Spatial Structure Relationship Research Articles

Tensor completion algorithm-aided structural color design.

In recent years, structural color has developed rapidly due to its distinct advantages, such as low loss, high spatial resolution and environmental friendliness. Various inverse design methods have been extensively investigated to efficiently design optical structures. However, the optimization method for the inverse design of structural color remains a formidable challenge. Traditional optimization approaches, such as genetic algorithms require time-consuming repetitions of structural simulations. Deep learning-assisted design necessitates prior simulations and large amounts of data, making it less efficient for systems with a small number of features. This study proposes a tensor completion algorithm capable of swiftly and accurately predicting missing datasets based on partially obtained datasets to assist in structural color design. Transforming the complex physical problem of structural color design into a spatial structure relationship problem linking geometric parameters and spectral data. The method utilizes tensor multilinear data analysis to effectively capture the complex relationships associated with geometric parameters and spectral data in higher-order data. Numerical and experimental results demonstrate that the algorithm exhibits high reliability in terms of speed and accuracy for diverse structures, datasets of varying sizes, and different materials, significantly enhancing design efficiency. The proposed algorithm offers a viable solution for inverse design problems involving complex physical systems, thereby introducing a novel approach to the design of photonic devices. Additionally, numerical experiments illustrate that the structural color of cruciform resonators with diamond can overcome the high loss issues observed in traditional dielectric materials within the blue wavelength region and enhance the corrosion resistance of the structure. We achieve a wide color gamut and a high-narrow reflection spectrum nearing 1 by this structure, and the theoretical analysis further verifies that diamond holds great promise in the realm of optics.

Structure, mechanical properties and tribocorrosion behaviours of superhard TiSiCN nanocomposite coatings

Superhard coatings are no longer very difficult to obtain, but the direct observation and prediction of their failure are still a challenge. 3D tomography of the microcracks in coating failure offers lots of new information on the geometric shape, size, and spatial structure relationship of microcracks which was not available up to now. Hence, in this work, cross-scale visualization analysis of the microcrack network in superhard yet toughness TiSiCN coatings around indentations was achieved through advanced 3D tomography combined with HRTEM characterization. TiSiCN nanocomposite coatings were deposited by high-power pulsed magnetron sputtering. Superhard TiSiCN coatings exhibit a simultaneous superhardness of 48 ± 6.99 GPa and favorable toughness due to fully dense microstructure and smooth surface morphology. Due to the excellent crack initiation resistance and super-high hardness, superhard TiSiCN coatings show enhanced tribocorrosion performance in 3.5 wt% NaCl aqueous solution, with a low coefficient of friction of 0.11 and low specific wear rates of 3.23 × 10−6 mm−3N−1m−1.

SSR-Net: A Spatial Structural Relation Network for Vehicle Re-identification

Vehicle re-identification (Re-ID) represents the task aiming to identify the same vehicle from images captured by different cameras. Recent years have seen various feature learning-based approaches merely focusing on feature representations including global features or local features to obtain more subtle details to identify highly similar vehicles. However, few such methods consider the spatial geometrical structure relationship among local regions or between the global and local regions. By contrast, in this study, we propose a Spatial Structural Relation Network (SSR-Net) that explores the above-mentioned two kinds of relations simultaneously to learn more discriminative features by modeling the spatial structure information and global context information. In this article, we propose to adopt a Graph Convolution Network (GCN), for modeling spatial structural relationships among characteristic features. The GCN model aggregating the local and global features is shown to be more discriminative and robust to several car image transformations. To improve the performance of our proposed network, we jointly combine the classification loss with metric learning loss. Extensive experiments conducted on the public VehicleID and VeRi-776 datasets validate the effectiveness of our approach in comparison with recent works.

Three-Dimensional Human Pose Estimation with Spatial–Temporal Interaction Enhancement Transformer

Three-dimensional human pose estimation is a hot research topic in the field of computer vision. In recent years, significant progress has been made in estimating 3D human pose from monocular video, but there is still much room for improvement in this task owing to the issues of self-occlusion and depth ambiguity. Some previous work has addressed the above problems by investigating spatio-temporal relationships and has made great progress. Based on this, we further explored the spatio-temporal relationship and propose a new method, called STFormer. Our whole framework consists of two main stages: (1) extract features independently from the temporal and spatial domains; (2) modeling the communication of information across domains. The temporal dependencies were injected into the spatial domain to dynamically modify the spatial structure relationships between joints. Then, the results were used to refine the temporal features. After the preceding steps, both spatial and temporal features were strengthened, and the estimated final pose will be more precise. We conducted substantial experiments on a well-known dataset (Human3.6), and the results indicated that STFormer outperformed recent methods with an input of nine frames. Compared to PoseFormer, the performance of our method reduced the MPJPE by 2.1%. Furthermore, we performed numerous ablation studies to analyze and prove the validity of the various constituent modules of STFormer.

Three-dimensional Visualization in Diagnostic Imaging: A Renaissance in Radiology.

Supplemental material is available for this article.

LESSL: Can LEGO sampling and collaborative optimization contribute to self-supervised learning?

Self-supervised visual representation learning (SSL) aims to extract the most distinctive features from unlabeled datasets to overcome challenges of labor-intensive and time-consuming data annotation. Recent advances in this area are dominated by contrastive learning-based methods with various sampling strategies and optimization objectives. Unfortunately, these methods are limited by larger batch sizes and longer training epochs, resulting in non-negligible computational resource consumption and memory footprint. Unlike prior sampling manners and inspired by LEGO bricks, we design a comprehensive interleave sampling module to fully exploit unlabeled datasets, which seamlessly integrates the advantages of semantic and spatial relations complementarity from the unlabeled image via optimizing two objectives. Specifically, the unlabeled images are sampled in a dense and interleaved manner, which breaks down barriers of under-utilization of datasets existing in state-of-the-art methods. Meanwhile, the redefined collaborative optimization term is designed to alleviate feature-specificity and implicitly explore sample relationships during training, which effectively improves the performance of downstream tasks on various datasets. Besides, we focus on high-level semantic features and utilize the spatial structure relationships provided by the unlabeled datasets to ensure the learned features with low-level texture characteristics. Extensive experiments on four types of datasets demonstrate that our method performs favorably against the state-of-the-art SSL methods.

Deep transfer learning with metric structure for fault diagnosis

Maximum Mean Discrepancy (MMD) is a classical method of measuring distribution distance and is widely used in Deep Transfer Learning (DTL) for fault diagnosis. However, in practice, MMD neglects the label information and the spatial structure relationship of the data and performs unsatisfied. To address this issue, a novel DTL fault diagnosis method with metric structure is proposed in this paper. Besides MMD, a novel domain adaptation module is designed through learning a common metric matrix from the Two Transferred Domains (TTD). Thereby, it can constrain TTD in the same geometric structure and make the distribution of TTD similar by both the mean discrepancy and the metric structure. Furthermore, a novel soft pseudo-label trick is proposed for dividing similar and dissimilar sample pairs of the unlabeled samples in the target domain. Experimental results have demonstrated that this method has well diagnostic performance in multiple transfer tasks.

Exploring the Visual Space Structure of Oil Painting Based on Visual Importance.

For present and future design, research on visual-spatial language and the mastery of its fundamental concepts is extremely valuable. Since a more comprehensive research system has not yet been established to investigate the aesthetics of visual form and its significance, the research fields of spatial color and spatial structure relationships are currently relatively unique. The visual forms of design will be fundamentally affected by the diverse development of the design thinking ideology, and the expansion of design language will become a given. Design thinking and methods are significantly impacted by the emphasis on digital technology in online media. Design is no longer restricted to the arrangement of two-dimensional planes thanks to the visual space of oil paintings; three-dimensional space and time organization structures are combined with visual elements, and content layout evolves from static layout to dynamic interactive layout. In order to better understand visual space and intuition, this study begins by examining the structure of visual space in oil paintings. It does this by using artistic psychology and spatial space as research hints. The experimental results demonstrate that the depth-aware image quality evaluation algorithm in this study consumes more time than the MSEPF + 30 algorithm and the MSEPF + 50 algorithm by 12 ms and 15 ms, respectively, while the time consumption of individual particles is increased by 0.08 ms and 0.04 ms, respectively. This algorithm can thus meet the real-time requirements. As a result, it is anticipated that this study will address more comprehensive and detailed data as well as theoretical underpinnings that will be helpful in the study of painting art.

A Graph Attention Mechanism-Based Multiagent Reinforcement-Learning Method for Task Scheduling in Edge Computing

Multi-access edge computing (MEC) enables end devices with limited computing power to provide effective solutions while dealing with tasks that are computationally challenging. When each end device in an MEC scenario generates multiple tasks, how to reasonably and effectively schedule these tasks is a large-scale discrete action space problem. In addition, how to exploit the objectively existing spatial structure relationships in the given scenario is also an important factor to be considered in task-scheduling algorithms. In this work, we consider indivisible, time-sensitive tasks under this scenario and formalize the task-scheduling problem to minimize the long-term losses. We propose a multiagent collaborative deep reinforcement learning (DRL)-based distributed scheduling algorithm based on graph attention neural networks (GATs) to solve task-scheduling problems in the MEC scenario. Each end device creates a graph representation agent to extract potential spatial features in the scenario and a scheduling agent to extract the timing-related features of the tasks and make scheduling decisions using a gated recurrent unit (GRU). The simulation results show that, compared with several baseline algorithms, our proposed algorithm can take advantage of the spatial positional relationship of devices in the environment, significantly reduce the average delay and drop rate, and improve link utilization.

Cross-sectional representations of the central nervous system in Pirogov’s “Ice Anatomy”

ABSTRACT Russian surgeon Nikolay Ivanovich Pirogov (Pirogoff; 1810–1881) introduced the teaching of applied topographical anatomy in Russia. Pirogov’s monumental four-part atlas, Anatome topographica sectionibus per corporis humanum congelatum triplici directione ductis illustrate (An Illustrated Topographic Anatomy of Saw Cuts Made in Three Dimensions Across the Frozen Human Body), colloquially known as the “Ice Anatomy,” was published in Latin in folio in the 1850s. Pirogov sought to investigate “the normal and pathological positions of different organs and body parts using sections made in the three principal directions [anatomical planes] … throughout all regions.” To accomplish this, he froze cadavers “to the density of the thickest wood” and then cut them into thin plates with a special mechanical saw. His approach was reportedly inspired by his observations of butchers sawing across frozen pig carcasses at the meat market in St. Petersburg during winter. Pirogov systemically obtained full-size representations of more than 1,000 sections. A painter made a representative copy of the cross-sectional contours of each section, using ruled glass overlain on the sections. The final lithographs were of high artistic quality and execution, resembling modern high-resolution medical imaging (i.e., CT or MRI). Moreover, structures were serially sectioned and systematically illustrated along all three anatomical planes, something that had never previously been attempted. This allowed clinicians and anatomists to scrutinize the spatial relationships of structures from multiple perspectives and at a much more detailed level than was previously possible, although the cost, massiveness, and complexity of the completed work precluded wide dissemination.

Precision analysis of series-parallel machine tool based on hierarchy motion function units

This paper takes the motion units of the series-parallel machine tool (SPMT) as the basic research objects. Combined with the meta-action theory, the FMA-MBS structure model is proposed to describe the spatial structure relationships and motion transfer relationships of SPMT. Based on the simplified model, the kinematics characteristics of the machine tool are gradually researched from four layers: joint layer, limb layer, subsystem layer, and the whole machine. According to the twist system of limbs and moving platform, the parallel motion system can be simplified into a virtual motion chain with specific constraints. Then, the kinematics model of SPMT is constructed in twist exponential form. By introducing two virtual degrees of freedom (DOFs), the unified expressions of active and passive joint errors are obtained. All motion systems error models are derived based on three basic error terms of the meta-action layer. Based on the error mapping relationships, the precision of the machine tool can be calculated and evaluated. In order to illustrate the applicability and effectiveness of the proposed method, a sample library of common SPMTs and a case study of a 5-axis series-parallel machining center are provided.

Research Techniques Made Simple: Volume Scanning Electron Microscopy

Volume scanning electron microscopy (VSEM) involves the serial sectioning and imaging of a sample using scanning electron microscopy (SEM), followed by segmentation and three-dimensional (3D) reconstruction using computer software packages to allow visualization of 3D structures. VSEM can reveal qualitative and quantitative properties of organelles and cells within tissues at nanoscale. The ability to visualize spatial relationships of structures of interest within and across cells in 3D space in particular sets VSEM apart from conventional SEM and transmission electron microscopy. Here, we provide an overview of VSEM platforms and image processing, highlighting characteristics that will aid selection of a method to address specific research questions in dermatological research.

Initial experiences with preoperative three-dimensional image reconstruction technology in laparoscopic pyeloplasty for ureteropelvic junction obstruction.

BackgroundTo explore the clinical value of three-dimensional image reconstruction technology (3DIT) on preoperative surgical planning and perioperative outcomes in laparoscopic pyeloplasty (LP).MethodsData of 25 patients with ureteropelvic junction obstruction (UPJO) admitted to our hospital from January 2018 to January 2019 was analyzed retrospectively. All patients underwent preoperative enhanced computed tomography (CT) scanning. In the 12 cases in the 3DIT group, preoperative planning involved the use of virtual operation and morphometry based on reconstruction of the CT data into three-dimensional (3D) images. Surgery in the other 13 cases was performed with traditional CT examination. Demographic, surgical outcome, and postoperative parameters were compared between these two groups.ResultsReconstructed 3D images clearly showed the spatial structural relationships between the UPJO and surrounding blood vessels. In all 25 cases surgery was completed with no conversion to open surgery. Preoperative 3DIT analyses resulted in significant improvements to mean operation time (107.76 vs. 141.58 min, P=0.024), mean time of dissociating ureteropelvic junction (UPJ) (11.26 vs. 19.40 min, P=0.020), and mean estimated blood loss volume (23.84 vs. 49.16 mL, P=0.028). There were no statistically significant differences in perioperative complications, postoperative hospital stays or postoperative drainage time.Conclusions3DIT based on enhanced CT scans is of clinical value in the treatment of UPJO, as it can provide accurate anatomical information and reliable guidance for preoperative operation planning, and it facilitates image-guided LP.

Crashing from cadaver to computer: Covid-driven crisis-mode pedagogy spawns active online substitute for teaching gross anatomy.

In early 2020, the Covid-19 crisis forced medical institutions worldwide to convert quickly to online platforms for content delivery. Although many components of medical education were adaptable to that format, anatomical dissection laboratory lost substantial content in that conversion, including features of active student participation, three-dimensional spatial relationships of structures, and the perception of texture, variation, and scale. The present study aimed to develop and assess online anatomy laboratory sessions that sought to preserve benefits of the dissection experience for first-year medical students. The online teaching package was based on a novel form of active videography that emulates eye movement patterns that occur during processes of visual identification, scene analysis, and learning. Using this video-image library of dissected materials, content was presented through asynchronous narrated laboratory demonstrations and synchronous/active video conference sessions and included a novel, video-based assessment tool. Data were obtained using summative assessments and a final course evaluation. Test scores for the online practical examination were significantly improved over those for previous in-person dissection-based examinations, as evidenced by several measures of performance (Mean: 2015-2019: 82.5%; 2020: 94.9%; P=0.003). Concurrently, didactic test scores were slightly, but not significantly, improved (Mean: 2015-2019: 88.0%; 2020: 89.9%). Student evaluations of online sessions and overall course were highly positive. Results indicated that this innovative online teaching package can provide an effective alternative when in-person dissection laboratory is unavailable. Although this approach consumed considerable faculty time for video editing, further development will include video conference breakout rooms to emulate dissection small-group teamwork.

Hyperspectral Image Unmixing Method Based on Multiple Kernel Graph Non-negative Matrix Factorization

In order to improve the accuracy of non-negative matrix factorization (NMF) in hyperspectral images unmixing, this paper proposes a hyperspectral image unmixing method based on multiple kernel graph NMF. Using multiple kernel learning (MKL), the hyperspectral data is mapped to the high-dimensional feature space, so that the linearly inseparable hyperspectral data in the low-dimensional space is linearly separable. Introducing the graph regular terms that characterizes the spatial structure relationship between pixels, the internal manifold structure of the hyperspectral image is effectively expressed. The results of both simulated dataset and real dataset show that compared with other method, the proposed method can improve the unmixing accuracy.

Creating a 3D Printed Model of the Female Perineum to Supplement Traditional Cadaveric Education

Cadaveric materials are the cornerstone of anatomy education as they provide students the opportunity to understand the complex spatial relationships of structures within their native environment. Despite their widespread use, cadaveric dissection in the hands of inexperienced medical students rarely provides adequate visualization of delicate anatomy. An excellent example of this is the female perineum, where the structures to be studied are relatively small and often not well preserved. Continued pressure to condense anatomy courses may result in the elimination of “low‐yield” dissections such as the perineum from the anatomy curriculum in favor of laboratory sessions involving easier to dissect or more prominent regions. Additionally, the sensitive nature of the area may pose personal, ethical, and cultural concerns for some students. With these concerns in mind, our aim was to create a supplementary model to assist in teaching the anatomy of the female perineum. 3D‐modeling and printing technology was chosen as it has a short learning curve for novice users, and it allows for cost effective design, production, and dissemination of models. Standard methods of auto‐creating virtual 3D‐models from CT or MRI utilizing medical processing software were not feasible given that the structures of the perineum and pelvic diaphragm are not well visualized by standard imaging techniques. Therefore, referencing anatomical atlases and cadaveric prosections, we created a virtual model of the female perineum, pelvis and internal organs. While a majority of the structures in this model were created with Fusion 360 (Autodesk), we did incorporate structures from the opensource 3D‐print library Thinigverse, demonstrating the ease at which users can collaborate using 3D‐printing technology. This model was then 3D‐printed on a Makerbot2x Replicator using 1.5mm ABS filament. Currently, the model is being employed in small‐group self‐directed laboratory stations, supplementing prosected specimens. Here, we demonstrate that is it possible to use 3D‐printing technology to create a detailed anatomic model when cadaveric material and MRI is inadequate. The implications of this are vital for the future of anatomical education. While specialized training is always beneficial, online tutorials, active support communities and freely available software make it possible for anyone to create 3D‐printed models. Additionally, open access libraries allow users to collaborate on models and share designs. The low cost of producing 3D‐printed models enables faculty to provide a larger number and variety of models, not to replace cadaveric materials but to act as supplements for regions or concepts that are difficult to visualize or learn. Models can also be modified to represent specific pathology, or anatomical variations which are unlikely to be observed in cadavers. Finally, regions that are difficult to understand can be enlarged or color‐coded to facilitate learning. Future studies will validate the effectiveness of the model as a supplemental teaching tool.Support or Funding InformationMassachusetts Medical Society Information Technology in Medicine Award

Plastination-A scientific method for teaching and research.

Over the last four decades, plastination has been one of the best processes of preservation for organic tissue. In this process, water and lipids in biological tissues are replaced by polymers (silicone, epoxy, polyester) which are hardened, resulting in dry, odourless and durable specimens. Nowadays, after more than 40years of its development, plastination is applied in more than 400 departments of anatomy, pathology, forensic sciences and biology all over the world. The most known polymers used in plastination are silicone (S10), epoxy (E12) and polyester (P40). The key element in plastination is the impregnation stage, and therefore depending on the polymer that is used, the optical quality of specimens differs. The S10 silicone technique is the most common technique used in plastination. Specimens can be used, especially in teaching, as they are easy to handle and display a realistic topography. Plastinated silicone specimens are used for displaying whole bodies, or body parts for exhibition. Transparent tissue sections, with a thickness between 1 and 4mm, are usually produced by using epoxy (E12) or polyester (P40) polymer. These sections can be used to study both macroscopic and microscopic structures. Compared with the usual methods of dissection or corrosion, plastinated slices have the advantage of not destroying or altering the spatial relationships of structures. Plastination can be used as a teaching and research tool. Besides the teaching and scientific sector, plastination becomes a common resource for exhibitions, as worldwide more and more exhibitions use plastinated specimens.

Sea Ice Classification with First-Order Logic Refined Sliding Bagging

Ren, P.; Yu, Z.-Q.; Dong, G.S.; Wang, G.-X., and, Wei, K., 2019. Sea ice classification with first-order logic refined sliding bagging. In: Jung, H.-S.; Lee, S.; Ryu, J.-H., and Cui, T. (eds.), Advances in Remote Sensing and Geoscience Information Systems of Coastal Environments. Journal of Coastal Research, Special Issue No. 90, pp. 129-134. Coconut Creek (Florida), ISSN 0749-0208.This paper proposes an automatic framework for classifying sea ice types using remote sensing imageries. Firstly, polarization features are extracted to form polarization characteristic vectors for representing individual pixels in a remote sensing image. Secondly, a multiple classifier ensemble strategy, i.e., sliding bagging, is exploited for classifying each polarization characteristic vector into a sea ice type. The sliding bagging strategy not only avoids the limitation of one single classifier for characterizing the sea ice variability over a large-scale region, but also alleviates the data imbalance over different sea ice types. Finally, the spatial structural relationships of sea ice types are encoded with first-order logic (FOL) to refine the sea ice classification resulted from the sliding bagging scheme. Experimental evaluations on RADARSAT-2 data validate the effectiveness of the proposed framework.

Matching Real-World Facilities to Building Information Modeling Data Using Natural Language Processing

Building Information Modeling (BIM) is a promising technology for building informatics. Currently, an increasing number of applications adopt BIM to improve the building operations and facility management. In these applications, matching real-world facilities to the corresponding BIM items is a fundamental yet challenging task. This study addresses this issue using Natural Language Processing. Firstly, a novel BIM hierarchy tree (HiTree) is proposed to model the original spatial structure relationships of a BIM. Then, the locations of facilities are extracted from natural language through processes of word segmentation, keyword extraction, and semantic disambiguation. Thirdly, an algorithm that matches real-world facilities to the BIM data is developed using the HiTree and the extracted locations. Finally, a concrete case for a 35,000 m 2 library is presented to verify the effectiveness of the proposed solution. BIM has become a common paradigm in the construction industry, and our scheme can facilitate more applications of BIM in building operations and facility management. One of the most representative applications is integrating the BIM data and information within IoT (Internet of Things) system intelligently by matching the BIM data to real-world facilities.

Weak-structure-aware visual object tracking with bottom-up and top-down context exploration

It poses great challenges to model-free trackers that the object undergoes large appearance variations due to motion, shape deformation, occlusion and surrounding environments. In this paper, we investigate a novel method for modeling and locating the object by being aware of the weak structures of discriminative parts of both the object and its surroundings. The discriminative parts are modeled based on keypoints and feature descriptors. We separate the discriminative parts into two sets corresponding to object and background, and model their spatial structure relationship with the object. While tracking, the successfully localized parts will contribute to potential centers of the object. Aware of the weak structures, we further cluster potential centers to locate the object. The object scale is also updated adaptively. To increase the accuracy of this weak-structure-aware location inference, we fully explore context in both bottom-up and top-down procedures. In the bottom-up stage, we explore the local motion estimation of low-level pixels. The bottom-up information produces consistent tracking of discriminative parts. In the top-down stage, we build a superpixel kernel model to roughly distinguish the object from its surroundings, which provides guided information for location inference and model update. The effectiveness of the proposed method is verified by evaluation on a popular benchmark and comparison with recent tracking methods.