Search Results (488)

Anomaly detection is crucial in large-scale industrial manufacturing as it helps to detect and localize defective parts. Pre-training feature extractors on large-scale datasets is a popular approach for this task. Stringent data security, privacy regulations, high costs, and long acquisition time hinder the development of large-scale datasets for training and benchmarking. Despite recent work focusing primarily on the development of new anomaly detection methods based on such extractors, not much attention has been paid to the importance of the data used for pre-training. This study compares representative models pre-trained with fractal images against those pre-trained with ImageNet, without subsequent task-specific fine-tuning. We evaluated the performance of eleven state-of-the-art methods on MVTecAD, MVTec LOCO AD, and VisA, well-known benchmark datasets inspired by real-world industrial inspection scenarios. Further, we propose a novel method to create a dataset by combining the dynamically generated fractal images creating a “Multi-Formula” dataset. Even though pre-training with ImageNet leads to better results, fractals can achieve close performance to ImageNet under proper parametrization. This opens up the possibility for a new research direction where feature extractors could be trained on synthetically generated abstract datasets mitigating the ever-increasing demand for data in machine learning while circumventing privacy and security concerns. Full article

The classification and recognition of features play a vital role in production and daily life; however, the current semantic segmentation of remote sensing images is hampered by background interference and other factors, leading to issues such as fuzzy boundary segmentation. To address these challenges, we propose a novel module for encoding and reconstructing multi-dimensional feature layers. Our approach first utilizes a bilinear interpolation method to downsample the multi-dimensional feature layer in the coding stage of the U-shaped framework. Subsequently, we incorporate a fractal curve module into the encoder, which aggregates points on feature maps from different layers, effectively grouping points from diverse regions. Finally, we introduce an aggregation layer that combines the upsampling method from the UNet series, employing the multi-scale censoring of multi-dimensional feature map outputs from various layers to efficiently capture both spatial and feature information. The experimental results across diverse scenarios demonstrate that our model achieves excellent performance in aggregating point information from feature maps, significantly enhancing semantic segmentation tasks. Full article

The stability of open-pit mine slopes is crucial for safety, especially for spoil dump slopes, which are prone to cracks leading to landslides. This study investigates the use of xanthan gum (XG) to enhance the stability of mudstone in spoil dumps. Various concentrations of xanthan gum were mixed with mudstone and subjected to dry–wet cycle tests to assess the impact on crack development. Pore and crack analysis system (PCAS) was utilized for image recognition and crack analysis, comparing the efficiency of crack rate and length modification. The study found that xanthan gum addition significantly improved mudstone’s resistance to crack development post-drying shrinkage. A 2% xanthan gum content reduced the mudstone crack rate by 45% on average, while 1.5% xanthan gum reduced crack length by 46.2% and crack width by 26.3%. Xanthan gum also influenced the fractal dimension and water retention of mudstone cracks. The optimal xanthan gum content for mudstone modification was identified as between 1.5% and 2%. Scanning electron microscopy imaging and X-ray diffraction tests supported the findings, indicating that xanthan gum modifies mudstone by encapsulation and penetration in wet conditions and matrix concentration and connection in dry conditions. These results are expected to aid in the development of crack prevention methods and engineering applications for open-pit mine spoil dump slopes. Full article

To investigate the microscopic pore evolution characteristics of Zhanjiang Formation structural clay during the disturbance process, unconfined compressive strength tests, scanning electron microscopy (SEM), and X-ray diffraction (XRD) were conducted on disturbed samples subjected to various disturbance conditions after vibrational disturbance. Based on the evolution characteristics of the microstructure, the microscopic pore characteristics of the disturbance damage of Zhanjiang Formation structural clay were examined. The results indicate the following. (1) The porosity in three-dimensional visualization images of the microstructure reconstructed by ArcGIS 10.1 increases with the disturbance degree, showing a linear growth trend. (2) The correlation analysis between macroscopic mechanics and microscopic pores shows that the unconfined compressive strength of Zhanjiang Formation structural clay is mainly affected by its porosity, with a significant linear negative correlation. Based on this, a reasonable regression model between the microscopic porosity and the unconfined compressive strength has been established. The model can rapidly estimate the unconfined compressive strength from porosity data, facilitating the assessment engineering properties of the soil. (3) The microscopic pore structure of Zhanjiang Formation structural clay exhibits prominent Menger fractal characteristics. The three-dimensional pore fractal dimension has a certain positive correlation with the disturbance degree, and can be utilized to characterize the pore structure and complexity, serving as a significant parameter for the quantitative evaluation of the pore structure characteristics of Zhanjiang Formation structural clay. Consequently, the complexity of the pore structure of the engineering soil can be evaluated by the pore fractal dimension. (4) The impact of disturbance on soil is primarily manifested in the structural changes in secondary clay minerals, transitioning from a relatively intact to a fully adjusted state. During this process, interparticle pores continuously increase, pore structure complexity increases, and interparticle cementation weakens, resulting in the continuous degradation of unconfined compressive strength. This study contributes to a deeper understanding of the disturbance damage characteristics of Zhanjiang Formation structured clays from a microscopic pore perspective, providing a theoretical basis for the engineering construction and operational maintenance in regions with Zhanjiang Formation structural clay. Full article

This review considers the use of fractal concepts to improve the development, fabrication, and characterisation of catalytic materials and supports. First, the theory of fractals is discussed, as well as how it can be used to better describe often highly complex catalytic materials and enhance structural characterisation via a variety of different methods, including gas sorption, mercury porosimetry, NMR, and several imaging modalities. The review then surveys various synthesis and fabrication methods that can be used to create catalytic materials that are fractals or possess fractal character. It then goes on to consider how the fractal properties of catalysts affect their performance, especially their overall activity, selectivity for desired products, and resistance to deactivation. Finally, this review describes how the optimum fractal catalyst material for a given reaction system can be designed on a computer. Full article

With recent advancements in deep learning, spoofing techniques have developed and generative adversarial networks (GANs) have become an emerging threat to finger-vein recognition systems. Therefore, previous research has been performed to generate finger-vein images for training spoof detectors. However, these are limited and researchers still cannot generate elaborate fake finger-vein images. Therefore, we develop a new densely updated contrastive learning-based self-attention generative adversarial network (DCS-GAN) to create elaborate fake finger-vein images, enabling the training of corresponding spoof detectors. Additionally, we propose an enhanced convolutional network for a next-dimension (ConvNeXt)-Small model with a large kernel attention module as a new spoof detector capable of distinguishing the generated fake finger-vein images. To improve the spoof detection performance of the proposed method, we introduce fractal dimension estimation to analyze the complexity and irregularity of class activation maps from real and fake finger-vein images, enabling the generation of more realistic and sophisticated fake finger-vein images. Experimental results obtained using two open databases showed that the fake images by the DCS-GAN exhibited Frechet inception distances (FID) of 7.601 and 23.351, with Wasserstein distances (WD) of 18.158 and 10.123, respectively, confirming the possibility of spoof attacks when using existing state-of-the-art (SOTA) frameworks of spoof detection. Furthermore, experiments conducted with the proposed spoof detector yielded average classification error rates of 0.4% and 0.12% on the two aforementioned open databases, respectively, outperforming existing SOTA methods for spoof detection. Full article

Urban land use classification plays a significant role in urban studies and provides key guidance for urban development. However, existing methods predominantly rely on either raster structure deep features through convolutional neural networks (CNNs) or topological structure deep features through graph neural networks (GNNs), making it challenging to comprehensively capture the rich semantic information in remote sensing images. To address this limitation, we propose a novel urban land use classification model by integrating both raster and topological structure deep features to enhance the accuracy and robustness of the classification model. First, we divide the urban area into block units based on road network data and further subdivide these units using the fractal network evolution algorithm (FNEA). Next, the K-nearest neighbors (KNN) graph construction method with adaptive fusion coefficients is employed to generate both global and local graphs of the blocks and sub-units. The spectral features and subgraph features are then constructed, and a graph convolutional network (GCN) is utilized to extract the node relational features from both the global and local graphs, forming the topological structure deep features while aggregating local features into global ones. Subsequently, VGG-16 (Visual Geometry Group 16) is used to extract the image convolutional features of the block units, obtaining the raster structure deep features. Finally, the transformer is used to fuse both topological and raster structure deep features, and land use classification is completed using the softmax function. Experiments were conducted using high-resolution Google images and Open Street Map (OSM) data, with study areas on the third ring road of Shenyang and the fourth ring road of Chengdu. The results demonstrate that the proposed method improves the overall accuracy and Kappa coefficient by 9.32% and 0.17, respectively, compared to single deep learning models. Incorporating subgraph structure features further enhances the overall accuracy and Kappa by 1.13% and 0.1. The adaptive KNN graph construction method achieves accuracy comparable to that of the empirical threshold method. This study enables accurate large-scale urban land use classification with reduced manual intervention, improving urban planning efficiency. The experimental results verify the effectiveness of the proposed method, particularly in terms of classification accuracy and feature representation completeness. Full article

The natural fracture network within the coal body serves as the main pathway for gas migration, with its geometric characteristics significantly impacting coalbed methane flow. In order to enhance the numerical model for simulating coalbed methane flow based on fracture network images, we define porosity and permeability functions for these images and improve upon existing methods. By employing a pixel probability decomposition algorithm, we establish a geometric model of a rough discrete fracture network, which is imported into COMSOL Multiphysics to build a numerical model of gas flow. We analyze the impact of different fracture structures on coal seam permeability and find that gas primarily flows through interconnected fractures at much higher velocities compared to matrix pores. Furthermore, we observe that fracture network permeability increases with increasing porosity (0.0635–0.164), fractal dimension (1.571–1.755), maximum fracture branch length (0.0111–0.0249 m), and connectivity (0.808–2.789). Conversely, it decreases with an increasing fracture dip angle (1.61–88.39°) and tortuosity fractal dimension (1.0018–1.0195). Our simulation method based on fracture network imaging provides a simple yet feasible approach to simulate gas extraction while accurately capturing various stages in the extraction process, including the temporal and spatial evolution of gas velocity and pressure as well as differences between fractures and the coal matrix. Full article

Existing research on image-based plant classification has demonstrated high performance using artificial intelligence algorithms. However, limited camera viewing angles can cause parts of the plant to be invisible in the acquired images, leading to an inaccurate classification. However, this issue has not been addressed by previous research. Hence, our study aims to introduce a method to improve classification performance by taking these limitations into account; specifically, we incorporated both segmentation and classification networks structured as shallow networks to expedite the processing times. The proposed shallow plant segmentation network (Shal-PSN) performs adversarial learning based on a discriminator network; and a shallow plant classification network (Shal-PCN) with applied residual connections was also implemented. Moreover, the fractal dimension estimation is used in this study for analyzing the segmentation results. Additionally, this study evaluated the performance of the proposed Shal-PSN that achieved the dice scores (DSs) of 87.43% and 85.71% with PlantVillage and open leaf image (OLID-I) open datasets, respectively, in instances where 40–60% of plant parts were missing. Moreover, the results demonstrate that the proposed method increased the classification accuracy from 41.16% to 90.51% in the same instances. Overall, our approach achieved superior performance compared to the existing state-of-the-art classification methods. Full article

Background: The aim of this review is to highlight the new advance of predictive and explainable artificial intelligence for neuroimaging applications. Methods: Data came from 30 original studies in PubMed with the following search terms: “neuroimaging” (title) together with “machine learning” (title) or ”deep learning” (title). The 30 original studies were eligible according to the following criteria: the participants with the dependent variable of brain image or associated disease; the interventions/comparisons of artificial intelligence; the outcomes of accuracy, the area under the curve (AUC), and/or variable importance; the publication year of 2019 or later; and the publication language of English. Results: The performance outcomes reported were within 58–96 for accuracy (%), 66–97 for sensitivity (%), 76–98 for specificity (%), and 70–98 for the AUC (%). The support vector machine and the convolutional neural network registered the best performance (AUC 98%) for the classifications of low- vs. high-grade glioma and brain conditions, respectively. Likewise, the random forest delivered the best performance (root mean square error 1) for the regression of brain conditions. The following factors were discovered to be major predictors of brain image or associated disease: (demographic) age, education, sex; (health-related) alpha desynchronization, Alzheimer’s disease stage, CD4, depression, distress, mild behavioral impairment, RNA sequencing; (neuroimaging) abnormal amyloid-β, amplitude of low-frequency fluctuation, cortical thickness, functional connectivity, fractal dimension measure, gray matter volume, left amygdala activity, left hippocampal volume, plasma neurofilament light, right cerebellum, regional homogeneity, right middle occipital gyrus, surface area, sub-cortical volume. Conclusion: Predictive and explainable artificial intelligence provide an effective, non-invasive decision support system for neuroimaging applications. Full article

Fractal analysis is a mathematical approach employed to study and describe complex patterns and structures across various disciplines, including mathematics, physics, computer science, biology and finance. Introduced by mathematician Benoit Mandelbrot in the 1970s, fractals are intricate, self-similar patterns that repeat at different scales, exhibiting consistent structures upon magnification or reduction. This analysis involves generating fractals through iterative processes or recursive equations, resulting in highly detailed and complex formations. Fractal analysis enhances medical images by removing noise while preserving details and improving diagnostic quality in magnetic resonance and computed tomography scans. However, there is a lack of comprehensive studies on its application in ultrasound imaging, prompting this narrative review to investigate its use and methodology in this context. Selected papers on the use of fractal analysis in ultrasound imaging were analyzed. Out of 186 records screened, 60 duplicates were removed and 28 were discarded. The text content of 98 potentially eligible papers was checked, with 65 not meeting inclusion criteria. Finally, 33 studies were included in the review. Fractal analysis enhances ultrasound imaging by providing detailed tissue texture characterization, aiding in the diagnosis of conditions like breast and lung cancer, osteoporosis and hypertensive disorders in pregnancy. It quantifies biological structure complexity and improves diagnostic accuracy and reliability. This technique supports clinicians in making informed decisions by offering critical insights into various medical conditions. Full article

The accurate modeling of single-particle breakage based on three-dimensional (3D) images is crucial for understanding the particle-level mechanics of granular materials. This study aims to propose a systematic framework incorporating single-particle breakage experiments and numerical simulations based on a novel 3D particle reconstruction technique for fractal analysis of particle size and morphology in single-particle breakage. First, the vision foundation model is used to generate accurate particles from 3D images. The numerical approach is validated by simulating the single-particle breakage test with multiple Fujian sand particles. Then, the breakage processes of reconstructed sand particles under axial compression are numerically modeled. The relationship between 3D fractal dimensions and particle size, particle crushing strength, and morphology is meticulously investigated. Furthermore, the implications of these relationships on the particle breakage processes are thoroughly discussed, shedding light on the underlying mechanisms that govern particle breakage. The framework offers an effective way to investigate the breakage behavior of single sand particles, which will enhance understanding of the mechanism of the whole particle breakage process. Full article

In response to the challenges posed by long-term cyclic loading and unloading in underground rock engineering, this study systematically investigates the macro- and meso-mechanical response mechanisms of fractured rock masses under cyclic loading conditions. We performed graded cyclic loading–unloading tests on parallel double-fractured sandstone samples with varying spatial distribution configurations. These tests were integrated with digital image correlation (DIC) technology, fractal dimension analysis, and discrete element method (DEM) numerical simulations to analyze the mechanical properties, deformation characteristics, crack propagation features, and meso-fracture mechanisms of the fractured rock masses. The findings indicate that the diverse spatial distribution characteristics of the double fractures exert a significant influence on the loading–unloading processes, surface deformation fields, and fracture states of the rock. Cyclic loading leads to an increase in the fractal dimension of the fractured samples, resulting in more intricate and chaotic crack propagation patterns. Furthermore, DEM simulations reveal the impact of fracture spatial configurations on the force chain distribution within the rock bridges. The equivalent stress nephogram effectively represents the stress field distribution. This offers valuable insights for predicting meso-fracture trends in rocks. This paper comprehensively integrates both experimental and numerical simulation methodologies to deliver a thorough analysis of the complex mechanical behavior of fractured rock masses under cyclic loading conditions, with direct relevance to engineering applications such as mine excavation and slope stabilization. Full article

The damage degree of a reinforced concrete structure is closely related to the generation and expansion of cracks. However, the traditional damage assessment methods of reinforced concrete structures have defects, including low efficiency of crack detection, low accuracy of crack extraction, and dependence on the experience of inspectors to evaluate the damage of structures. Because of the above problems, this paper proposes a damage assessment method for concrete members combining the U-Net convolutional neural network and crack fractal features. Firstly, the collected test crack images are input into U-Net for segmenting and extracting the output cracks. The damage to the concrete structure is then classified into four empirical levels according to the damage index (DI). Subsequently, a linear regression equation is constructed between the fractal dimension (D) of the cracks and the damage index (DI) of the reinforced concrete members. The damage assessment is then performed by predicting the damage index using linear regression. The method was subsequently employed to predict the damage level of a reinforced concrete shear wall–beam combination specimen, which was then compared with the actual damage level. The results demonstrate that the damage assessment method for concrete members proposed in this study is capable of effectively identifying the damage degree of the concrete members, indicating that the method is both robust and generalizable. Full article

Using Landsat TM/OLI remote sensing images and social statistical data from 1995, 2000, 2005, 2010, 2015, and 2020, construction land information in Beijing’s main urban area was extracted with ArcGIS 10.4.1 and other software. Based on calculations of the expansion speed, expansion intensity, fractal dimension, and elasticity coefficient, the spatiotemporal expansion characteristics of the urban area of Beijing were analyzed to reveal the laws and driving forces of urban expansion in Beijing. The results showed that the urban construction land area in Beijing expanded by a factor of 0.53 from 1995 to 2020, and its expansion speed and intensity gradually slowed. The overall expansion trend is that the central urban area remains basically unchanged, while the peripheral areas are rapidly expanding, showing a trend of rapid growth first and then stable growth, and the urban layout is basically stable. The urban expansion of Beijing has led to increasingly complex, tortuous, and unstable boundaries. Overall, the center of gravity of Beijing is moving toward the northeast, and the elasticity coefficient of urban expansion is 1.67 times that of a reasonable coefficient. The intensity and direction of urban expansion in Beijing are most significantly related to population mobility. Research on the expansion of Beijing lies the foundation for the integration and coordinated planning of resources in the various districts of Beijing and provides a basis for its sustainable development. Full article