PCEDNet: A Lightweight Neural Network for Fast and Interactive Edge Detection in 3D Point Clouds

In recent years, Convolutional Neural Networks (CNN) have proven to be efficient analysis tools for processing point clouds, e.g., for reconstruction, segmentation, and classification. In this article, we focus on the classification of edges in point clouds, where both edges and their surrounding are described. We propose a new parameterization adding to each point a set of differential information on its surrounding shape reconstructed at different scales. These parameters, stored in a Scale-Space Matrix (SSM) , provide a well-suited information from which an adequate neural network can learn the description of edges and use it to efficiently detect them in acquired point clouds. After successfully applying a multi-scale CNN on SSMs for the efficient classification of edges and their neighborhood, we propose a new lightweight neural network architecture outperforming the CNN in learning time, processing time, and classification capabilities. Our architecture is compact, requires small learning sets, is very fast to train, and classifies millions of points in seconds.

Technical aspects of creating radiotechnical systems of information and command interface to ensure the functioning of complexes with unmanned aerial vehicles

One of the key components of the complexes with unmanned aerial vehicles (CUAV), which significantly affects the safety and reliability of operation, is the radio link for transmitting command-telemetric and target information of direct radio visibility (radio engineering system of information and command interface) between unmanned aerial vehicles (UAV) and the ground control and information processing point (GCSP). Heavy-class unmanned aerial vehicles can have several target loads simultaneously – a multispectral optical-electronic system, a digital aerial photo system, a radar system, a radio monitoring system, etc. Due to the fact that the flow of information with the payloads to the input of the radio link during peak time moments can be quite large, severe requirements are imposed as to the equipment of primary processing of the target information aboard the UAV and it’s compression in the recording information, and the bandwidth of the radio link in terms of transmission target information on GCSP for subsequent analysis and processing by the operator of the complex. Technical articles are the justification of modern approaches to the development of radio systems of information and team mates in the functioning of complexes with unmanned aerial vehicles, modeling and analysis of directional properties of the antenna system on the basis of the active phased array antenna (APAA) to ensure that decisions adopted targets CUAV, calculation and modeling of the use of various signal-code structures in the information and command-telemetry channels of the CUAV in order to assess the achievable data transfer rates and the energy budget of radio lines. The main tasks of the article are consideration of the nomenclature of possible target tasks of complexes with UAV, proposals for the technical implementation of radio systems of information and team mates, analysis of the advantages of using APAA as part of an unmanned aerial vehicle and a ground control and information processing point in order to improve the efficiency of UAV complexes in the end, modeling of directional diagrams in the azimuth and angular planes for receiving and transmitting modules and receiving and transmitting panels from the APAA, calculation of the energy budget of information and command radio lines, plotting the dependences of the error probability on the signal-to-noise ratio for various combinations of signal modulations and noise-tolerant encoding for various models of communication channels (additive white Gaussian noise – AWGN and binary-symmetric channel – BSC). The main results of the article include specific practical suggestions and recommendations on hardware and software approaches to the creation of antenna systems based on APAA as part of the CUAV, the effect of using modern approaches (reliability, reduced labor intensity and cost, improved performance characteristics), calculated signal-to-noise ratios for various methods of modulation and encoding, obtained as a result of calculations and mathematical modeling, as well as suggestions and recommendations for the rational use of specific types of signals and codes in the command-telemetry and information channels.

Conceptual Features of a Complex Vision of the World

Today, almost all spheres of human existence are interpreted – directly or indi­rectly – as permanently becoming, interpreted from a processing point of view realities that do not imply either final fixation or predetermined ultimate goals or states. The world appears not so much in the form of difficult composite dy­namic formation in mechanistic sense, but in the form of mobile, continuously becoming environment, which presupposes special technical researches and ways of staying in it. Such techniques and methods lead to the formation of a non-trivial vision of the universe. And such a vision, aimed at comprehending of emerging realities, presupposing conceptual shifts in modern natural science, technology, humanitarian activity, and more broadly in the very perception of na­ture and society, V.I. Arshinov endows with the epithet “complexity”. In the pro­posed text, a small fragment from the creative heritage of one of the most influ­ential philosophers J. Simondon will be considered, allowing to partially reveal the features of such complexitly oriented thinking. The central theme of Simon­don's philosophical strategy is the conceptualization of how the becomings of beings are realized, or how beings (inanimate, living, technical, mental, social) are individuated. Simondon begins the discussion of this plot with criticism of the hylemorphic scheme, which posits the genetic principle of existence in the form-matter dichotomy and, above all, in the interpretation and theoretical use of such a dichotomy by Aristotle, since, according to Simondon, it was this pair that contributed to the formation of a static view on the world, man and society. In different performances, the form-matter dichotomy can be interpreted in the form of mind-body dichotomies, artificial-natural, living-nonliving, etc. Note that Simondon begins his criticism with the technological substantiation of the hylemorphic scheme, referring to the operation of making a parallelepiped brick from clay

Epoxy and epoxy novolac adhesive binders for structural reinforcement systems of building structures

Today, along with the conventional materials and technologies for reinforcement of building structures, structural reinforcement systems based on polymer composites are on the rise. A structural reinforcement system consists of reinforcing filler (normally, carbon fiber fabric or cloth) and adhesive binder. This paper investigated the modification of epoxy nobake adhesive binders to provide a higher thermal resistance and mechanical strength. The influence of epoxy novolac resins on the processing and physical mechanical properties of adhesives, thermal resistance is studied. It is found that the substitution of 20-40 % of epoxy resin for epoxy novolac resin in the binder increases tensile strength by 40-50 %, flexural modulus by 20 %, reduces tensile strain, and increases glass transition temperature. From the processing point of view, it is more reasonable to use liquid epoxy novolac resins with a functionality of f=2.5 than semisolid resins with a functionality of f=3.6.

Study of the Properties of a Biodegradable Polymer Filled with Different Wood Flour Particles

Lignocellulosic wood flour particles with three different sizes were used to reinforce Solanyl® type bioplastic in three compositions (10, 20, and 30 wt.%) and further processed by melt-extrusion and injection molding to simulate industrial conditions. The wood flour particles were morphologically and granulometric analyzed to evaluate their use as reinforcing filler. The Fuller method on wood flour particles was successfully applied and the obtained results were subsequently corroborated by the mechanical characterization. The rheological studies allowed observing how the viscosity was affected by the addition of wood flour and to recover information about the processing conditions of the biocomposites. Results suggest that all particles can be employed in extrusion processes (shear rate less than 1000 s−1). However, under injection molding conditions, biocomposites with high percentages of wood flour or excessively large particles may cause an increase in defective injected-parts due to obstruction of the gate in the mold. From a processing point of view and based on the biocomposites performance, the best combination resulted in Solanyl® type biopolymer reinforced with wood flour particles loaded up to 20 wt.% of small and medium particles size. The obtained biocomposites are of interest for injected molding parts for several industrial applications.

Laryngeal Image Processing of Vocal Folds Motion

This review provides a comprehensive compilation, from a digital image processing point of view of the most important techniques currently developed to characterize and quantify the vibration behaviour of the vocal folds, along with a detailed description of the laryngeal image modalities currently used in the clinic. The review presents an overview of the most significant glottal-gap segmentation and facilitative playbacks techniques used in the literature for the mentioned purpose, and shows the drawbacks and challenges that still remain unsolved to develop robust vocal folds vibration function analysis tools based on digital image processing.

Processing of signals at the multiradar system on the basis of surveillance two coordinate radars

The subject of research in this work is the problem of developing methods for signal processing in a multi-radar system based on two-coordinate surveillance radar stations with mechanical azimuth rotation. The aim of the article is to improve the quality of airborne objects detection by combining the same type of two coordinate radars in a multi-radar system. It is proposed to combine the existing surveillance radar stations into a spatially separated coherent multi-radar system. The processing of radar information from individual positions of such spatially separated systems is carried out in the central information processing point, which can be combined with one of the positions or located separately. At such a point, it is advisable to carry out joint processing of radar information, the efficiency of which depends on the degree of coherence that is provided in the system. The synthesis of optimal detectors of coherent and incoherent signals is carried out in the work. The characteristics of air object detection in a multi-radar system with compatible signal reception are evaluated. The results obtained: increasing the number of radar stations in the system, regardless of the degree of signal coherence, showed the greatest efficiency in terms of increasing the signal-to-noise ratio when moving from a stand-alone radar station to a two element system, the rational number of radar stations in a multi-radar system should not exceed four. The expected signal-to-noise gain in a system of four radars can be up to eighteen decibels for a system with coherent signals and up to eleven decibels for a system with incoherent signals. The using of more than four radars is impractical.

Application of lexical bundles from biotechnical research articles in ESP teaching

This paper presents the possibility of applying a list of lexical bundles that have been singled out as the most common or typical in the texts of native English speakers. These bundles belong to biotechnical research articles and this study reveals how lists of pedagogically useful lexical bundles could be transferred to teaching materials intended for non-native speakers who aim to master their writing skills for the purpose of writing research papers in English in the field of biotechnology. Lexical bundles occurring in all subcorpora of biotechnical articles, i.e. the subcorpora of forestry, landscape architecture, ecological engineering and wood processing point to those groups of lexical bundles whose proper use should first be adopted by the authors of biotechnical articles, while the results of their classification and further analyses provide materials for exercises which aim to teach their use. In this way, our research responds to the invitation of the study Chen and Baker (2010), which urges pedagogues and publishers to use more results of automated corpus research in teaching.