Morphological Filtering Research Articles

An intelligent system for lung CT image denoising using a hybrid WT-NLM filter

Image denoising is an extensively researched problem in computer vision that is commonly used as a benchmark for many imaging tasks. Noise distorts an image throughout the acquisition and transmission cycle. The elimination of noise from the original image is a significant challenge for scientists and this work considers Gaussian, Salt and pepper noises, as medical images are prone to it. Hence, this work presents ways for mitigating noise, while preserving the relevant image information. In this study, a Hybrid Wavelet Transform with Non-Local Mean (Hybrid WT-NLM) filter is proposed for reducing noise in lung CT images. The final denoised images are formed by the summation of NLM-filtered images and wavelet coefficients. The proposed study offers a straightforward method for removing noise to attain better visual quality in a reasonable period. The efficacy of the proposed denoising approach is proven by comparing it with traditional denoising approaches such as Morphological filtering, Mean Filter, Median filter, Adaptive median filter and Non-Local Means (NLM) upon two public datasets such as Public LIDC database and in-house clinical ICCN database. The average PSNR of the proposed work is 25.47 dB at a time consumption of 1.03 s.

Solving Multi-Criteria Shortest Path by Optimization with Morphological Filters

The challenge of determining the shortest path within a multimodal transportation network involves identifying the most efficient travel route while considering various interconnected modes of transportation, such as roads, railways, and public transit. This problem becomes increasingly complex when numerous criteria and modes are involved, complicating the decision-making process. This study proposes a novel approach to computing the shortest path in multimodal networks, focusing on four modes of transportation: metro, trams, buses, and taxis. The optimization criteria include distance, travel time, and monetary cost. The proposed method utilizes a new metaheuristic called Optimization by Morphological Filters (OMF), inspired by image processing techniques. This approach was compared with the Genetic Algorithms (GA) and the Non-Dominated Sorting Genetic Algorithm II (NSGA-II). Experiments were carried out using graph models of multimodal transport networks that closely resemble real-world scenarios varying in size. Furthermore, the proposed method was evaluated using a real network from the city of Lyon, France. The results demonstrate that the OMF approach performs well in terms of convergence to optimal solutions and computation time.

MixCFormer: A CNN–Transformer Hybrid with Mixup Augmentation for Enhanced Finger Vein Attack Detection

Finger vein recognition has gained significant attention for its importance in enhancing security, safeguarding privacy, and ensuring reliable liveness detection. As a foundation of vein recognition systems, vein detection faces challenges, including low feature extraction efficiency, limited robustness, and a heavy reliance on real-world data. Additionally, environmental variability and advancements in spoofing technologies further exacerbate data privacy and security concerns. To address these challenges, this paper proposes MixCFormer, a hybrid CNN–transformer architecture that incorporates Mixup data augmentation to improve the accuracy of finger vein liveness detection and reduce dependency on large-scale real datasets. First, the MixCFormer model applies baseline drift elimination, morphological filtering, and Butterworth filtering techniques to minimize the impact of background noise and illumination variations, thereby enhancing the clarity and recognizability of vein features. Next, finger vein video data are transformed into feature sequences, optimizing feature extraction and matching efficiency, effectively capturing dynamic time-series information and improving discrimination between live and forged samples. Furthermore, Mixup data augmentation is used to expand sample diversity and decrease dependency on extensive real datasets, thereby enhancing the model’s ability to recognize forged samples across diverse attack scenarios. Finally, the CNN and transformer architecture leverages both local and global feature extraction capabilities to capture vein feature correlations and dependencies. Residual connections improve feature propagation, enhancing the stability of feature representations in liveness detection. Rigorous experimental evaluations demonstrate that MixCFormer achieves a detection accuracy of 99.51% on finger vein datasets, significantly outperforming existing methods.

Mitigating the ferroresonance resulting from HILGF clearing in a remote area 10 kV power distribution network

AbstractTo solve the problem of high impedance line‐to‐ground fault, a solution using isolation transformers for subnetwork divisions was previously implemented in the power distribution network. As a result of that, the network has been experiencing ferroresonance more often. Through modelling and simulation based on its actual parameters on the PSCAD/EMTDC, the understanding of the situation is deepened. The ferroresonance originates from the network response to the high impedance line‐to‐ground fault clearing. The subnetwork division has certainly improved the network fault damping factor, but the line capacitance reduction it caused has exposed the system to ferroresonance. To mitigate the ferroresonance, high‐frequency components are removed from the zero‐sequence current using the morphological filtering; then, through edge detection, it is used as the input signal to the control unit. The control unit then takes the ferroresonance mitigation action intelligently. The simulation results indicate that the proposed method can restore ferroresonance‐induced overvoltage and overcurrent to normal levels within 0.05 s.

Automated mold defects classification in paintings: A comparison of machine learning and rule-based techniques.

Mold defects pose a significant risk to the preservation of valuable fine art paintings, typically arising from fungal growth in humid environments. This paper presents a novel approach for detecting and categorizing mold defects in fine art paintings. The technique leverages a feature extraction method called Derivative Level Thresholding to pinpoint suspicious regions within an image. Subsequently, these regions are classified as mold defects using either morphological filtering or machine learning models such as Classification and Regression Trees (CART) and Linear Discriminant Analysis (LDA). The efficacy of these methods was evaluated using the Mold Features Dataset (MFD) and a separate set of test images. Results indicate that both methods improve the accuracy and precision of mold defect detection compared to no classifier. However, the CART algorithm exhibits superior performance, increasing precision by 32% to 53% while maintaining high accuracy (96%) even with an imbalanced dataset. This innovative method has the potential to transform the approach to managing mold defects in fine art paintings by offering a more precise and efficient means of identification. By enabling early detection of mold defects, this method can play a crucial role in safeguarding these invaluable artworks for future generations.

Vectorized solar photovoltaic installation dataset across China in 2015 and 2020

To achieve carbon neutrality, solar photovoltaic (PV) in China has undergone enormous development over the past few years. PV datasets with high accuracy and fine temporal span are crucial to assess the corresponding carbon reductions. In this study, we employed the random forest classifier to extract PV installations throughout China in 2015 and 2020 using Landsat-8 imagery in Google Earth Engine. The results were further visually inspected and refined by morphological filtering, cavity filling and manual adjustment. Validation analysis revealed that the initial classification achieved an overall accuracy over 96% for both 2015 and 2020. Further validation using independent test samples demonstrated that the final dataset outperformed the accuracies of existing PV datasets. In 2015, the total area of installed PV in China was 663.09 km2, which were mainly distributed in the northwest, Beijing-Tianjin-Hebei, and the Yangtze River Delta region. By 2020, the total area of PV reached to 2847.36 km2, with net increase of almost 3.3 times. Installed PV was intensified in the northwest and extended to eastern China.

Enhancing vector control: AI-based identification and counting of Aedes albopictus (Diptera: Culicidae) mosquito eggs

BackgroundDengue fever poses a significant global public health concern, necessitating the monitoring of Aedes mosquito population density. These mosquitoes serve as the disease vectors, making their surveillance crucial for dengue prevention. The objective of this study was to address the difficulty associated with identifying and counting mosquito eggs of wild strains during the monitoring of Aedes albopictus (Diptera: Culicidae) density via ovitraps in field surveys.MethodsWe constructed a dataset comprising 1729 images of Ae. albopictus mosquito eggs from wild strains and employed the Segment Anything Model to enhance the applicability of the detection model in complex environments. A two-stage Faster Region-based Convolutional Neural Network model was used to establish a detection model for Ae. albopictus mosquito eggs. The identification and counting process involved applying the tile overlapping method, while morphological filtering was employed to remove impurities. The model’s performance was evaluated in terms of precision, recall, and F1 score, and counting accuracy was assessed using R-squared and root mean square error (RMSE).ResultsThe experimental results revealed the model’s remarkable identification capabilities, achieving precision of 0.977, recall of 0.978, and an F1 score of 0.977. The R-squared value between the actual and identified egg counts was 0.997, with an RMSE of 1.742. The average detection time for a single tile was 0.48 s, which was more than 10 times as fast as the human–computer interaction method in counting an entire image.ConclusionsThe model demonstrated excellent performance in recognizing and counting Ae. albopictus mosquito eggs, indicating great application potential. This study offers novel technological support for enhancing vector control effectiveness and public health standards.Graphical

Investigating the potential of ICEYE-SAR data in storm damage detection in a coniferous forest with rugged terrain

ABSTRACT Adaptive forest management strategies require accurate detection of forest disturbance, in various spatial scales. Synthetic Aperture Radar (SAR) data can provide information about the forest attributes, penetrating the canopy at different levels under any weather and lighting conditions. ICEYE consists of the largest constellation of SAR satellites, enabling very high spatial and temporal resolution. In this study, ICEYE data were investigated in the detection of storm damage, in a fir forest with complex topography which was recently hit by the Daniel storm, resulting in severe damage to the forest structure (FS) and alluvium depositions (AD). To identify the best potential for storm damage detection using ICEYE data, an unsupervised change detection approach was employed combining wavelet transform and adaptive thresholds at spatial scales of 0.5 m (R05), 1 m (R1), 2 m (R2) and 3 m (R3). Additionally, two morphological filters were applied in best-performing resolutions to assess the impact of post-processing on the detection accuracy. Finally, FS and AD damage were investigated separately in order to provide detailed information about the detection capabilities of ICEYE. The results showcased that R1 (UA = 32.35, PA = 18.34 and K = 0.31) provided the best detection performance, followed by R05 (UA = 20.62, PA = 20.12 and K = 0.19). Furthermore, the employment of morphological filters slightly increased UA and Kappa metrics in both R1(UA = 33.40 and K = 0.32) and R05 (UA = 25.60 and K = 0.24), suggesting that post-processing is necessary to mitigate false detections. Regarding the investigation of AD and FS damage, it was revealed that post-processed ICEYE data in both R05 and R1 are capable of identifying AD with satisfying accuracy (UA = 47.41, PA = 22.36, K = 0.47), while FS damage detection is more challenging (UA = 10.15%-14.40%, PA = 14.55%-15.11%, and Kappa = 0.10-0.17). Overall, this study demonstrated that ICEYE can be used to detect storm-affected areas in mountainous forest ecosystems, especially in cases where detection with other methods is not feasible.

Investigation on morphological filtering via enhanced adaptive time-varying structural element for bearing fault diagnosis

Investigation on morphological filtering via enhanced adaptive time-varying structural element for bearing fault diagnosis

An automated system for 2D building detection from UAV-based geospatial datasets

The focus of this manuscript is on integrating optical images and laser point clouds carried on low-cost UAVs to create an automated system capable of generating urban city models. After pre-processing both datasets, we co-registered both datasets using the DLT transformation model. We estimated structure heights from the LiDAR dataset through a progressive morphological filter followed by removing bare ground. Unsupervised and supervised image classification techniques were applied to a six-band image created from the optical and LiDAR datasets. After finding building footprints, we traced their edges, outlined their borderlines, and identified their geometric boundaries through several image processing and rule-based feature identification algorithms. Comparison between manually digitized and automatically extracted buildings showed a detection rate of about 92.3 % with an average of 7.4 % falsely identified areas with the six-band image in contrast to classifying only the RGB image that detected about 63.2 % of the building pixels with 25.3 % pixels incorrectly identified. Moreover, our building detection rate with the 6-band image was superior to that attained by performing traditional image segmentation for only the LiDAR DEM. Shifts in the horizontal coordinates between corner points identified by a human operator and those detected by the proposed system were in the range of 10–15 cm. This is an improvement over traditional satellite and manned-aerial large mapping systems that have lower accuracies due to sensor limitations and platform altitude. These findings demonstrate the benefits of fusing multiple UAV remote sensing datasets over utilizing a single dataset for urban area mapping and 3D city modeling.

How relevant is the relationship between ostiole size and wasp head shape in the Ficus-Agaonidae mutualistic interaction?

The mutualism between Ficus species (Moraceae) and their pollinating wasps (Agaonidae) is a widely recognized coevolutionary model. In Ficus species from the Paleotropic, it has been determined that the ostiole acts as a morphological filter that affects the head dimensions of pollinating female wasps. Here, for the first time, the allometric relationship between ostiole size (diameter and length) and the shape of the head (length/width) of pollinating wasps is quantitatively explored in 6 Neotropical Ficus species (3 sect. Americanae and 3 sect. Pharmacosycea). In the case of sect. Americanae, wasp head shape was significantly correlated only with ostiole length, while in sect. Pharmacosycea both ostiole variables were correlated with head shape. The ordination analysis (NMDS) clearly reflected associations of these traits in species for both sections. The results support what has been interpreted in previous studies as reciprocal evolution between the analyzed traits, which contribute, along with other morphological and ecological traits, to the specificity between Ficus species and their pollinating wasps.

Automatic visual recognition, detection and classification of weeds in cotton fields based on machine vision

Crops and weeds are involved in a continuous competition for equal resources, which may result in a potential decrease in crop yields by up to 31% and an increase in the costs of agricultural inputs by up to 22% of cultivation. Weeds further impact crop production, and their detection is crucial for effective crop management. In this research, we targeted common weeds of cotton field, specifically i) Digitaria sanguinalis (L.) Scop, ii) Amaranthus retroflexus L., iii) Acalypha australis, L., iv) Cephalanoplos segetum, and v) Chenopodium album L. Additionally, image processing techniques such as grayscale conversion, binarization, and Gaussian and morphological filters were also utilized. These methods are based on machine vision and facilitate rapid and straightforward weed detection by segmenting, scrutinizing, and comparing input images. The plant height and area were obtained during cotton planting within 32 days and fitted to develop the growth law concerning planting days for achieving the function of distinguishing cotton from weeds. We conducted recognition experiments by dividing images into four quadrants and categorizing weeds as either inter-row or intra-row. Meanwhile, the inter-row planting information was used to identify weeds, and the leaf pixel area and circularity were used as the identification methods for intra-row weeds, which reduced the algorithm's running time and improved real-time performance. The experimental results indicated that the inter-row weed recognition rate was 89.4%, with an average processing time of 102ms. Whereas in the case of intra-row weeds, the recognition rate was measured at 84.6%, and the overall recognition rate for cotton was 85.0%, with a mean time consumption of 437ms. Furthermore, the present research underscores recent advancements such as machine vision and high-resolution imaging, which have significantly improved the accuracy of automated weed identification in cotton fields while acknowledging ongoing challenges and outlining future opportunities. By Integrating state-of-the-art technology with sustainable agricultural practices, implementing an intelligent system offers a viable approach toward efficient and environmentally friendly weed management in modern agriculture.

An improved morphological filtering and feature enhancement method for rolling bearing fault diagnosis

Abstract The defects-induced periodic pulse is one of the important indices for the characterization of bearing failure. To solve the problem that the weak impact features caused by the early fault of the rolling bearing are easily to be interfered with by noise and strong background signal and are difficult to extract, an improved morphological filtering method combined with the Teager energy operator (TEO) is proposed to extract weak shock features. Firstly, according to the correlation between the periodic pulse induced by defects and the Morlet wavelet, the Morlet wavelet is used as the model to construct the structural elements. Then, capturing the Pearson correlation coefficient of the structural elements and the original signal and the signal is filtered by the variable scale morphological filter after threshold screening. Finally, the TEO is used as the post-enhancement link to suppress the noise in the signal after morphological processing and further highlight the fault characteristics. Simulation signals, experimental signals, and field signals verify the effectiveness and robustness of the proposed method.

Visual Enhancement of Aerial Flight in High Plateau Low Visibility Conditions

A method for automatic detection of airport runways in low visibility, forward and downward looking aerial airport images is proposed. Firstly, the recursive Otsu segmentation method is used to extract the main contour of the airport from the complex background; then the morphological filtering is performed by using the morphology-based open and swell operations, the open operation is used to filter out the isolated noise points and the swell operation is used to connect the segmented airport components into a complete whole; finally, the morphological filtered image is subjected to the region growth to localize the main runway of the airport. Finally, the morphology filtered image is processed to localize the main airport runway by its region growth. It is proved that this method can quickly and accurately detect and localize the main and auxiliary runways of airports in low visibility and forward and downward looking aerial images of airports.

Deep learning-based detection of qanat underground water distribution systems using HEXAGON spy satellite imagery

Qanats are a remarkable type of ancient hydraulic structure for sustainable water distribution in arid environments that use subterranean channels to transport water from highland or mountainous areas. The presence of the qanat system is marked by a line of regularly spaced shafts visible from the surface, which can be used to detect qanats using satellite imagery. Typically, qanats have been documented by field mapping or manual digitisation within a Geographic Information System (GIS) environment. This process is time-consuming due to the numerous shafts within each qanat line. However, several automated methods for detecting qanat structures have been explored, using techniques such as morphological filters, custom convolutional neural networks (CNN) and, more recently, YOLOv5 and Mask R-CNN. These approaches used high-resolution RGB images and CORONA images. However, the use of black and white CORONA in CNNs has been limited in its applicability due to a high rate of false positives.This paper explores the potential of YOLOv9 in processing the black and white HEXAGON (KH-9) high-resolution spy satellite system launched in 1971. Two areas in Afghanistan (Maiwand) and Iran (Gorgan Plain) were selected to train the system images extracted from HEXAGON imagery and artificial synthetic data. The training dataset was augmented using the Albumentation library, which increased the number of tiles used. The model was tested using two types of HEXAGON imagery for selected areas in Afghanistan (Maiwand), Iran (Gorgan Plain) and Morocco (Rissani), and CORONA imagery in Iran (Gorgan Plain).Our study provided a model capable of predicting the location of qanat shafts with a precision of over 0.881 and a recall of 0.627 for most of the case studies tested. This is the first case study aimed at detecting qanats in different landscapes using different types of satellite imagery. Using real, augmented, and artificial data allowed us to generalise the representation of qanats into lineal groups of circular features. Thanks to applying labelling for individual qanats and their pairs as separate classes, our approach eliminated most of the isolated and clustered false positives.

Mathematical Morphology on Directional Data

We define morphological operators and filters for directional images whose pixel values are unit vectors. This requires an ordering relation for unit vectors which is obtained by using depth functions. They provide a centre-outward ordering with respect to a specified centre vector. We apply our operators on synthetic directional images and compare them with classical morphological operators for grey-scale images. As application examples, we enhance the fault region in a compressed glass foam and segment misaligned fibre regions of glass fibre-reinforced polymers.

Monitoring sedimentation of magnetorheological fluids using an infrared night vision wide-angle lens visual monitoring system with dynamic calibration method

Magnetorheological fluids (MRFs) have demonstrated remarkable potential in engineering vibration damping. Nonetheless, the substantial density difference between the dispersed phase and the dispersing medium causes inevitable sedimentation in MRFs, significantly affecting the material’s service life and restricting its engineering applications. Monitoring technology for MRF sedimentation plays a crucial role in comprehending sedimentation behavior and holds significant importance for the subsequent redispersion of MRFs. Consequently, the monitoring technology for MRFs has been gaining increasing attention in recent times. This study proposes a machine vision-based real-time and in situ monitoring method to test and evaluate the suspension stability of MRFs. Firstly, an infrared wide-angle night vision lens (IWNVL) is fixed at the side of a MRF column for capturing the sequential MRF images during its sediment process. Then, the gray values (GVs) could be obtained by utilizing Gaussian filtering and morphological filtering. Secondly, by combining the Kynch sedimentation theory and Lambert-Beer composite material optical transmission theory, an analytic model is established in this works to find the relationship between the MRFs’ concentration gradient to its GVs. In what follows, the concentration distribution varied with time could be tested by the GVs of the sequential MRF images. Finally, a sentimental experiment for prepared MRF column is carried out in this works, and the testing results are verified and discussed by a capacitance sensor and a simple visual observation. The experimental results demonstrate that the proposed method could accurately measure the concentration distribution during MRFs’ sedimentation process under different initial light intensities.

GENETIC ALGORITHM FOR SELECTING OPERATIONS FOR LOCALIZATION OF OBJECTS ON COMPLEX STRUCTURED IMAGES

A genetic algorithm has been developed that implements the automatic selection of operations for processing complex structured images at the stage of localization of point objects. The algorithm is distinguished by the representation of a set of operations in the form of a variable-length chromosome, combining the performance of both halftone and binary morphological transformations and filters, a developed fitness function that takes into account the recall and IoU measure together, and a set of genetic operators (crossover and special types of mutation) corresponding to the task. Dilation, erosion, opening, closing, and skeletonization were studied as morphological operations. Clipping by area, eccentricity, and dimensions of the major and minor axes of the equivalent ellipse were used as filters. The genetic algorithm was launched and experimentally tested on scanned topographic maps of the USSR from 1965−1982, as typical representatives of images with a complex structure, together with sets of morphological operations obtained manually to determine objects of the “elevation mark” type. The quality criteria were the proportion of localized pixels, the number of missed images, and the processing time. For training on multiple images, the fitness function was modified in two ways: soft (averaging calculation for all images) and hard (zeroing the value of fitness function with a decrease in recall on at least one training image). The hard calculation of the fitness function turned out to be more effective, the final set of operations consists of 10 operations with a decrease in the number of localized pixels by 2,6 times. The results show that the use of a genetic algorithm for the selection of localization operations makes it possible to reduce the total number of localized pixels with reducing the overall operating time, taking into account the recognition stage, which allows us to conclude that it is consistent for image processing.

Cardboard packaging minimization: Autonomous box selection by RGB-depth vision system and complementary deep learning classification

Amidst the dramatic expansion of online shopping, wasted cardboard packaging has become a significant challenge in modern-day society. In this paper, we propose an innovative approach to address the issue of box packaging waste by efficiently selecting an optimal cardboard box for a set of items. Employing depth vision and deep learning classification techniques, the proposed system analyzes depth images to compute an optimal cuboid for unpackaged items while accounting for fragility characteristics. The system, facilitated by two Intel RealSense RGB-D cameras, first captures the overhead and sideview images to determine the dimensions of an unknown object. Then, the developed vision system utilizes averaging and morphological opening filters to denoise the acquired images, calculates the difference between the background image and object image, and employs thresholding by Otsu’s algorithm to find the minimum cuboid of the irregular object. After, a two-step deep learning model is used to determine the general category and characteristics of an object, considering factors, such as fragility, that may necessitate wrapping or separate packaging. Finally, a modified largest area first fit algorithm uses the outputted dimensions for irregular-shaped objects and packaged objects to determine the optimal fitting box from the standard Amazon box catalog. The computer simulation shows that the proposed method can be implemented in a cardboard packaging warehouse for optimal box selection, effectively minimizing cardboard waste.

Dual-Structure Elements Morphological Filtering and Local Z-Score Normalization for Infrared Small Target Detection against Heavy Clouds

Infrared (IR) small target detection in sky scenes is crucial for aerospace, border security, and atmospheric monitoring. Most current works are typically designed for generalized IR scenes, which may not be optimal for the specific scenario of sky backgrounds, particularly for detecting small and dim targets at long ranges. In these scenarios, the presence of heavy clouds usually causes significant false alarms due to factors such as strong edges, streaks, large undulations, and isolated floating clouds. To address these challenges, we propose an infrared dim and small target detection algorithm based on morphological filtering with dual-structure elements. First, we design directional dual-structure element morphological filters, which enhance the grayscale difference between the target and the background in various directions, thus highlighting the region of interest. The grayscale difference is then normalized in each direction to mitigate the interference of false alarms in complex cloud backgrounds. Second, we employ a dynamic scale awareness strategy, effectively preventing the loss of small targets near cloud edges. We enhance the target features by multiplying and fusing the local response values in all directions, which is followed by threshold segmentation to achieve target detection results. Experimental results demonstrate that our method achieves strong detection performance across various complex cloud backgrounds. Notably, it outperforms other state-of-the-art methods in detecting targets with a low signal-to-clutter ratio (MSCR ≤ 2). Furthermore, the algorithm does not rely on specific parameter settings and is suitable for parallel processing in real-time systems.