform sample angle based on the four-parameter roundness measurement model, when eccentricity is above 16 μm, probe offset is approximately 1000 μm, tilt error is approximately 1″, the probe tip head radius is 1 mm and the cylindrical component radius is approximately 37 mm. The accuracy and reliability of roundness measurements are improved by using the proposed method for cylindrical components with a small radius, especially if the eccentricity and probe offset are large, and the tilt error and probe tip head radius are small. The proposed morphological filter method can be used for precision and ultra-precision roundness measurements, especially for functional assessments of roundness profiles.

A Novel Multi-scale 3D Area Morphological Filtering Method for Airborne LiDAR Building Extraction

Automatic building extraction is one of the most important issues in the fields of geoscience and remote sensing. In this letter, by introducing the idea of area morphology to the analysis of 3-D point clouds, a novel approach for automatic building extraction from airborne LiDAR data was proposed. At first, single scale area opening and area closing operator was used to produce normalized point clouds. With the normalized point clouds as input, multi-scale area morphology was employed to obtain connected regions, and then tree points were removed by PCA based local structural analyzing technique. Finally, building regions were extracted by analyzing geometry properties of the obtained connected regions without tree points. Experiments for different terrains were conducted. And the corresponding experimental results are very promising.

Bone‐marrow densitometry: Assessment of marrow space of human vertebrae by single energy high resolution‐quantitative computed tomography

Accurate noninvasive assessment of vertebral bone marrow fat fraction is important for diagnostic assessment of a variety of disorders and therapies known to affect marrow composition. Moreover, it provides a means to correct fat-induced bias of single energy quantitative computed tomography (QCT) based bone mineral density (BMD) measurements. The authors developed new segmentation and calibration methods to obtain quantitative surrogate measures of marrow-fat density in the axial skeleton. The authors developed and tested two high resolution-QCT (HR-QCT) based methods which permit segmentation of bone voids in between trabeculae hypothesizing that they are representative of bone marrow space. The methods permit calculation of marrow content in units of mineral equivalent marrow density (MeMD). The first method is based on global thresholding and peeling (GTP) to define a volume of interest away from the transition between trabecular bone and marrow. The second method, morphological filtering (MF), uses spherical elements of different radii (0.1-1.2 mm) and automatically places them in between trabeculae to identify regions with large trabecular interspace, the bone-void space. To determine their performance, data were compared ex vivo to high-resolution peripheral CT (HR-pQCT) images as the gold-standard. The performance of the methods was tested on a set of excised human vertebrae with intact bone marrow tissue representative of an elderly population with low BMD. 86% (GTP) and 87% (MF) of the voxels identified as true marrow space on HR-pQCT images were correctly identified on HR-QCT images and thus these volumes of interest can be considered to be representative of true marrow space. Within this volume, MeMD was estimated with residual errors of 4.8 mg/cm(3) corresponding to accuracy errors in fat fraction on the order of 5% both for GTP and MF methods. The GTP and MF methods on HR-QCT images permit noninvasive localization and densitometric assessment of marrow fat with residual accuracy errors sufficient to study disorders and therapies known to affect bone marrow composition. Additionally, the methods can be used to correct BMD for fat induced bias. Application and testing in vivo and in longitudinal studies are warranted to determine the clinical performance and value of these methods.

Development of the hybrid adaptive method for noise reduction in the bitmap image of the part drawing

An original hybrid method FILTRATOR that allows to eliminate the noise in monochrome and binary bitmap images of the part drawings was proposed. The analysis of the characteristics of the scanned images of the part drawings and characteristic noise, not allowing to get a satisfactory result by standard noise reduction methods was performed. The main features of bitmap images of drawings are the presence of fine lines, filtering of which by standard methods leads to disruption of the integrity and coherence of the part contours, as well as specific types of noise and distortion.The peculiarity of the proposed hybrid method is its phased implementation. At the first stage, monochrome noise is eliminated by the automatic tone adjustment of the monochrome bitmap image of the part drawing based on the analysis of its brightness and contrast histograms. At the second stage, the remaining binary noise is eliminated by adaptive method, which involves selecting an effective combination of filtering methods: the contour mask method, modified aperture method kFill, median, morphological and logical filtering methods, as well as selecting structural elements based on the block-by-block evaluation of the image parameters: contour line thickness, noise type and level.The proposed hybrid method FILTRATOR effectively eliminates noise in binary bitmap images of the part drawings, while maintaining the integrity and continuity of contours.The comparative qualitative and quantitative testing results of the FILTRATOR method and spatial, frequency, and morphological filtering methods were given. The comparison was performed using the MSE, PSNR and UQI criteria. The result of the comparison showed the superiority of the FILTRATOR method in terms of filtering quality of artifacts in scanned bitmap images of the part drawings.

Application of Morphological Filtering Technology in Underwater Acoustic Signal Preprocessing

In order to extract useful information of underwater acoustic signals from noise, preprocessing steps are unavoidable. In view of trend and transient pulse interference exited in the original data inevitability, morphological filter method is applied and analyzed in this paper. Data processing results show that the morphological filter method is superior to the traditional filter method.

Shadow Detection and Reconstruction in High-Resolution Satellite Images via Morphological Filtering and Example-Based Learning

The shadows in high-resolution satellite images are usually caused by the constraints of imaging conditions and the existence of high-rise objects, and this is particularly so in urban areas. To alleviate the shadow effects in high-resolution images for their further applications, this paper proposes a novel shadow detection algorithm based on the morphological filtering and a novel shadow reconstruction algorithm based on the example learning method. In the shadow detection stage, an initial shadow mask is generated by the thresholding method, and then, the noise and wrong shadow regions are removed by the morphological filtering method. The shadow reconstruction stage consists of two phases: the example-based learning phase and the inference phase. During the example-based learning phase, the shadow and the corresponding nonshadow pixels are first manually sampled from the study scene, and then, these samples form a shadow library and a nonshadow library, which are correlated by a Markov random field (MRF). During the inference phase, the underlying land-cover pixels are reconstructed from the corresponding shadow pixels by adopting the Bayesian belief propagation algorithm to solve the MRF. Experimental results on QuickBird and WorldView-2 satellite images have demonstrated that the proposed shadow detection algorithm can generate accurate and continuous shadow masks and also that the estimated nonshadow regions from the proposed shadow reconstruction algorithm are highly compatible with their surrounding nonshadow regions. Finally, we examine the effects of the reconstructed image on the application of classification by comparing the classification maps of images before and after shadow reconstruction.