Polygonal Contour Research Articles

A convolutional neural network (Seg-CNN) for recognition, classification, and segmentation of apple fruit branches and stems in images

This article reports the results of research studies conducted in 2023–2024 on transfer learning of Segmentation Convolutional Neural Networks (Seg-CNN) models for classification, recognition, and segmentation of branches with apple fruits and stems in images. State-of-the-art convolutional neural network architectures, i.e., YOLOv8(n,s,m,l,x)-seg, were used for a detailed segmentation of biological objects in images of varying complexity and scale at the pixel level. An image dataset collected in the field using a GoPro HERO 11 camera was marked up for transfer model training. Data augmentation was performed, producing a total of 2500 images. Image markup was performed using the polygon annotation tool. As a result, polygonal contours around objects were created, outlines of branches, apple tree fruits, and stems were outlined, and segments of objects in the images were indicated. The objects were assigned the following classes: Apple branch, Apple fruit, and Apple stem. Binary classification metrics, such as Precision and Recall, as well as Mean Average Precision (mAP), were used to evaluate the performance of the trained models in recognizing branches with apple fruits and stems in images. The YOLOv8x-seg (mAP50 0.758) and YOLOv8l-seg (mAP50 0.74) models showed high performance in terms of all metrics in recognizing branches, apple fruit, and fruit stems in images, outperforming the YOLOv8n-seg (mAP50 0.7) model due to their more complex architecture. The YOLOv8n-seg model has a faster frame processing speed (11.39 frames/s), rendering it a preferred choice for computing systems with limited resources. The results obtained confirm the prospects of using machine learning algorithms and convolutional neural networks for segmentation and pixel-by-pixel classification of branches with apple fruits and stems on RGB images for monitoring the condition of plants and determining their geometric characteristics.

From lines to Polygons: Polygonal building contour extraction from High-Resolution remote sensing imagery

Automated extraction of polygonal building contours from high-resolution remote sensing images is important for various applications. However, it remains a difficult task to achieve automated extraction of polygonal buildings at the level of human delineation due to diverse building structures and imperfect image conditions. In this paper, we propose Line2Poly, an end-to-end approach that uses feature lines as geometric primitives to achieve polygonal building extraction by recovering topological relationships among these lines within an individual building. To extract building feature lines with precision, we adopt a two-stage strategy that combines Convolutional Neural Network (CNN) and transformer architectures. A CNN-based module extracts preliminary feature lines, which serve as positional priors for initializing positional queries in the subsequent transformer-based module. For polygonal building contour reconstruction, we devise a learnable polygon topology reconstruction module that predicts adjacency relationships among discrete lines, and integrates lines into building polygons. The resultant building polygons, based on feature lines, exhibit inherent regularity that aligns with manual labeling standards. Extensive experiments on the Vectorizing World Buildings dataset, the WHU aerial building dataset and the WHU-Mix (vector) dataset validate Line2Poly’s impressive performance in building feature line extraction and instance-level building detection. Moreover, Line2Poly’s predictions exhibit the highest level of concurrence with manual delineations, with over 83% agreement on the WHU aerial building test set and 68.7/59.7% on the WHU-Mix (vector) test set I and II, respectively.

The Embodiment of Digital Art Elements in Traditional Cultural and Creative Product Designs

Abstract In order to extract the image path of traditional cultural and creative items, we set the binary image target of such products to be black and the backdrop to be white in this research. We followed the use of image vectorization processing technology. In order to address the issue of edge jaggedness in the traditional cultural and creative product image path extraction process, polygonal contour fitting is utilized to optimize the image paths of these products, resulting in the creation of a vector map of these products. The PCA point cloud alignment technique is used, based on the vector map of traditional cultural and creative products, to identify the attributes of these items and build a model of their traditional cultural and creative product design that incorporates digital art. After choosing dependent and independent variables, traditional cultural and creative product design, including digital art aspects, is studied and analyzed using linear quantile regression. The findings demonstrate that all of the value-added ratio’s coefficients are significant and bigger at the 0.2 and 0.95 quantile points, suggesting that raising the ratio can aid in the merger of digital art and more established creative and cultural sectors. The medium (M=3.586) and small standard deviation of the traditional cultural and creative items incorporating digital art suggest that the use function and cost-effectiveness of these products have a stronger influence on purchase motivation. This research encourages the fusion of digital art components with conventional cultural and artistic output in order to give traditional culture a fresh lease of life in the information era.

Application of bridge information modelling using laser scanning for static and dynamic analysis with concrete damage plasticity

A new proposal has been applied using Bridge Information Modeling associated to Terrestrial Laser Scanning TLS and photogrammetry to obtain the digital model of an existing bridge. For the laser scanning process. Thus, it was also possible to identify the pathologies and damages in the concrete of the bridge. Furthermore, it was shown how geo-information with remote sensing using the station Trimble SX 10 can be applied in area of bridge projects with less time and lower cost, in relationship to a conventional survey Asbuilt with manual survey in field. One of the innovations of this paper was to obtain the digital model of the bridge by integration of TLS and photogrammetry to obtain the 3D digital model. The new methodology consisted of acquisition of the bridge scenes through points previously implanted, making the polygonal contour of the entire bridge. Furthermore, the execution of the field survey from georeferenced points to capture the scenes during the sweep with TLS facilitated the refinement and, consequently, it was possible to obtain the digital model with the concrete damage. It was also possible to perform a complex nonlinear analysis with Finite Element with the digital model of the bridge. Therefore, it was applied a reverse analysis of the strength of the structure using the Concrete Damage Plasticity Model using ABAQUS software. Two numerical analyses were performed on the bridge; one considering only the static behavior of the bridge and another considering the dynamic behavior. In the static analysis, the behavior of the dapped-end beam of the bridge, was investigated. On the other hand, in the dynamic evaluation, the natural frequency and damping ratio for bridges with Dapped-end beams was obtained in ABAQUS through Rayleigh damping and compared with the technical literature, which presented satisfactory results.

A graph-based algorithm for slicing unstructured mesh files

• This paper describes an efficient algorithm for slicing files with unstructured mesh. • A novel graph-based algorithm was described and implemented for polygon assembly and contour generation. • Up to 10 times speedups were gained by using the new serial algorithm. • A multicore parallel algorithm was implemented to further speedup the slicing of files with a large number of triangles. Process planning is an important step in additive manufacturing processes that involves the creation of a deposition toolpath from input geometry. The most commonly used input geometry file formats are in the form of unstructured triangular meshes. Previous research works have presented different algorithms for slicing an unstructured triangular mesh by a series of uniformly spaced parallel planes. Recent research works show that the use of hashtable data structures results in an optimal algorithm that improves the efficiency of the slicing process. The presented algorithm minimizes the use of hashtable in the slicing process by using graph-based data structures. It was found that the slicing can be sped up by about 10 times by using the graph-based algorithm compared to the previous algorithm. The use of parallel processing on multi-core processors was explored to further speed up the slicing process. The proposed efficient algorithm can be useful in process planning optimization.

CHMM Object Detection Based on Polygon Contour Features by PSM.

Since the conventional split–merge algorithm is sensitive to the object scale variance and splitting starting point, a piecewise split–merge polygon-approximation method is proposed to extract the object contour features. Specifically, the contour corner is used as the starting point for the contour piecewise approximation to reduce the sensitivity of the contour segment for the starting point; then, the split–merge algorithm is used to implement the polygon approximation for each contour segment. Both the distance ratio and the arc length ratio instead of the distance error are used as the iterative stop condition to improve the robustness to the object scale variance. Both the angle and length as two features describe the shape of the contour polygon; they have a strong coupling relationship since they affect each other along the contour order relationship. To improve the description correction of the contour, these two features are combined to construct a Coupled Hidden Markov Model to detect the object by calculating the probability of the contour feature. The proposed algorithm is validated on ETHZ Shape Classes and INRIA Horses standard datasets. Compared with other contour-based object-detection algorithms, the proposed algorithm reduces the feature number and improves the object-detection rate.

Bending Stresses and Deformations in Prismatic Profiled Shafts with Noncircular Contours Based on Higher Hybrid Trochoids

This paper presents an analytical method for determining the bending stresses and deformations in prismatic, noncircular profile shafts with trochoidal cross sections. The so-called higher trochoids can be used as form-fit shaft-hub connections. Hybrid (mixed) higher trochoids (M-profiles) were developed for the special application as a profile contour for the form-fit shaft and hub connections in an earlier work by the author. M-profiles combine the advantages of the two standardised polygonal and spline contours, which are used as shaft-hub connections for the transmission of high torques. In this study, the geometric and mechanical properties of the higher hybrid trochoids were investigated using complex functions to simplify the calculations. The pure bending stress and shaft deflection were determined for M-profiles using bending theory based on the theory of mathematical elasticity. The loading cases consisted of static and rotating bends. Analytical, numerical, and experimental results agreed well. The calculation formulas developed in this work enable reliable and low-cost dimensioning with regard to the stresses and elastic deformations of profile shafts subjected to bending loads.

DrawnNet: Offline Hand-Drawn Diagram Recognition Based on Keypoint Prediction of Aggregating Geometric Characteristics.

Offline hand-drawn diagram recognition is concerned with digitizing diagrams sketched on paper or whiteboard to enable further editing. Some existing models can identify the individual objects like arrows and symbols, but they become involved in the dilemma of being unable to understand a diagram’s structure. Such a shortage may be inconvenient to digitalization or reconstruction of a diagram from its hand-drawn version. Other methods can accomplish this goal, but they live on stroke temporary information and time-consuming post-processing, which somehow hinders the practicability of these methods. Recently, Convolutional Neural Networks (CNN) have been proved that they perform the state-of-the-art across many visual tasks. In this paper, we propose DrawnNet, a unified CNN-based keypoint-based detector, for recognizing individual symbols and understanding the structure of offline hand-drawn diagrams. DrawnNet is designed upon CornerNet with extensions of two novel keypoint pooling modules which serve to extract and aggregate geometric characteristics existing in polygonal contours such as rectangle, square, and diamond within hand-drawn diagrams, and an arrow orientation prediction branch which aims to predict which direction an arrow points to through predicting arrow keypoints. We conducted wide experiments on public diagram benchmarks to evaluate our proposed method. Results show that DrawnNet achieves , , and recognition rate improvements compared with the state-of-the-art methods across benchmarks of FC-A, FC-B, and FA, respectively, outperforming existing diagram recognition systems on each metric. Ablation study reveals that our proposed method can effectively enable hand-drawn diagram recognition.

Generation Algorithm of a Novel Platform Attached Support Structure for FDM-Fused Deposition Modeling

In the process of 3D printing, the setting of the initial layer is the most important step to achieve perfect printing results, and its main function is to prevent warping in 3D printing. There are three main types of the first layer: Raft, Brim, and Skirt. The Brim is a special skirt and is usually the preferred starting layer for the printing process. In order to increase the adhesion, multiple Brim contour edges will be printed, which increases the difficulty of removal. Otherwise, poor adhesion effect will occur. By analyzing the distinction between inner and outer contours of polygons, the offset process of contour points in different cases is calculated in this paper, and a simplified processing method of contour offset line is designed by using the idea of Douglas–Peucker simplifying curve feature points. According to the functional characteristics of the Brim edge structure, a novel platform attached support structure is generated based on ray-casting algorithm and contour offset algorithm. The proposed algorithm is encapsulated into the slicing engine and compared with three common initial layer structures for conducting part printing comparison experiments. The experimental results show that the edge structure realized by the proposed generation optimization algorithm of a novel platform attached support can effectively reduce the slicing execution time, and the auxiliary structure has little impact on the forming quality. It can be easier to remove on the premise and the warpage of the printing model is small.

Building Classification Algorithm Based on Improved Hough Transform and Geographic Information System

Although determining the type of building can be useful for 3D reconstruction, image retrieval, and other applications, it is often difficult to do so automatically. A building classification algorithm using a geographic information system (GIS) is proposed in this paper. The typical adjustment of building polygon, local convex-concave simplification of large-area polygon contour, and local exaggeration of competing polygon groups merging in small-area polygon is the main topics of this paper. The preliminary outline area is obtained by analyzing the corner characteristics of two basic types of buildings in urban buildings using a large-scale morphological screen. The classification statistics of all kinds of corners are carried out using the improved Hough transform and the proposed line segment and corner optimization algorithm, and the automatic classification of flat-roofed and nonflat-roofed buildings photographed by digital cameras is realized. The experimental results show that the algorithm simplifies the polygons in a reasonable way while maintaining the block shape. Furthermore, the optimization algorithm proposed in this paper effectively eliminates the influence of false contours, allowing for high-accuracy building type judgment.

Perturbative Approach for the Analysis of Charge Distribution on Arbitrarily Shaped Conductors

In this paper, we analyze the electrostatic charge distribution on arbitrarily shaped conductor surfaces. Following a perturbative approach, we derive an approximate analytical formulation of the problem. We start from the known case of a conducting ellipsoid, we adopt a deformed ellipsoidal coordinate system, and we search for the zero-order approximated solution of the problem. We also focus on arbitrary-shaped thin foils, showing that the charge density is divergent on their borders. We then define the applicability range of the proposed approach expressing the contour equation as the Fourier series. Finally, we present a detailed error analysis for several polygonal contours, comparing the analytical results with those obtained via a numerical analysis based on the Finite Element Methods (FEM).

Large-size expansion for triangular Wilson loops in confining gauge theories

The asymptotic behavior of Wilson loops in the large-size limit (L → ∞) in confining gauge theories with area law is controlled by effective string theory (EST). The L−2 term of the large-size expansion for the logarithm of Wilson loop appears within EST as a 2-loop correction. Ultraviolet divergences of this 2-loop correction for polygonal contours can be renormalized using an analytical regularization constructed in terms of Schwarz-Christoffel mapping. In the case of triangular Wilson loops this method leads to a simple final expression for the L−2 term.

Electrostatic field of angular-dependent surface electrodes

We present an analytic strategy to find the electric field generated by surface electrode SE with angular dependent potential. This system is a planar region $\mathcal{A}$ kept at a fixed but non-uniform electric potential $V(\phi)$ with an arbitrary angular dependence. We show that the generated electric field is due to the contribution of two fields: one that depends on the circulation on the contour of the planar region ---in a Biot-Savart-Like (BSL) term---, and another one that accounts for the angular variations of the potential in $\mathcal{A}$. This approach can be used to find exact solutions of the BSL electric field for circular or polygonal contours of the planar region with periodic distributions of the electric potential. Analytic results are validated with numerical computations and the Finite Element Method. Keywords: Surface-electrode, Biot-Savart law, electrostatic problems, exactly solvable models.

Prognostic of the Using Wood of Erythroxylum myrsinites Mart. Under the Anatomic Prism

The composition of the wood and its elements are considered determinants in the wood properties, standing out the fibers, axial parenchyma, number of vessels and rays. In this context, we aimed to investigate and describe the wood anatomical structure of the Erythroxylum myrsinites Mart. and identify its potential uses in the area of wood technology. From the tree, were removed three wood samples for the confection of anatomical slides, witch 3 × 3 × 3 cm, oriented we plans transversal, longitudinal and tangential longitudinal. For the preparation of histological blades, the method of Burger and Richter was adopted; the anatomical description was based on IAWA Committee recommendations. The species presents vessels numerous, solitary, composing 19% of the volume of wood. Diffuse porosity of polygonal contour and thin walls. Mean vascular elements, axial parenchyma representing 3% of the volume of the wood, in a paratracheal vasicentric arrangement. The rays are numerous, occupying 23% of the volume of the wood, narrow and heterogeneous, consisting of procumbent, square and erect cells. The fibers are libriform, of length medium and thick walls, occupying 55% of the volume of the wood, have spiral thickening and septate fibers, absent tracheids, and gelatinous fibers abundant. The characteristics presented give high flexibility to the stem and branches, and the wood with low specific mass, permeable and resistant to mechanical loads has potential for use in energy generation.

ABJM flux-tube and scattering amplitudes

There is a number of indications that scattering amplitudes in the Aharony-Bergman-Jafferis-Maldacena theory might have a dual description in terms of a holonomy of a supergauge connection on a null polygonal contour in a way analogous to the four-dimensional maximally supersymmetric Yang-Mills theory. However, so far its explicit implementations evaded a successful completion. The difficulty is intimately tied to the lack of the T-self-duality of the sigma model on the string side of the gauge/string correspondence. Unscathed by the last misfortune, we initiate with this study an application of the pentagon paradigm to scattering amplitudes of the theory. With the language being democratic and nondiscriminatory to whether one considers a Wilson loop expectation value or an amplitude, the success in the application of the program points towards a possible unified observable on the field theory side. Our present consideration is focused on two-loop perturbation theory in the planar limit, begging for higher loop data in order to bootstrap current analysis to all orders in the ’t Hooft coupling.

On Direct and Inverse Problems of Logarithmic Potential With Finite Number of Parameters

We describe possibilities of realizing V. K. Ivanov’s method for solving inverse problems of logarithmic potential and that of A. V. Tsirul’skii’s method for solving direct problems of logarithmic potential. We solve two direct problems for logarithmic potential and for simple layer potential in the case of polygonal contour.

A variant of the Hough Transform for the combined detection of corners, segments, and polylines

The Hough Transform (HT) is an effective and popular technique for detecting image features such as lines and curves. From its standard form, numerous variants have emerged with the objective, in many cases, of extending the kind of image features that could be detected. Particularly, corner and line segment detection using HT has been separately addressed by several approaches. To deal with the combined detection of both image features (corners and segments), this paper presents a new variant of the Hough Transform. The proposed method provides an accurate detection of segment endpoints, even if they do not correspond to intersection points between line segments. Segments are detected from their endpoints, producing not only a set of isolated segments but also a collection of polylines. This provides a direct representation of the polygonal contours of the image despite imperfections in the input data such as missing or noisy feature points. It is also shown how this proposal can be extended to detect predefined polygonal shapes. The paper describes in detail every stage of the proposed method and includes experimental results obtained from real images showing the benefits of the proposal in comparison with other approaches.

Image-Based Slicing and Tool Path Planning for Hybrid Stereolithography Additive Manufacturing

The hybrid stereolithography (SLA) process integrates a laser scanning-based system and a mask projection-based system. Multiple laser paths are used to scan the border of a 2D pattern, whereas a mask image is adopted to solidify the interior area. By integrating merits of two subsystems, the hybrid SLA process can achieve high surface quality without sacrificing productivity. For the hybrid system, closed polygonal contours are required to direct the laser scanning, and a binary image is also needed for the mask projection. We proposed a novel image-based slicing method. This approach can convert a 3D model into a series of binary images directly, and each image is corresponding to the cross section of the model at a specific height. Based on the resultant binary image, we use an image processing method to gradually shrink the pattern in the image. Boundaries of the shrunk image are traced and then restored as polygons to direct the laser spot movement. The final shrunk image serves as the input for the mask projection. Experimental results of test cases demonstrate that the proposed method is substantially more efficient than the traditional approaches. Its accuracy is also studied and discussed.

Contour analysis of a fine structure in an electroencephalogram

An electroencephalogram (EEG) is examined as an image limited by a polygonal contour. A sequence of elementary vectors presented in analytical form, which approximates the contour line, is accepted as an electroencephalogram model. A concept on a fine structure in the electroencephalogram as a set of elementary vectors that defines the segmented parts obtained as a result of electroencephalogram decomposition is introduced. The expressions for a measure of similarity of two segments of the fine structure in a form of magnitude of their normalized scalar product are obtained. An example on detecting and analyzing the "Sleep Spindle" pattern, which is observed in the frequency range of 9---13 Hz, by using the contour model, is presented.

The Perceived Size and Shape of Objects in Peripheral Vision

Little is known about how we perceive the size and shape of objects in far peripheral vision. Observations made during an artistic study of visual space suggest that objects appear smaller and compressed in the periphery compared with central vision. To test this, we conducted three experiments. In Experiment 1, we asked participants to draw how a set of peripheral discs appeared when viewed peripherally without time or eye movement constraints. In Experiment 2, we used the method of constant stimuli to measure when a briefly presented peripheral stimulus appeared bigger or smaller compared with a central fixated one. In Experiment 3, we measured how accurate participants were in discriminating shapes presented briefly in the periphery. In Experiment 1, the peripheral discs were reported as appearing significantly smaller than the central disc, and as having an elliptical or polygonal contour. In Experiment 2, participants judged the size of peripheral discs as being significantly smaller when compared with the central disc across most of the peripheral field, and in Experiment 3, participants were quite accurate in reporting the shape of the peripheral object, except in the far periphery. Our results show that objects in the visual periphery are perceived as diminished in size when presented for long and brief exposures, suggesting diminution is an intrinsic feature of the structure of the visual space. Shape distortions, however, are reported only with longer exposures.