Edge Discontinuities Research Articles

Real-time tunnel lining leakage image semantic segmentation via multiple attention mechanisms

One of the key objectives in tunnel illness detection is identifying tunnel lining leakage, and deep learning-based image semantic segmentation approaches can automatically locate tunnel lining leakage. However, in order to meet the real-time processing needs of professional mobile inspection equipment, existing leakage image segmentation approaches have difficulties in identifying real-time, dealing with voids, and dealing with edge discontinuities in the leaking zone. To address the aforementioned issues, this study introduces the PP-LiteSeg-Attn model, which takes the real-time semantic segmentation model PP-LiteSeg-B as baseline model, and combines the multi-layer CBAM attention mechanism and the CoT attention mechanism. Using the publically available dataset Water-Leakage, we trained and validated the PP-LiteSeg-Attn model, and attained IoU and F1 values of 88.18% and 93.72%, respectively, outperforming similar models in both measures. Extensive experiments show that the segmentation speed of the PP-LiteSeg-Attn model reaches 112.28 FPS, which meets real-time requirements, and that the model can effectively solve problems such as the appearance of voids in the seepage area, discontinuity, and fuzzy segmentation of seepage edges. The PP-LiteSeg-Attn model is better applicable to complicated tunnel settings, offering technical references for real-time diagnosis of tunnel illnesses.

HORIZON: High-Resolution Semantically Controlled Panorama Synthesis

Panorama synthesis endeavors to craft captivating 360-degree visual landscapes, immersing users in the heart of virtual worlds. Nevertheless, contemporary panoramic synthesis techniques grapple with the challenge of semantically guiding the content generation process. Although recent breakthroughs in visual synthesis have unlocked the potential for semantic control in 2D flat images, a direct application of these methods to panorama synthesis yields distorted content. In this study, we unveil an innovative framework for generating high-resolution panoramas, adeptly addressing the issues of spherical distortion and edge discontinuity through sophisticated spherical modeling. Our pioneering approach empowers users with semantic control, harnessing both image and text inputs, while concurrently streamlining the generation of high-resolution panoramas using parallel decoding. We rigorously evaluate our methodology on a diverse array of indoor and outdoor datasets, establishing its superiority over recent related work, in terms of both quantitative and qualitative performance metrics. Our research elevates the controllability, efficiency, and fidelity of panorama synthesis to new levels.

The joint learning of multi-resolution feature for multi-class retinal vessel segmentation

The task of multi-class vessel segmentation on retinal images is the basis for the arteriovenous quantitative analysis, and plays an important role in the diagnosis and treatment of cerebrovascular diseases. Due to the intricate details and intertwining of the retinal vessels, traditional feature learning networks based on single-level resolution images are prone to the troubles from arteriovenous confusion and vascular edge discontinuity. To this end, we develop a paradigm of multi-level image resolution joint learning. This scheme overcomes the limitation of the methods depending on single-level image resolution on feature modeling. Specifically, we designed a cross-scale feature fusion network with a dual-branch structure that integrates global and local perspectives. This approach enables the extraction of retinal image features across multiple resolutions, effectively compensating for the vascular feature gaps inherent in single-resolution network models. This framework not only corrects the intra-segment misclassification, but also improves continuity by supplementing the details of vascular edge. Furthermore, the cross-scale fusion process of the network at multiple stages is conducive to its optimization and enhances the collaborative learning ability of dual-branch. Meanwhile, we use the generative adversarial structure as the backbone to supervise and constrain the aforementioned feature fusion results. Finally, extensive experiments are conducted on three publicly available datasets, DRIVE-AV, LES-AV, and HRF-AV. It is shown that the proposed scheme outperforms the current state-of-the-art methods significantly. The source code is available at https://github.com/Tang9867/Multi-Resolution-Learning.

Determining the stresses near crack tips in composite plates based on the Williams series and singular integral equations

The paper presents a numerical-analytical method developed to determine the stresses formed in composite plates near the tips of curved cracks by means of the Williams series. The coefficients of the terms of the Williams series and T-stresses have been calculated based on the derivatives of the displacement discontinuity of the crack edges. It has been shown that the calculations can be simplified implementing the Green's solutions by the Lekhnitskii method. The numerical algorithms based on Lobatto quadrature formulas have been used to solve the singular equations and determine both the coefficients the Williams power series and T-stresses. The proposed approach has been used to analyze rectilinear and curvilinear cracks in composite plates. In addition, the Green's solutions have been implemented to investigate the cracks located in a half-plane, near an elliptical hole, in rectangular plates, and in strip. One- and two-period problems have been considered for plates with cracks in the present calculations. The high accuracy of the performed calculations of the stresses near the crack tips in composite plates using elementary formulas contained by the Williams series has been confirmed on the test examples.

Bioverse: The Habitable Zone Inner Edge Discontinuity as an Imprint of Runaway Greenhouse Climates on Exoplanet Demographics

Long-term magma ocean phases on rocky exoplanets orbiting closer to their star than the runaway greenhouse threshold—the inner edge of the classical habitable zone—may offer insights into the physical and chemical processes that distinguish potentially habitable worlds from others. The thermal stratification of runaway planets is expected to significantly inflate their atmospheres, potentially providing observational access to the runaway greenhouse transition in the form of a habitable zone inner edge discontinuity in radius–density space. Here, we use Bioverse, a statistical framework combining contextual information from the overall planet population with a survey simulator, to assess the ability of ground- and space-based telescopes to test this hypothesis. We find that the demographic imprint of the runaway greenhouse transition is likely detectable with high-precision transit photometry for sample sizes ≳100 planets if at least ∼10% of those orbiting closer than the habitable zone inner edge harbor runaway climates. Our survey simulations suggest that, in the near future, ESA’s PLATO mission will be the most promising survey to probe the habitable zone inner edge discontinuity. We determine the survey strategies that maximize the diagnostic power of the obtained data and identify as key mission design drivers: (1) a follow-up campaign of planetary mass measurements and (2) the fraction of low-mass stars in the target sample. Observational constraints on the runaway greenhouse transition will provide crucial insights into the distribution of atmospheric volatiles among rocky exoplanets, which may help to identify the nearest potentially habitable worlds.

A grinding process method of solid end mills groove with segmented unequal helical angles by cutting edges offset and interference

Solid end mills with variable helical angles within a certain range can effectively mitigate cutting vibration and enhance the machining quality. However, the grinding process for these helical grooves is complex, and the existing disclosed process method cannot grind those helical grooves accurately. This study proposes a two-pass grinding method for the helical groove with variable helical angles. First, a cutting edge model with variable helical angles is established by edge offset to guide the grinding process. Second, to ensure the required cross-section parameters while avoiding overcutting and edge discontinuity, a novel optimization algorithm for determining the grinding wheel posture, and calculation rules of related process parameters for the two-pass grinding process are proposed. Finally, the verification is conducted and the results show that the maximum relative error of cross-section parameters is 1.254 % in simulation machining, which indicates the theoretical modeling of posture optimization and cutting edge are correct, and the two-pass grinding process is effective. Furthermore, this approach can be extended to grinding other complex structures with helical grooves.

Resilient EDGE Detection Method for Significant IMPULSE Noise

The present study introduces a novel technique for detecting edges in images by means of the Switching Adaptive Median and Fixed Weighted Mean (SAMFWM) filter, which proves to be highly effective in removing impulse noise compared to conventional denoising filters that are currently available, while preserving edge details, thus ensuring optimal edge detection. The performance of the proposed approach is assessed using a comprehensive analysis of different performance metrics, including Mean Square Error (MSE), Structural Similarity Index (SSIM), and Peak Signal-to-Noise Ratio (PSNR). In addition, the Sobel operator is used to detect the edges and Non-Maximum Suppression is used to track and thin the edges. These techniques are utilized to handle edge discontinuities and detect edges in the presence of high-intensity noise. Furthermore, the proposed approach outperforms other alternative techniques such as the Robert, Prewitt, and Canny edge detectors in effectively removing impulse noise, even at high levels.

Feature pyramid with attention fusion for edge discontinuity classification

Feature pyramid with attention fusion for edge discontinuity classification

Learning Traces by Yourself: Blind Image Forgery Localization via Anomaly Detection With ViT-VAE

Most of existing deep learning models for image forgery localization rely on a large number of high-quality labeled samples for training. The training procedures are performed off-line and without adaptation to the image under scrutiny. In this paper, we propose to perform run-time learning of the forgery traces from the suspicious image itself. To this aim, a Variational Auto-Encoder (VAE) model is learned to reconstruct small cliques of the suspicious image, and those cliques with anomalous larger reconstruction errors are therein identified as forged. To further enhance performance, Vision Transform (ViT) is employed as the VAE encoder, and multi-modal input information is explored by considering noise inconsistency, high-pass residual inconsistency, and edge discontinuity. Evaluation on widely used benchmark datasets shows our method outperforms existing blind methods by a large margin, and is competitive against approaches that use ground-truth for supervised training.

Fruit Classification using Colorized Depth Images

Fruit classification is a computer vision task that aims to classify fruit classes correctly, given an image. Nearly all fruit classification studies have used RGB color images as inputs, a few have used costly hyperspectral images, and a few classical ML-based have used colorized depth images. Depth images have apparent benefits such as invariance to lighting, less storage requirement, better foreground-background separation, and more pronounced curvature details and object edge discontinuities. However, the use of depth images in CNN-based fruit classification remains unexplored. The purpose of this study is to investigate the use of colorized depth images in fruit classification with four CNN models, namely, AlexNet, GoogleNet, ResNet101, and VGG16, and compare their performance and computational efficiency, as well as the impact of transfer learning. Depth images of apple, orange, mango, banana and rambutan (Nephelium Lappaceum) were manually collected using a depth sensor with sub-millimeter accuracy and subjected to jet, uniform, and inverse colorization to produce three sets of dataset. Results show that depth images can be used to train CNN models for fruit classification with ResNet101 achieving the best accuracy of 96%on the inverse dataset. It achieved 100% accuracy after transfer learning. GoogleNet showed the most significant improvement after transfer learning on the uniform dataset, at 12.27%. It also exhibited the lowest training and inference times. The results show the potential use of depth images for fruit classification and similar computer vision tasks.

Classification and Object Detection of 360° Omnidirectional Images Based on Continuity-Distortion Processing and Attention Mechanism

The use of 360° omnidirectional images has occurred widely in areas where comprehensive visual information is required due to their large visual field coverage. However, many extant convolutional neural networks based on 360° omnidirectional images have not performed well in computer vision tasks. This occurs because 360° omnidirectional images are processed into plane images by equirectangular projection, which generates discontinuities at the edges and can result in serious distortion. At present, most methods to alleviate these problems are based on multi-projection and resampling, which can result in huge computational overhead. Therefore, a novel edge continuity distortion-aware block (ECDAB) for 360° omnidirectional images is proposed here, which prevents the discontinuity of edges and distortion by recombining and segmenting features. To further improve the performance of the network, a novel convolutional row-column attention block (CRCAB) is also proposed. CRCAB captures row-to-row and column-to-column dependencies to aggregate global information, enabling stronger representation of the extracted features. Moreover, to reduce the memory overhead of CRCAB, we propose an improved convolutional row-column attention block (ICRCAB), which can adjust the number of vectors in the row-column direction. Finally, to verify the effectiveness of the proposed networks, we conducted experiments on both traditional images and 360° omnidirectional image datasets. The experimental results demonstrated that better performance than for the baseline model was obtained by the network using ECDAB or CRCAB.

The Constraints between Edge Depth and Uncertainty for Monocular Depth Estimation

The self-supervised monocular depth estimation paradigm has become an important branch of computer vision depth-estimation tasks. However, the depth estimation problem arising from object edge depth pulling or occlusion is still unsolved. The grayscale discontinuity of object edges leads to a relatively high depth uncertainty of pixels in these regions. We improve the geometric edge prediction results by taking uncertainty into account in the depth-estimation task. To this end, we explore how uncertainty affects this task and propose a new self-supervised monocular depth estimation technique based on multi-scale uncertainty. In addition, we introduce a teacher–student architecture in models and investigate the impact of different teacher networks on the depth and uncertainty results. We evaluate the performance of our paradigm in detail on the standard KITTI dataset. The experimental results show that the accuracy of our method increased from 87.7% to 88.2%, the AbsRel error rate decreased from 0.115 to 0.11, the SqRel error rate decreased from 0.903 to 0.822, and the RMSE error rate decreased from 4.863 to 4.686 compared with the benchmark Monodepth2. Our approach has a positive impact on the problem of texture replication or inaccurate object boundaries, producing sharper and smoother depth images.

Assessment of Optimization Methods for Aeroacoustic Prediction of Trailing-Edge Interaction Noise in Axisymmetric Jets

Our concern in this paper is in the fine-tuning of the arbitrary parameters within the upstream turbulence structure for the acoustic spectrum of a rapid-distortion theory (RDT)-based model of trailing-edge noise. RDT models are based on an appropriate asymptotic limit of the Linearized Euler Equations and apply when the interaction time of the turbulence with the surface edge discontinuity is small compared to the eddy turnover time. When an arbitrary transversely sheared jet mean flow convects a finite region of nonhomogeneous turbulence, the acoustic spectrum of the pressure field scattered by the trailing-edge depends on (among other things) the upstream turbulence via the Fourier transform of the correlation function, R22 (where subscript 2 refers to a co-ordinate surface normal to the plate). We show that the length and time scale parameters that govern the spatial and temporal de-correlation of R22 can be found using formal optimization methods to avoid any uncertainty in their selection by hand-tuning. We assess various optimization methods that are broadly categorized into an ‘evolutionary’ and ‘non-evolutionary’ paradigm. That is, we optimize the acoustic spectrum using the Multi-Start algorithm, Particle Swarm Optimization and the Multi-Population Adaptive Inflationary Differential Evolution Algorithm. The optimization is based upon different objective functions for the acoustic spectrum and/or turbulence structure. We show that this approach, while resulting in the total modest increase in computation time (on average 2 h), gives excellent prediction over most frequencies (within 2–4 dB) where the trailing-edge noise associated amplification in sound exists.

Infra-red line camera data-driven edge detector in UAV forest fire monitoring

The accurate prediction of the wildfire spread-rate is a challenging task, due to the high number of parameters involved and the underlying complex dynamic multi-physics processes which drive the phenomenon. For these reasons, data-driven prediction tools could be useful to provide a more accurate prediction of the fire front. In this scenario, systematic fire data gathering becomes crucial and using an Unmanned Aircraft Vehicle (UAV) is strategic to reduce considerably the risk associated with flying a manned aircraft into low visibility and extremely turbulent air, sustained by the fire-induced convective motions.Moreover the employment of the UAV is beneficial, as the possibility of flying at very low altitudes maximizes the on-board Electro-Optical (EO) sensor effectiveness. The aim is to develop a real time data-driven fire propagator to support wildfire fighting operations and to facilitate the risk assessment and decision making process. In order to collect data, the fire front position has to be measured using an infra-red (IR) camera so as to overcome the limitations associated to a visible camera in low visibility (smoky)conditions and night operations. To reduce the computational cost associated to the image processing, a Line Camera (LC) configuration has been preferred. Because of the mono-dimensionality of the measure, classical edge detector, like the Canny method, or contour algorithms, developed for 2D images, can not be applied. In this paper, a mono-dimensional noise-resistant algorithm for edge detection is presented. The generality of the proposed method opens the possibility to a variety of heterogeneous problems of different nature. The robustness of this algorithm resides in the use of known physical characteristics of the target of interest, to increase the feature edge discontinuity. Its straightforwardness guarantees fast computation, making it very attractive for real time image processing, remote sensing applications and for UAV surveillance tasks.

Wave Scattering by a Planar Junction Between Anomalously Reflecting Metasurface and Impedance Sheets

Metasurface technology has become one of the promising structures for controlling and manipulating of the phase and amplitude characteristics of the electromagnetic waves in recent years. For this reason, the determination of the exact diffracted fields is also crucial especially in the edge or discontinuity points of these types of materials. In the paper, the discontinuity problem by the junction between the layers of an anomalously reflecting metasurface and an impedance half-screen is taken into consideration. These two planar layers have a common edge discontinuity where the diffracted waves occur. The scattered waves from the structure are obtained by the novel technique of the method of transition boundary, which especially offers the relation between the diffracted and scattered geometrical optical waves at the transition boundaries. The diffracted wave expressions are derived by the technique of the method of transition boundary and the uniform wave expressions are obtained by the method of uniform geometrical theory of diffraction. The behaviors of the total wave and its components are analyzed numerically.

ZnO as an anti-reflective layer for GaAs based heterojunction solar cell

Currently, how to improve the efficiency of solar cells has attracted wide attention. ZnO film is one of the most effective films today, which can act as both emitter and anti-reflective coating of solar cells. In this paper, n-type ZnO/p-type GaAs solar cell is modeled by analyzing the band edge discontinuities, electric field distributions at the ZnO/GaAs interface and cell parameters with varying ZnO layer thickness, affinity values and carrier concentration. Moreover, in order to improve the band offset alignment at the heterojunction, Mg doped ZnO emitter is a possible alternative. Then, the thickness and carrier concentration of MgZnO emitter layer are studied and simulation results show stronger electric field, better fill factor and higher efficiency. After optimization of two solar cells by using Silvaco Atlas, it is observed that the conversion efficiencies of ZnO/GaAs and MgZnO/GaAs solar cells are 22.84% and 23.44% respectively.

Novel magic angle-step state and mechanism for restraining the path ripple of magnetorheological finishing

Computer-controlled sub-aperture polishing technology is limited by its propensity to introduce mid-spatial-frequency (MSF) error (ripple error), which significantly inhibits the performance improvement of the optical system. Here, we propose a novel processing form of the magnetorheological finishing (MRF) tool and its generation mechanism. There are specific path angles and steps, with only tens of microns bandwidth (‘magic’ angle step), under which a surface without significant path ripple can be stably realized without affecting the convergence of other frequency errors. The magic angle step is precisely obtained via the scale-variant filtered spectrum analysis. Using the two-dimensional (2-D) Taylor approximation of the tool influence function (TIF), we verify that in the magic angle step state, owing to the shape asymmetry and the edge discontinuity of the TIF, the convolution of TIFs presents a complementary effect. It is first observed in experiments that the MSF error can be abnormally increased by decreasing the path step around the magic angle step state. Therefore, the magic angle step and corresponding analysis model have the potential of being extended to other polishing tools. Furthermore, they can significantly benefit the manufacturing efficiency of high-power laser systems, huge telescopes, and lithography systems, as well as improve existing polishing technologies.

Scattering of electromagnetic waves by an impedance sheet junction in anisotropic plasma

The diffraction of electromagnetic plane waves by the edge discontinuity of an impedance sheet junction, which is located in an anisotropic plasma, is investigated. The problem is solved by the integral theory of diffraction. The scattering integral is obtained from the scattered geometric optics wave. The diffracted fields are derived by the edge point evaluation of the scattering integral. The evaluated field expressions are analyzed numerically.

Interaction of plane waves with the edge discontinuity of a half-sheet at the interface of two different media

The present work considers the interaction of plane waves with a conductive half-sheet at the boundary of left- and right-handed media. The shadow boundary shifts toward the lower surface of the half-plane because of the anomalous refraction property of left-handed medium. For this reason, the diffracted fields are modified slightly by dividing into its subfields in order to satisfy the boundary conditions. The uniform version of fields is obtained and the results are plotted and discussed for different sets of parameters.

Low-level multiscale image segmentation and a benchmark for its evaluation

In this paper, we present a segmentation algorithm to detect low-level structure present in images. The algorithm is designed to partition a given image into regions, corresponding to image structures, regardless of their shapes, sizes, and levels of interior homogeneity. We model a region as a connected set of pixels that is surrounded by ramp edge discontinuities where the magnitude of these discontinuities is large compared to the variation inside the region. Each region is associated with a scale that depends upon the fraction of the strong and weak parts of its boundary. Traversing through the range of all possible scales, we obtain all regions present in the image. Regions strictly merge as the scale increases; hence a tree is formed where the root node corresponds to the whole image, nodes closer to the root along a path correspond to larger regions and those further from the root capture smaller regions representing embedded details. To evaluate the accuracy and precision of our algorithm, as well as to compare it to the existing algorithms, we present a new benchmark dataset for low-level image segmentation. We provide evaluation methods for both boundary-based and region-based performance of algorithms. We also annotate different parts of the images that are difficult to segment with the types of segmentation challenges they pose. This enables our benchmark to give an account of which algorithm fails where. We show that our proposed algorithm performs better than the widely used low-level segmentation algorithms on this benchmark.