Search

We propose a new video vectorization approach for converting videos in the raster format to vector representation with the benefits of resolution independence and compact storage. Through classifying extracted curves in each video frame into salient ones and non-salient ones, we introduce a novel bipartite diffusion curves (BDCs) representation in order to preserve both important image features such as sharp boundaries and regions with smooth color variation. This bipartite representation allows us to propagate non-salient curves across frames such that the propagation, in conjunction with geometry optimization and color optimization of salient curves, ensures the preservation of fine details within each frame and across different frames, and meanwhile, achieves good spatial-temporal coherence. Thorough experiments on a variety of videos show that our method is capable of converting videos to the vector representation with low reconstruction errors, low computational cost, and fine details, demonstrating our superior performance over the state of the art. We also show that, when used for video upsampling, our method produces results comparable to video super-resolution.

To meet the urgent demands in a wide range of geospatial applications, such as road management, intelligent transportation systems, road safety evaluation, and traffic accident analysis, automatic and accurate extraction of 3D roads and associated geometric parameters from point clouds is receiving wide attention. In this paper, we propose an accurate 3D road boundary extraction and vectorization method to bridge the gap from unstructured mobile laser scanning (MLS) point clouds to the vector-based representation of road boundary. Firstly, we propose a supervoxel generation method to extract candidate curbs with fine border preservation and high computation efficiencies. Then the candidate curb supervoxels are recognized and clustered to produce continuous road boundary segments with a contracted distance clustering strategy. Finally, the vectorized road boundary is represented by fitting, tracking, and completion from the extracted road boundary segments, resulting in road geometric parameters including boundary location, road widths, turning radius, and slopes. The performance of the proposed method was evaluated on two large-scale datasets collected in urban and industrial areas. Comprehensive experiments reveal that the proposed method is robust to various road shapes and point densities, in terms of precision of 95.0% and recall of 91.0%, respectively.

In this paper, we evaluate the risk of auxiliary engineering of mine geological environment utilizing multiobjective wireless sensing detection. After vectorization of each evaluation factor, we evaluate the geological environment of a high-top mountain mine by fuzzy comprehensive evaluation method and get the evaluation results of our geological environment. Firstly, an in-depth analysis is carried out for the tracking gate that most association methods pay little attention to. In terms of optimization of traditional tracking gate, an attempt is made to propose a genetic algorithm fusion coding for tracking gate optimization, and then, the principal characteristics of traditional data association algorithm are focused on, and a particle swarm annealing association algorithm is proposed. Finally, the image information fusion of multisensors is studied, and the image information fusion level, fusion technology, and fusion process are discussed in detail, and the optimized discrete wavelet and color space alignment method is proposed, and good results are achieved through experiments. Among them, the area of poor geological environment evaluation is 0.32 km2, accounting for 6.6% of the whole study area, the area of average geological environment evaluation is 0.63 km2, accounting for 12.7% of the whole study area, and the area of good geological environment evaluation is 3.98 km2, accounting for 80.7% of the whole study area. The evaluation results can fully reflect the degree of mining influence on the Gaodingshan mining area and provide a theoretical basis for the study of mine environment restoration and management.

In this study, we present a new automatic deep learning-based network named Road Vectorization Network (RoadVecNet), which comprises interlinked UNet networks to simultaneously perform road segment...

Understanding the relationship between urban areas is essential to the extraction of human mobility patterns. Most popular algorithms use word embedding technology to vectorize urban areas to calculate the similarity between them. However, noise generated by the traffic hub area is often ignored, such as crossroads and road toll stations, which affect the training time of the model and the accuracy of vectorization. In this paper, we propose a vectorized representation method of urban geographic areas based on the distribution of trajectories, combining with the filtering of traffic hub areas based on the law of similarity to eliminate their noise. A novel visual analytics system is designed, which enables experts to analyse the relationship between urban areas and explore human mobility patterns based on this vectorized representation method. Finally, we analyse the relevance to long-distance areas and campus crowd mobility patterns on real datasets, which proves the effectiveness of our method.

Physiological studies have illustrated that speech can be used as a biomarker to analyze the severity of depression and different frequency bands of the speech spectrum contribute unequally for depression detection. To this end, we propose a Time-Frequency Attention (TFA) component and combine it with the Squeeze-and-Excitation (SE) component to form our Time-Frequency Channel Attention (TFCA) block for emphasizing those discriminative timestamps, frequency bands and channels. In addition, considering the time-frequency attributes of the data, a Time-Frequency Channel Vectorization (TFCV) block is proposed to vectorize the tensor. Furthermore, we merge the proposed blocks (i.e., TFCA and TFCV blocks) and the two blocks (i.e., Dense block and Transition Layer) of the DenseNet into a unified architecture to form our Time-Frequency Channel Attention and Vectorization (TFCAV) network. In this way, to predict the depression level of an individual, we firstly introduce the sphere embedding normalization method to preprocess the long-term logarithmic amplitude spectrum for maintaining the time-frequency attributes and divide it into segments. Then, these segments are input into the TFCAV network to obtain the depression scores. Finally, the average of scores is taken as the result corresponding to the long-term spectrum. Our method is validated on two challenging databases, i.e., AVEC2013 and AVEC2014 depression databases. The experimental performance illustrates the superiority of the proposed network over some previous methods.

Reliable urban road vector maps are essential for urban analysis because the spatial distribution of road networks reflects urban development under the combined effects of nature and socio-economics. Diverse very high resolution (VHR) remote sensing images are now available, enabling explicit extraction of urban road vector maps over wide areas. Urban road vectorization mapping consists of two separate tasks: road extraction and road vectorization. The traditional methods focus on the road extraction task, and can obtain a good performance when using a pixel-based metric. However, the road vectorization methods are faced with the problem of road connectivity. In this work, to implement urban road vectorization mapping in a unified way, an end-to-end road vectorization mapping network (RVMNet) framework is proposed. The proposed RVMNet framework consists of a node proposal network (NPN) module and a node connectivity based road refinement module. In the NPN module, a fully convolutional network is adopted for the road node extraction. This improves the connectivity of the road mask by providing supervised information in the form of the road nodes, which are actually part of the road mask. The road mask is then converted into a road vector map by vectorization. In the node connectivity based road refinement module, road nodes are inserted into the road vector map to improve the connectivity. We compared RVMNet with the other state-of-the-art road detection methods on two public road datasets (SpaceNet 3 and DeepGlobe). The results of this comparison showed that combining road extraction and road vectorization into a unified framework is an efficient and accurate strategy for urban road vectorization mapping because it can propose road nodes that help to improve the road connectivity. Moreover, we constructed the novel UrbanRoadNet dataset, covering six cities (Beijing city center; Helsinki; Wuhan; Macao; the Wan Chai area of Hong Kong; and Shanghai). We then applied the RVMNet framework to the data from the six cities, obtaining an improvement in the vector-based average path length similarity (APLS) value of 4.1%. The spatial transfer assessments from both the qualitative and quantitative aspects corroborated the robust generalizability of the proposed method, and further verified the effectiveness of the proposed approach for large-scale road vectorization mapping at a very high resolution. It was also found that road vector spatial distribution is a useful way to reflect urban development.

The article is devoted to the traditional sphere of application of Geoinformation systems – logistics and use of Geoinformation technologies in the process of preparation of future logisticians on the basis of the offered technique of carrying out laboratory employment with use of vectorization of a rural settlement map. It is claimed that Geoinformation technologies allow to present statistical information on maps and provide tools for processing spatial statistical information with taking into account its qualitative and quantitative properties. The reasons of efficiency of the mechanism of the decision of a logistic problem are defined, some widespread ways of optimization of logistic activity on the basis of the geographical approach are considered and the directions of its optimization are established. The main professionally necessary tasks that represent a valuable part of cartographic literacy of future professionals in the field of logistics are highlighted. Due to the priority of development and detailing the maps of urban settlements, there is a decrease in attention to rural and remoted areas, which, to some extent, creates some inconvenience in designing transport routes to the necessary points of location. The sequence of laboratory work on vectorization of the selected part of the map of the rural settlement is offered. Comparing it with the modes "Satellite" and "Map" and adhering to the selected fixed scale, additional graphic prompts are applied. This work is performed in the training of specialists in the specialty 015.38 Professional Education (Transport) at Ternopil Volodymyr Hnatiuk National Pedagogical

The article presents the results of research on the creation of information support for predicting flooding of the territories of the Saratov Left Bank region, which includes a database of data necessary for calculations, consisting of 14 relational files, including reference books, normative files and files of source data; algorithms and computer programs forecasting. Flooding boundaries vectorization is carried out in the QGIS environment based on the exported forecast results and digital elevation model. Development and implementation of the proposed information support will provide a improvement in the quality of forecasts and will pay off in 4 years of operation.

In Computer Graphics, the two main approaches to rendering and visibility involve ray tracing and rasterization. However, a limitation of both approaches is that they essentially use point sampling. This is the source of noise and aliasing, and also leads to significant difficulties for differentiable rendering. In this work, we present a new rendering method, which we call vectorization, that computes 2D point-to-region integrals analytically, thus eliminating point sampling in the 2D integration domain such as for pixel footprints and area lights. Our vectorization revisits the concept of beam tracing, and handles the hidden surface removal problem robustly and accurately. That is, for each intersecting triangle inserted into the viewport of a beam in an arbitrary order, we are able to maintain all the visible regions formed by intersections and occlusions, thanks to our Visibility Bounding Volume Hierarchy structure. As a result, our vectorization produces perfectly anti-aliased visibility, accurate and analytic shading and shadows, and most important, fast and noise-free gradients with Automatic Differentiation or Finite Differences that directly enables differentiable rendering without any changes to our rendering pipeline. Our results are inherently high-quality and noise-free, and our gradients are one to two orders of magnitude faster than those computed with existing differentiable rendering methods.

Surveys of roadways with Mobile Laser Scanning (MLS) are nowadays the faster and more secured way to collect topographic data compared with conventional techniques. To deliver topographic plans, the voluminous data collected by the MLS device need to be processed. If the acquisition step is quite fast, the second part of interpretation and vectorization of the LiDAR data and the panoramic images is laborious and time consuming. This paper proposes two approaches that have been developed in order to reduce the time required to process roadway MLS data. The first one is about automatic detection of pole like objects, and the second one is about the detection of linear objects. The presented workflow try to automatically extract a 3D position for each object from MLS Data.

Increasing knowledge about what actions are most effective in sport is essential to improve performance in the game. This study aims to analyse the offensive finalization in soccer during the World Cup '18. Specifically, the finalist teams of the tournament were analysed. It will take into account the regular patterns of play and tactics that occur in the finalization actions of the finalist teams in the tournament. Observational methodology and polar-coordinate analysis have been used for dramatic data reduction and behavioural vectorization. An ad hoc observation tool has been designed and validated in the form of field formats and category systems, comprehensive and mutually exclusive (E/ME), with 8 criteria and 59 categories. The polar coordinate analysis used is the "genuine" technique with significant relationships (p

The paper analyses the support for vectorization that can be found in some programming languages, and the ways it could also be used in Ada. A proposal for an Ada extension for enhanced vectorization support is included.

Analiza sentimenta je naučno polje koje se bavi analizom mišljenja, stavova i emocija ljudi koji su napisali određeni tekst. Ovakva analiza može se koristiti u svrhe praćenja brendova, poboljšanja korisničke podrške, analize proizvoda, istraživanja tržišta ili u svrhe kreiranja sistema za preporuke. U ovom radu obučeno je i upoređeno više modela mašinskog učenja za zadatku analize sentimenta. Recenzije korišćene kao skup podataka prikupljene su sa jednog od sajtova za dostavu hrane u Srbiji. Recenzije sadrže skup ocena koje će biti korišćene kao pokazatelj emocionalne polarnosti. Trenirana su i upoređena tri modela mašinskog učenja sa različitim vrstama vektorizacije i rezultati su upoređeni sa pristupom transfera učenja. Transfer učenja je metoda dubokog učenja gde se model obučen da reši jedan problem koristi kao polazna tačka pri rešavanju drugog problema. Za transfer učenja korišćen je model dubokog učenja zasnovan na transformerima.

In this paper, we propose an original manner to employ a tangential cover algorithm - minDSS - in order to vectorize noisy digital contours. To do so, we exploit the representation of graphical objects by maximal primitives we have introduced in previous work. By calculating multi-scale and irregular isothetic representations of the contour, we obtained 1-D (one-dimensional) intervals, and achieved afterwards a decomposition into maximal line segments or circular arcs. By adapting minDSS to this sparse and irregular data of 1-D intervals supporting the maximal primitives, we are now able to reconstruct the input noisy objects into cyclic contours made of lines or arcs with a minimal number of primitives. We explain our novel complete pipeline in this work, and present its experimental evaluation by considering both synthetic and real image data.

Automatic themes-based classification of Quran verses is the process of classifying verses to predefined categorizes or themes. It is an essential task for all Muslims and people interested in studying the Quran. Quran themes-based classification could be used in many natural language processing (NLP) fields such as search engines, data mining, question–answering systems, and information retrieval applications. This paper presents an ensemble multi-label classification model that automatically identifies and classifies the Quran verses based on themes/topics. The model is composed of four phases: pre-processing, data vectorization, binary relevance classifier, and voting module. Firstly, the verses of the second chapter of the Quran (Al-Baqarah) are tokenized and normalized. Then, the topics of these verses are manually labeled based on “Mushaf Al-Tajweed” classification. Secondly, verses are converted into features’ vectors using term frequency-inverse document frequency (TF-IDF) and word2vec techniques. Word2vec is used to consider the semantic meaning of Quranic words and to improve performance. Also, they are trained on a collected classic Arabic corpus of 200 million words. Then, the binary relevance multi-label classification technique is applied using three different classifiers: logistic regression, support vector machine, and random forest, which categorize verses into 393 topics/tags. Finally, the voting module is applied by picking the tags with the maximum prediction probability among the three classifiers. The results of the three classifiers and the ensemble model are compared against “Mushaf Al-Tajweed.” The ensemble model outperforms the three classifiers. Its average hamming loss, recall, precision, and F1-Score are 0.224, 81%, 75%, and 77%, respectively.

This study describes the semi-automated pipeline created for the comprehensive analysis of the urban areas with the extremely low and extremely high popularity levels. It includes the geo-frequency analysis of the Russian-language Instagram publications for the St. Petersburg area and selection of areas with the extreme values of the popularity level according to the number of publications in them. Semantic analysis of the urban areas with an extremely low number of publications includes comparing of algorithms for descriptions extraction and classification for these areas and results of such descriptions extraction and classification using TF-IDF vectorization technique and most valuable words extraction. Semantic analysis of areas with an extremely high number of publications includes the structure description of such areas, comparing of algorithms for advertisement publications extraction, results of the advertisement extraction using BigARTM model and further development and implementation of the algorithm for extracting events related to the the points of attraction in extremely popular urban areas, which is based on the strong time binding hypothesis and the idea of similarity queries using combination of LDA models for revealing semantic structure and algorithm based on frequency analysis. Developed algorithm was tested to extract events in the urban area of St. Petersburg where Ice Palace is placed and showed interpretable results and allow us to correctly extract 89 events out of 102 which occurred in this area in 2019. Finally, SeSAM pipeline for comprehensive urban analysis was created that combined the described algorithms.

Widespread application of supercomputer technologies in various spheres of life, as well as the need of high-performance calculations allows us to speak about the relevance of the problem of increasing the performance of computer codes on supercomputers of modern architectures. Vectorization of program code is a low-level optimization that can, with a relatively local and compact application, increase the productivity of computational codes by several times. Modern Intel microprocessors have support for a unique set of instructions AVX-512, which, due to its features, allows you to vectorize almost any kind of code written in a predicate form. A set of simple restrictions when developing programs along with vectorization tools to enable the use of the AVX-512 instruction set can significantly speed up the resulting program. The article discusses approaches to vectorization of flat loops—a special-purpose program context, the successful vectorization of which allows to increase the productivity of supercomputer applications even for such program code for which optimizing compilers are powerless.

A utilização de sistemas de informação geográfica (SIG) surge como uma ferramenta útil para compreender as alterações ambientais, uma vez que permite uma análise rápida, fiável e de baixo custo através de técnicas especializadas de processamento de imagens. Este artigo, demonstra a aplicação de ferramentas de geoprocessamento para identificar características básicas relacionadas à morfologia costeira. estudos. A área de estudo, a planície costeira do Balneário Piçarras, apresenta problemas associados aos processos de erosão costeira, que acarreta perdas econômicas que requerem, constantemente, o controle de intervenções humanas ou mitigação dos danos associados. O transporte de sedimentos não consolidados causa mudanças em diferentes escalas de tempo geológicas em um equilíbrio de variáveis ​​naturais e antropogênicas. A área de estudo foi limitada a um polígono obtido a partir de uma imagem de satélite (Quickmapservice /Google Satélite). A localização dos elementos e pontos de amostragem, com destaque para a importância da inserção de polígonos e pontos, foi realizada por meio do software Q-GIS 3.10. A vetorização e detalhamento das imagens permitiram reconhecer melhor os aspectos da área estudada, pela criação de diferentes camadas para o estudo da morfologia da faixa arenosa da praia central do Balneário Piçarras. a partir de imagens de satélite predefinidas, usando a camada de base Google Quickmapservices disponível no Q-GIS 3.10. O mapa resultante produzido serve de base para novos estudos e aprendizados durante a área de processos destacada.

Visual secret sharing (VSS) has the advantage that the decryption is based on our human visual system without participation of any computational devices. However, traditional VSS schemes are only for sharing raster images with regard to pixels, which lead to that the secret image will be aliased when enlarged and its pixels are shared in rectangular way only. In this paper, we will introduce a VSS scheme for vectorized images based on Penrose tiling. Penrose tiling is with the merits of vectorization and nonperiodicity. These properties are applied to the proposed scheme so as to share those vectorized images; the basic unit of secret sharing could be any graphical shapes instead of pixels or rectangular regions only in the traditional methods. Our experiments show the effectiveness of the proposed scheme.

Introduction: In recent years, sentiment analysis has found practical application in many areas, such as evaluating the quality of products and services based on customers’ online reviews, analyzing negative emotions in messages, forecasting stock markets or political situations based on news data. In this regard, a large number of systems and methods for Russian text sentiment analysis are being developed. Purpose: A detailed review of approaches, and comparative analysis of available databases in the field of Russian text sentimental analysis. Results: Our analytical review of the approaches to Russian text data sentiment analysis has shown that there are a large number of ways for preprocessing, vectorization and machine classification of the text data. Studying the available databases shows that the Russian text sentimental analysis is less developed than that for other major world languages. Studying the existing software systems for Russian text analysis reveals their low accuracy compared to English, which can be caused by the sophisticated structure of Russian. Discussion: In our further research, we plan to implement sentiment analysis of spoken speech using audio data. To do this, we will need to obtain a spelling transcription of speech for each speaker.

In urban planning, a numerical and spatially based approach is expected to drive to the “best” choice. In this work a GIS-based procedure is proposed to model territorial dynamics by comparing maps of two different periods (1830 and 2000). The study area is located in the urban fringe of Torino (NW Italy) that suffered from important changes especially in the last 60 years. A workflow was defined and applied, based on a multi-criteria approach implemented by GIS. With reference to existing maps, strength and direction of forces, opposing urban to rural/semi-natural surroundings, were mapped and described operating through the following scheme: a) vectorization and qualification of both impacted (rural) and impacting (urban) landscape elements; b) implementation of spatially dependent functions representing strength and direction of urban pushes against rural; c) qualification of rural areas majorly exposed to urban growth. Accordingly, some maps, useful to read the playing urban growth dynamics, were generated and some applications proposed. Nevertheless some limitations still persist: the proposed methodology is based on simplified hypotheses, mainly related to the definition of spatial indices that somehow depend on the type of information that available maps contain. A second limitation is related to the persistence of a great component of subjectivity during the extraction of the starting information from the available maps and weights assignation.

A major challenge in line drawing vectorization is segmenting the input bitmap into separate curves. This segmentation is especially problematic for rough sketches, where curves are depicted using multiple overdrawn strokes. Inspired by feature-aligned mesh quadrangulation methods in geometry processing, we propose to extract vector curve networks by parametrizing the image with local drawing-aligned integer grids. The regular structure of the grid facilitates the extraction of clean line junctions; due to the grid's discrete nature, nearby strokes are implicitly grouped together. We demonstrate that our method successfully vectorizes both clean and rough line drawings, whereas previous methods focused on only one of those drawing types.

This paper describes a sequence of actions developed to guarantee a reliable and suitable dataset for the creation of detailed ortho images of Nativity Church mosaics. During acquisition campaigns, different photogrammetric techniques were tested, and different survey instruments were compared to improve the quality of the data obtained. The different outputs allow the adjustment of the instrument parameters and the acquisition methods, to structure a methodological process aimed at obtaining an accurate level of detail to describe the individual mosaic tile.From the realization of a reliable photomosaics an automatic vectorization system has been developed. This process, aimed at digitizing the tiles of the Church walls and the pavement mosaics, responds to a documentation and management purpose and to an objective of structuring a data acquisition method and post-production that can be replicated on other mosaic contexts.

As smart devices and cloud services are rapidly expanding, a large amount of location information can easily be gathered. However, there is a conflict between collecting location data and protecting personal data since obtaining and utilizing the data may be restricted due to privacy concerns. Various methods for anonymity and on the original location data have been studied, but these methods have excessively reduced data utility while stressing highly on privacy preservation. In this article, we suggest a novel model to overcome this fundamental dilemma via a surrogate vector based on the grid environment. Compared to the existing approaches, our model shows a new theoretical advancement in privacy protection, and outstanding performance with respect to time complexity and data utility has been achieved.

The research is about the problem of developing an effective way to determine the stylistic affiliation of texts. Styles such as scientific, journalistic and official-business are considered. Texts of one subject – about language – selected for the analysis. Different combinations of vectorization methods and architectures of artificial neural networks considered which would provide a high level of recognition. Among the architectures of artificial neural networks– Support Vector Machines (SVM) (C-Support Vector Classification (SVC), Epsilon-Support Vector Regression (SVR)) and Multi Layer Perseptron (MLP). Among the vectorization methods– HeshingVectorizer, CountVectorizer and TfidVectorizer. Studies have shown that all the approaches considered most effectively distinguish between official and business texts, due to their greatest standardization. Scientific and official business styles are especially effective. The considered methods show the least accuracy in determining the stylistic affiliation, when one of the styles is journalistic. The most effective approach to determining stylistic affiliation was the combination of the vectorization method and both architectures of artificial neural networks Support Vector Machines. In the previous stage, word stemming was increasing efficiency. In texts with at least 500 characters, this approach has ensured an accuracy of 94-98%, and the time for learning an artificial neural network does not exceed one second on computers of the standard configuration at this time. With the help of the Lime library, a visualization of the study of the operation of an artificial neural network presents, which is extremely important empirical material for philologists for further linguistic analysis.

In this paper, we propose a new parallel ordering method to vectorize and parallelize the sparse triangular solver, which is called hierarchical block multi-color ordering. In this method, the parallel forward and backward substitutions can be vectorized while preserving the advantages of block multi-color ordering, that is, fast convergence and fewer thread synchronizations. To evaluate the proposed method in a parallel ICCG (Incomplete Cholesky Conjugate Gradient) solver, numerical tests were conducted using five test matrices on three types of computational nodes. The numerical results indicate that the proposed method outperforms the conventional block and nodal multi-color ordering methods in 13 out of 15 test cases, which confirms the effectiveness of the method., Comment: 19 pages

PurposeThe purpose of this study is to investigate the effectiveness of qualitative information extracted from firm’s annual report in predicting corporate credit rating. Qualitative information represented by published reports or management interview has been known as an important source in addition to quantitative information represented by financial values in assigning corporate credit rating in practice. Nevertheless, prior studies have room for further research in that they rarely employed qualitative information in developing prediction model of corporate credit rating.Design/methodology/approachThis study adopted three document vectorization methods, Bag-Of-Words (BOW), Word to Vector (Word2Vec) and Document to Vector (Doc2Vec), to transform an unstructured textual data into a numeric vector, so that Machine Learning (ML) algorithms accept it as an input. For the experiments, we used the corpus of Management’s Discussion and Analysis (MD&A) section in 10-K financial reports as well as financial variables and corporate credit rating data.FindingsExperimental results from a series of multi-class classification experiments show the predictive models trained by both financial variables and vectors extracted from MD&A data outperform the benchmark models trained only by traditional financial variables.Originality/valueThis study proposed a new approach for corporate credit rating prediction by using qualitative information extracted from MD&A documents as an input to ML-based prediction models. Also, this research adopted and compared three textual vectorization methods in the domain of corporate credit rating prediction and showed that BOW mostly outperformed Word2Vec and Doc2Vec.

The problem of development of an effective method for text authorship identification (on the material of publications of well-known Ukrainian journalists) is explored. Most existing methods require text preprocessing, which entails new costs when solving the set problem. In the case where the number of possible authors can be minimized, this approach is often excessive. Another disadvantage of the existing approaches is that their vast majority was applied to texts in foreign languages and did not take into consideration the peculiarities of the Ukrainian language. Therefore, it was decided to develop an approach that makes it possible to identify the author of the text in Ukrainian without preprocessing and give high accuracy results, as well as to establish what types of artificial neural networks provide the minimum error for Ukrainian publicists. The developed method uses a multilayer perceptron of direct distribution, the algorithm of supervised learning, vectorization HashingVectorizer, and Adam optimizer. It was determined that with a small number of iterations (4–5 iterations) of artificial neural network learning, we obtain a rather high accuracy of identification of authorship of journalistic texts and rather small value of error. Over 1,000 fragments of texts by three Ukrainian authors were used. As a result of the conducted experiments, it was found that the application of the developed approach to solving the set problem enables achieving rather high results. In the texts containing not less than 500 characters, the accuracy reaches 91%, and the maximum number of iterations of artificial neural network learning does not exceed 15. Such results were achieved primarily due to the efficient selection of the vectorization method at the preparatory stage and the structure of an artificial neural network

A shoreline is a highly dynamic part of the earth’s surface. Advanced remote sensing (RS) and geographic information system (GIS) techniques are being used for detection of shoreline position and change analysis. In this paper, a new methodology for automatic shoreline extraction is demonstrated and analyzed using Landsat 8 Operational Land Imager (OLI)/Thermal Infrared Sensor (TIRS) images. The methodology involves several stages consisting of preprocessing of satellite images, band selection, coastal water index (CWI) preparation, normalization of binary images, Otsu thresholding technique (named after Nobuyuki Otsu) for the land and water separation, image noise correction with morphological filter (image morphology), seawater separation from waterbody, vectorization of classified binary image, polyline conversion from polygon vector, shoreline selection, and generalization of the final shoreline. The positional accuracy of the final shoreline is evaluated with expert captured shoreline. It was observed that the average positional difference between computer generated shoreline and expert digitized shoreline was less than a pixel resolution. The proposed methodology is very helpful in any coastal application where the shoreline is used as a parameter. It also reduces the time of human intervention.

Road markings are one of the most important safety elements in a road network, and they play a critical role in traffic safety. However, the automatic extraction of road markings remains a technical challenge in the fields of smart city construction and automatic driving. This paper presents an image-translation-based method of obtaining the 3D vectors of typical road markings from mobile laser point clouds. First, ground roughness is used as a criterion to extract ground points based on the topological relationship of adjacent scan lines, and the feature images of a road surface are generated using the adapted inverse distance weighted method. Second, by comparing objective functions based on the pix2pix framework, a finely adjusted image-to-image translation model named P2P_L1 is proposed for the segmentation of road markings. The proposed model outperforms the advanced DeepLab V3+ network in terms of precision, F1-score, and mean Intersection over Union indicators in the comparative segmentation results of ten types of road markings in the Shenzhen test area. Third, methods such as node averaging and optimized iterative closest point are developed for the 3D vectorization of road markings. This study presents a new approach for the automatic extraction of road markings to provide effective technical support for the construction of smart cities.

The vector graphics with gradient mesh can be attributed to their compactness and scalability; however, they tend to fall short when it comes to real-time editing due to a lack of real-time rasterization and an efficient editing tool for image details. In this paper, we encode global manipulation geometries and local image details within a hybrid vector structure, using parametric patches and detailed features for localized and parallelized thin-plate spline interpolation in order to achieve good compressibility, interactive expressibility, and editability. The proposed system then automatically extracts an optimal set of detailed color features while considering the compression ratio of the image as well as reconstruction error and its characteristics applicable to the preservation of structural and irregular saliency of the image. The proposed real-time vector representation makes it possible to construct an interactive editing system for detail-maintained image magnification and color editing as well as material replacement in cross mapping, without maintaining spatial and temporal consistency while editing in a raster space. Experiments demonstrate that our representation method is superior to several state-of-the-art methods and as good as JPEG, while providing real-time editability and preserving structural and irregular saliency information.

Natural language is ubiquitous in the workflow of medical imaging. Radiologists create and consume free text in their daily work, some of which can be amenable to enhancements through automatic processing. Recent advancements in deep learning and "artificial intelligence" have had a significant positive impact on natural language processing (NLP). This article discusses the history of how researchers have extracted data and encoded natural language information for analytical processing, starting from NLP's humble origins in hand-curated, linguistic rules. The evolution of medical NLP including vectorization, word embedding, classification, as well as its use in automated speech recognition, are also explored. Finally, the article will discuss the role of machine learning and neural networks in the context of significant, if incremental, improvements in NLP.

As Noncommunicable Diseases (NCDs) are affected or controlled by diverse factors such as age, regionalism, timeliness or seasonality, they are always challenging to be treated accurately, which has impacted on daily life and work of patients. Unfortunately, although a number of researchers have already made some achievements (including clinical or even computer-based) on certain diseases, current situation is eager to be improved via computer technologies such as data mining and Deep Learning. In addition, the progress of NCD research has been hampered by privacy of health and medical data. In this paper, a hierarchical idea has been proposed to study the effects of various factors on diseases, and a data-driven framework named d-DC with good extensibility is presented. d-DC is able to classify the disease according to the occupation on the premise where the disease is occurring in a certain region. During collecting data, we used a combination of personal or family medical records and traditional methods to build a data acquisition model. Not only can it realize automatic collection and replenishment of data, but it can also effectively tackle the cold start problem of the model with relatively few data effectively. The diversity of information gathering includes structured data and unstructured data (such as plain texts, images or videos), which contributes to improve the classification accuracy and new knowledge acquisition. Apart from adopting machine learning methods, d-DC has employed knowledge graph (KG) to classify diseases for the first time. The vectorization of medical texts by using knowledge embedding is a novel consideration in the classification of diseases. When results are singular, the medical expert system was proposed to address inconsistencies through knowledge bases or online experts. The results of d-DC are displayed by using a combination of KG and traditional methods, which intuitively provides a reasonable interpretation to the results (highly descriptive). Experiments show that d-DC achieved the improved accuracy than the other previous methods. Especially, a fusion method called RKRE based on both ResNet and the expert system attained an average correct proportion of 86.95%, which is a good feasibility study in the field of disease classification.

Risk is commonly found in corporate operation and management activities. It’s possible to turn risks into opportunities if companies can take effective measures to reduce mistakes or losses. This study proposes an improved method to reveal company-specific risk factors (take corporate crash risk as an example) in the view of cross analysis based on financial data and textual data. The core of the method is to propose a more suitable text vectorization approach for the quantitative measurement of risk fields in corporate annual reports, and on this basis, to conduct cross analysis based on financial data and risk description textual data. The experimental results show that the information content of risk description text fields is quite high, especially higher than that of the management discussion and analysis (MDA and that the information content of risk description text fields has positive effects on the stock price crash risk.

This paper discusses the issues of creating an aerial survey complex based on an unmanned aerial vehicle. The analysis of the efficiency of using aerial survey complexes based on unmanned aerial vehicles for operational monitoring of the state of natural and man-made territorial complexes is carried out. The paper considers the process of developing a modified control system for unmanned aerial vehicles, as well as the subsequent photogrammetric processing of aerial photographs. The quality of electronic units installed on unmanned aerial vehicles has been determined. The created modified control system for unmanned aerial vehicles and geographically referenced terrain orthophotomaps obtained by photogrammetric methods make it possible to classify the captured objects, perform their vectorization and provide each object with semantic information. The developed modified aerial survey complex will allow real-time monitoring of the state of natural and man-made territorial complexes based on remote sensing data from UAVs.

Sentiment analysis is one of the important fields in educational data mining. In this paper, a large dataset, more than 52 000 comments, was used during experiment to develop a state-of-art classification model. The correlation test was conducted on sentiment analysis results and scale-rated survey results, and the result (r(203)=.79, p

Semantic Web vision is to have machines interpret and understand the content of Web documents. There is a need to convert the existing Web of documents into an understandable format, which could be done by automatic semantic annotation. Annotation could be performed using a set of tools provided with general and domain-specific ontologies. The aim of this paper is to present a generic semantic annotation framework of Arabic text using deep learning models. The framework produces annotations using different output formats for a given set of Arabic documents and ontologies. With a prototype of the framework, the initial evaluation shows a promising performance using different public Arabic word embedding models with different vectorization and matching techniques.

Automatische Vektorisierungsalgorithmen eröffnen die Möglichkeit, Bitmap Grafiken mit sehr geringem Zeitaufwand und ohne spezieller Vorkenntnisse in Vektorgrafiken umzuwandeln. Vektorgrafiken bieten Vorteile in verschiedenen Gebieten durch deren unbegrenzte Skalierbarkeit sowie geringer Dateigröße bei simplen Inhalten wie etwa Logos, Grafiken, etc. Auch im Bereich der Kalligrafie ist die Vektorisierung der erstellten Schriftstücke teilweise notwendig, etwa wenn diese im Großformat gedruckt, effizient auf Webseiten eingebunden oder als Logos verwendet werden sollen. In dieser Bachelorarbeit sind die involvierten Schritte von automatischen Vektorisierungslösungen beschrieben und in einer praktischen Implementierung anhand bekannter Konzepte umgesetzt. Im Ausblick auf eine Verbesserung von automatischer Vektorisierung von Kalligrafie im Copperplate Stil ist ein neuer Ansatz zur Eckpunkterkennung vorgestellt und umgesetzt. Dieser basiert auf der Annahme, dass Ecken nur dort auftreten, wo die Kontur eine hohe Krümmung aufweist. Rund um solche Extrema werden Regressionslinien durch die benachbarten Konturpunkte gelegt. Anhand derer Kreuzungswinkel werden Eckpunkte bestimmt, die dann auf die Schnittpunkte der Linien gelegt werden. Der anschließende Vergleich mit der Referenzlösung als auch anderen Vektorisierungslösungen zeigt, dass der in dieser Arbeit vorgeschlagene Algorithmus gute Ergebnisse erzielt und ein hohes Potential zur Verbesserung von automatischer Vektorisierung von Kalligrafie im Copperplate Script Stil bietet. Automatic vectorization algorithms open the possibility to convert bitmap graphics into vector graphics with very little time and without special previous knowledge. Vector graphics offer advantages in various fields due to their unlimited scalability as well as small file size for simple content such as logos, graphics, etc. Also in the field of calligraphy, vectorization of the created artworks is sometimes necessary, for example if they are to be printed in large format, efficiently integrated on websites or used as logos. In this bachelor thesis the involved steps of automatic vectorization solutions are described and realized in a practical implementation based on known concepts. In the perspective of improving automatic vectorization of Copperplate calligraphy, a new approach for corner point detection is presented and implemented. The proposed corner detection algorithm is based on the assumption that corners only occur where the contour has a high curvature. Around such curvature extrema, regression lines are placed through the neighboring contour points. Based on their crossing angles, corner points are determined, which are then placed on the intersection points of the lines. The subsequent comparison with the reference solution as well as other vectorization solutions shows that the algorithm proposed in this thesis achieves good results and offers a high potential for improving automatic vectorization of calligraphy in Copperplate Script style. Martin Wechtitsch Bachelorarbeit FH JOANNEUM 2022

Internet has turned into an online learning network and a platform to exchange ideas and reviews. Social networking sites are quickly gaining popularity as it allows users to have a discussion, share, and express their views on subjects across the globe. It is generally known that social media and social networks are the best tools to collect knowledge about the viewpoint and the thoughts of people on entirely different subjects as they pay a considerable amount of time on social media networks to express opinions and interests. The proposed model detects the sentiments of tweets and using Twitter sentiment, and this research work has also attempted to perform classification on different models using TF-IDF vectorization. Python3 and its libraries are used for implementation. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

International audience; Maps have been a unique source of knowledge for centuries. Such historical documents provide invaluable information for analyzing the complex spatial transformation of landscapes over important time frames. This is particularly true for urban areas that encompass multiple interleaved research domains (social sciences, economy, etc.). The large amount and significant diversity of map sources call for automatic image processing techniques in order to extract the relevant objects under a vectorial shape. The complexity of maps (text, noise, digiti-zation artifacts, etc.) has hindered the capacity of proposing a versatile and efficient raster-to-vector approaches for decades. We propose alearnable, reproducible, and reusable solution for the automatic transformation of raster maps into vector objects (building blocks, streets,rivers). It is built upon the complementary strength of mathematical morphology and convolutional neural networks through efficient edge filtering. Even more, we modify ConnNet and combine with deep edgefiltering architecture to make use of pixel connectivity information and built an end-to-end system without requiring any post-processing techniques. In this paper, we focus on the comprehensive benchmark on various architectures on multiple datasets coupled with a novel vectorization step. Our experimental results on a new public dataset using COCO Panoptic metric exhibit very encouraging results confirmedby a qualitative analysis of the success and failure cases of our approach. Code, dataset, results and extra illustrations are freely available at https://github.com/soduco/ICDAR-2021-Vectorization

Bitmap vectorization conversion methodology in graphic design with OpenGL framework is designed in this paper. As noise and edges are both high-frequency components in image processing, figuring out how to minimize noise while boosting edges is a crucial challenge for edge extraction. Generally, the results of the source coordinates and the corresponding target coordinates of more than two points are required to obtain the transformation parameters between the two coordinate systems. With this model, the OpenGL is combined for the analysis with the graphic model. The author uses the difference between the coordinates of the obvious feature points such as features on the vectorized digital map and the coordinates of the actual location to evaluate the accuracy of the image geometric correction results. The data is collected from the Internet, and the proposed model is validated when compared with the other methods.

Efficient data flow optimization for the Internet middleware on application to the ideological online interactive system is studied in this manuscript. Our application research is mainly to discover the calling relationship and calling methods between the application and the database, to then clarify the compatibility characteristics of the application modules and the database calling middleware, and to clarify the transformation points of the application in the conversion of each module. The superword-level parallel vectorization is used at a finer granularity, and then it will be applied for the data flow analysis. The designed model is then applied to the online interactive system. The proposed model can effectively improve the online class efficiency and the feedback information is positive.

Accurate building footprint polygons provide essential data for a wide range of urban applications. While deep learning models have been proposed to extract pixel-based building areas from remote sensing imagery, the direct vectorization of pixel-based building maps often leads to building footprint polygons with irregular shapes that are inconsistent with real building boundaries, making it difficult to use them in geospatial analysis. In this study, we proposed a novel deep learning-based framework for automated extraction of building footprint polygons (DLEBFP) from very high-resolution aerial imagery by combining deep learning models for different tasks. Our approach uses the U-Net, Cascade R-CNN, and Cascade CNN deep learning models to obtain building segmentation maps, building bounding boxes, and building corners, respectively, from very high-resolution remote sensing images. We used Delaunay triangulation to construct building footprint polygons based on the detected building corners with the constraints of building bounding boxes and building segmentation maps. Experiments on the Wuhan University building dataset and ISPRS Vaihingen dataset indicate that DLEBFP can perform well in extracting high-quality building footprint polygons. Compared with the other semantic segmentation models and the vector map generalization method, DLEBFP is able to achieve comparable mapping accuracies with semantic segmentation models on a pixel basis and generate building footprint polygons with concise edges and vertices with regular shapes that are close to the reference data. The promising performance indicates that our method has the potential to extract accurate building footprint polygons from remote sensing images for applications in geospatial analysis.

Service Oriented Architecture (SOA), as a new architecture paradigm, has been dramatically developed. A growing number of Web services have been developed and published on the Web, which increases the burden of service selection in system construction. Nonfunctional attributes of services have an essential impact on service selection. Users may observe the dynamic nonfunctional property values of different web services in various locations due to the dynamic attributes of various web services. How to accurately predict user perception of nonfunctional attributes of services has become an important problem in services computing. Towards this issue, this paper improves the personalized LSTM model by adding the attention mechanism to obtain more accurate vectorization representations of users and services, so as to better forecast missing nonfunctional service attributes. The proposed approach is evaluated on a real-world dataset, and experiment results show its feasibility.

Bug handling processes aimed at efficient defect resolution provisioning is an important part of a software development lifecycle and usually has a very formal process definition in modern and professional large software development organizations. Improvements of such process may include automated bug assignment, which is a task of selecting a correct development team for further investigations of a bug report. As bug reports contain lots of natural language descriptions, the bug assignment becomes a non-trivial task, especially in testing of large-scale projects or complex systems. This research focuses on natural language preprocessing and vectorization impact on accuracy of bug report assignment based on real data captured in large software development projects. Experimentation results cover stemming and lemmatization techniques applied for bug description preprocessing and term frequency – inverse document frequency (TF-IDF) parametrization as vectorization method.

We describe a method for symbol spotting in born-digital architectural floor plans, which are a particularly challenging application domain of graphics recognition. We propose a hybrid method that capitalizes on strengths of both vector-based and pixel-based symbol spotting techniques. In the description phase, the salient geometric constituents of a symbol are extracted by a variety of vectorization techniques, including a proposed voting-based algorithm for finding partial ellipses. This enables us to better handle local shape irregularities and boundary discontinuities, as well as partial occlusion and overlap. In the matching phase, the spatial relationship between the geometric primitives is encoded via a primitive-aware proximity graph. A statistical approach is then used to rapidly yield a coarse localization of symbols within the plan. Localization is further refined with a pixel-based step implementing a modified cross-correlation function. Experimental results on the public SESYD synthetic dataset demonstrate that our approach clearly outperforms other popular symbol spotting approaches. Moreover, our approach yields promising results on real-world images, which are significantly more challenging in terms of overlap, occlusion, and presence of non-symbolic graphical data.