Image Fragments Research Articles

НЕЙРОСЕТЕВОЙ ПОДХОД К КОГНИТИВНОМУ МОДЕЛИРОВАНИЮ

Some features of cognitive modeling are presented, including the prerequisites for a cognitive approach to solving complex problems. Cognitive modeling involves the use of various artificial neural networks, including convolutional neural networks. The classification of artificial neural networks according to various characteristics is given. The features of self-organizing neural networks and networks using deep learning methods are considered. The artificial neural network, which is a three-layer unidirectional direct propagation network, the interface of a computer program used to approximate functions using the specified neural network, as well as the solution of the image recognition problem using an artificial convolutional neural network, in which the neural network parameters are adjusted for each recognizable image fragment in order to adaptively filter the image, are considered in detail. The analysis of images in video surveillance systems in order to detect fires allows them to be detected at an early stage and, thus, prevent the fire propogation.

Global-aware Fragment Representation Aggregation Network for image-text retrieval

Image-text retrieval is an important kind of cross-modal retrieval method and has recently attracted much attention. Existing image-text retrieval methods often ignore the relative importance of each fragment (region in an image or word in a sentence) on the global semantic of image or text when aggregating features of image or text fragments, resulting in the ineffectiveness of the learned image and text representations. To address this problem, we propose an image-text retrieval method named Global-aware Fragment Representation Aggregation Network (GFRAN). Specifically, it first designs a fine-grained multimodal information interaction module based on the self-attention mechanism to model both the intra-modality and inter-modality relationships between image regions and words. Then, with the guidance of the global image or text feature, it aggregates image or text fragment features conditioned on their attention weights over global feature, to highlight fragments that contribute more to the overall semantics of images and texts. Extensive experiments on two benchmark datasets Flickr30K and MS-COCO demonstrate the superiority of the proposed GFRAN model over several state-of-the-art baselines.

Development of Artificial Intelligence-Based Programs for the Diagnosis of Myocarditis in COVID-19 Using Chest Computed Tomography Data»

It has been established that 7.2% of patients hospitalized with coronavirus disease (COVID-19) exhibit signs of heart disease, with 23% of these patients experiencing heart failure. Currently, there is a lack of data on chest computed tomography (CT) for diagnosing myocarditis associated with COVID-19. The aim. To justify the feasibility and develop classification models for diagnosing myocarditis in COVID-19 patients based on chest CT data processing. Materials and methods. A retrospective analysis of data from 140 COVID-19 patients was conducted. Chest CT scans were analyzed using DRAGONFLY software, with permission from Object Research Systems. The COVID-CT-MD database, which includes CT data from 169 confirmed cases of SARS-CoV-2 infection, was used to build classification models. The regions of interest were fragments of heart CT images. Texture analysis methods were employed to create diagnostic models. Results. It was shown that the average density of the myocardium of a patient with a confirmed diagnosis of SARS-CoV-2 infection according to the Hounsfield scale does not essentially differ from the densitometric indicators of a healthy person. Therefore, the research was focused on finding structural changes in CT images for their use in constructing diagnostic models. The use of different classification algorithms had little effect on classification accuracy, probably due to the informational content of the input data. However, the obtained accuracy of the diagnostic models is acceptable and allows them to be used to support medical decision-making regarding diagnosis and treatment. Conclusions. Using classic methods, myocarditis was diagnosed in 7.1% of patients with severe pneumonia caused by the coronavirus. The global data closely aligns with the results of our clinical studies. The obtained results allowed for assessing structural changes in the myocardium characteristic of the acute form of SARS-CoV-2 infection. The constructed classification models indicate that specific changes in the myocardium during the acute form of SARS-CoV-2 infection can be identified using CT. The highest diagnostic accuracy on test samples reached 74%. The implementation of the developed diagnostic programs based on texture analysis of CT data and artificial intelligence technologies enables the diagnosis of myocarditis and the assessment of long-term treatment efficiency. Creation of these diagnostic programs using artificial intelligence technologies significantly simplifies the work of radiologists and improves the efficiency of myocarditis diagnosis in patients with SARS-CoV-2 infection.

Devising an edge effect compensation procedure to eliminate structural distortions during frequency filtering

The object of this study is the process of filtering astronomical frames that contain images of potential objects in the Solar System. To recognize the image of each such object in contrast with the background of the frame, it is necessary to carry out frequency filtering of the image. Any frequency filtering using various image filters is aimed at reducing the dynamic range of the background substrate. Also, frequency filtering leads to an increase in the signal-to-noise ratio of the entire image or its fragments, depending on the configuration. However, the identified problem area of each image during frequency filtering is the distortion of the structure of its edges. Therefore, to solve this problem, an edge effect compensation procedure has been proposed to eliminate structural distortions during frequency filtering. Complementing the image with borders on all sides and the augmented extended image made it possible to introduce a formal connection between the pixel values of the extended image fragment and the pixel values of the extended original image. Testing was carried out using a high-pass Gaussian filter. The use of the devised edge effect compensation procedure made it possible to remove distortion of the structure of the image edges. The devised edge effect compensation procedure was tested in practice within the CoLiTec project. It was implemented during the in-frame processing stage of the Lemur software. The study showed that the use of the devised edge effect compensation procedure makes it possible to remove image artifacts compared to conventional filtering without taking into account the edge effect. Also, owing to edge effect compensation, structural image distortions were eliminated, and the signal-to-noise ratio was increased by 7–10 times

Quantitative Edge Analysis Can Differentiate Pancreatic Carcinoma from Normal Pancreatic Parenchyma.

This study aimed to introduce specific image feature analysis, focusing on pancreatic margins, and to provide a quantitative measure of edge irregularity, evidencing correlations with the presence/absence of pancreatic adenocarcinoma. We selected 50 patients (36 men, 14 women; mean age 63.7 years) who underwent Multi-detector computed tomography (MDCT) for the staging of pancreatic adenocarcinoma of the tail of the pancreas. Computer-assisted quantitative edge analysis was performed on the border fragments in MDCT images of neoplastic and healthy glandular parenchyma, from which we obtained the root mean square deviation SD of the actual border from the average boundary line. The SD values relative to healthy and neoplastic borders were compared using a paired t-test. A significant SD difference was observed between healthy and neoplastic borders. A threshold SD value was also found, enabling the differentiation of adenocarcinoma with 96% specificity and sensitivity. We introduced a quantitative measure of boundary irregularity, which correlates with the presence/absence of pancreatic adenocarcinoma. Quantitative edge analysis can be promptly performed on select border fragments in MDCT images, providing a useful supporting tool for diagnostics and a possible starting point for machine learning recognition based on lower-dimensional feature space.

On estimating the size of informative fragments in the sea surface images

The paper proposes a solution to one of the problems arising in the construction of modern traffic safety systems in marine areas, namely, estimating the size of informative fragments in the image, which it seems advisable to use when detecting foreign objects on the sea surface image. It is proposed to estimate the size of informative fragments based on calculating the average distance between the contours of the wave elements visible in the image, such as their crests, depressions, etc. The contours of these wave elements are determined based on the Canny operator. The estimation of the sizes of informative fragments is performed along the columns and rows of the analyzed image. Computational experiments have been carried out to illustrate the developed algorithm efficiency. The obtained estimates of the sizes of informative fragments of sea surface images seem appropriate to use in their analysis, in particular, when solving problems of detecting foreign objects on sea surface images.

Restoration of Semantic-Based Super-Resolution Aerial Images

Currently, technologies for remote sensing image processing are actively developing, including both satellite images and aerial images obtained from video cameras of unmanned aerial vehicles. Often such images have artifacts such as low resolution, blurred image fragments, noise, etc. One way to overcome such limitations is to use modern technologies to restore super-resolution images based on deep learning methods. The specificity of aerial images is the presentation of texture and structural elements in a higher resolution than in satellite images, which objectively contributes to better results of restoration. The article provides a classification of super-resolution methods based on the main architectures of deep neural networks, namely convolutional neural networks, visual transformers and generative adversarial networks. The article proposes a method for reconstructing super-resolution aerial images SemESRGAN taking into account semantic features by using an additional deep network for semantic segmentation during the training stage. The total loss function, including adversarial losses, pixel-level losses, and perception losses (feature similarity), is minimized. Six annotated aerial and satellite image datasets CLCD, DOTA, LEVIR-CD, UAVid, AAD, and AID were used for the experiments. The results of image restoration using the proposed SemESRGAN method were compared with the basic architectures of convolutional neural networks, visual transformers and generative adversarial networks. Comparative results of image restoration were obtained using objective metrics PSNR and SSIM, which made it possible to evaluate the quality of restoration using various deep network models.

Convergence region size research of television signal superposition iteration method for measurement machine vision system

Background. Image processing is used in machine vision measurement systems, which are widely used in industrial applications. One of the important tasks in processing is image superposition. Registration methods must fulfill two basic requirements: satisfactory registration quality and practical processing time. One of the methods that meets these requirements is iterative superposition. Aim. The aim is to study the size of the area (region of convergence) for the values of the parameters of the initial approximation, which ensures satisfactory superposition. The size of the region determines the performance of the iterative algorithm. Methods. The size of the region of convergence was determined experimentally: correlation analysis and statistical simulation. Results. Experiments have shown if a sampling step is no more than 1/10 of the size of the image fragment, then the probability of correct registration is almost equal to one. If this threshold is exceeded, then the probability decreases. Conclusion. The size of the region of convergence of the iterative image registration method has been established. It is possible to design processing algorithms and predict processing time by obtaining results. The superposition method was introduced into a machine vision system to recognize contact wire clamps on the railway. Image superposition in this system is carried out in real time.

Explainable Histopathology Image Classification with Self-organizing Maps: A Granular Computing Perspective

The automatic analysis of histology images is an open research field where machine learning techniques and neural networks, especially deep architectures, are considered successful tools due to their abilities in image classification. This paper proposes a granular computing methodology for histopathological image classification. It is based on embedding tiles of histopathology images using deep metric learning, where a self-organizing map is adopted to generate the granular structure in this learned embedding space. The SOM enables the implementation of an explainable mechanism by visualizing a knowledge space that the experts can use to analyze and classify the new images. Additionally, it provides confidence in the classification results while highlighting each important image fragment, with the benefit of reducing the number of false negatives. An exemplary case is when an image detail is indicated, with small confidence, as malignant in an image globally classified as benign. Another implemented feature is the proposal of additional labelled image tiles sharing the same characteristics to specify the context of the output decision. The proposed system was tested using three histopathology image datasets, obtaining the accuracy of the state-of-the-art black-box methods based on deep learning neural networks. Differently from the methodologies proposed so far for the same purpose, this paper introduces a novel explainable method for medical image analysis where the advantages of the deep learning neural networks used to build the embedding space for the image tiles are combined with the intrinsic explainability of the granular process obtained using the clustering property of a self-organizing map.

Re-visioning and critically emphasizing the need for psychosocial intervention in addressing body image distress in head and neck cancer patients: A review of literature

Head and neck cancers (HNCs) patients undergo extensive debilitating surgeries on the part of their body which is considered as a unique characteristic feature and a central entity to communication and perception, instilling a sense of self: The face. Hereby, alterations and discontinuity in this sense of self lead to fragmentation of body image, thereby shattering the sense of identity and making it difficult for patients to recognize selves and return back to the precancerous identity. Body image dissatisfaction is associated with varied psychosocial adverse outcomes such as anxiety, depression, decreased quality of life affecting social dynamics, romantic relationship distress relating to sexual and intimate health, and excessive efforts spent on appearance-fixing behaviors and the need for constant reassurance. As body image goes way beyond just mere appreciation of self and is a multifaceted, intricate web crucially affecting an individual’s identity and existence, it is of utmost importance that medical health professionals aim at making oncologic treatment a more holistic model which considers a patient’s mental, emotional and physical well-being. Present literature offers a wide scope for studies focused on psychological distress in breast cancer patients; however, limited empirical research has been conducted on body image concerns and the need for psychosocial interventions in HNC patients. A need for an in-depth understanding into this aspect of psycho-oncology will help oncologic intervention to be more patient-centric, providing the benefits of both biomedical and psychosocial skills.

Image fusion algorithm for the collaborative use of unmanned aerial vehicles

Currently, the applicable scope of unmanned aircraft application is increasingly expanding. The promising field of unmanned aircraft enhancement is the implementation of some collaborative actions during controlled flight. This paper considers some issues of the group application of unmanned aerial vehicles (UAV) related to the coordinated planning and control of UAVs performing surveillance missions. Performing aerial search operations is technically complicated by the requirement to recognize a search object in arbitrary conditions, which can be both simple and severe environment. The search area is limited by the UAV capabilities, so, in order to improve the efficiency of search operations, UAVs are combined into groups. An algorithm for solving the problem of object search in arbitrary conditions by a group of unmanned aircraft is proposed. The advantage of search by a group of unmanned aircraft is the coverage of the larger search area in a conventional unit of time. This paper addresses the unmanned aircraft configuration, containing both the means of collaborative flight operation and a synthetic vision system. The image obtained by the synthetic vision system is both a source of navigation information and a means which reliably determines the result of search operations. Depending on the conditions of search operations, the image obtained by the synthetic vision system may require additional processing to use as intended. A fusion algorithm is proposed, which is characterized by adaptive adjustment of parameters in each frame individually for different image fragments. Based on the results obtained, it is planned to create a new product for commercial operation of unmanned aircraft.

TENSOR AND VECTOR APPROACHES TO OBJECTS RECOGNITION BY INVERSE FEATURE FILTERS

The investigation of the extraction of image objects features by filters based on tensor and vector data presentation is considered. The tensor data is obtained as a sum of rank-one tensors, given by the tensor product of the vector of lexicographic representation of image fragments pixels with itself. The accumulated tensor is approximated by one rank tensor obtained using singular values decomposition. It has been shown that the main vector of the decomposition can be considered as the object feature vector. The vector data is obtained by accumulating analogous vectors of image fragments pixels. The accumulated vector is also considered as an object feature. The filter banks of a set of objects are obtained by regularized inversion of the matrices compiled by object features vectors. Optimized regularization of the inversion is used to expand the regions of object features capture with minimal error. The object fragments and corresponding feature vectors are selected through a training iterative process. The tensor and vector approaches create two channels for recognition. High efficiency of object recognition can be achieved by choosing the filter capture band and creating filter branches according to the given bands. The filters create a convolutional network to recognize a set of objects. It has been shown that the obtained filters have an advantage over known correlation filters when recognizing objects with small fragments.

Definition of spectrum by colors metamerism using images of plants by ordinary camera

In several cases, studying biological objects requires spectral analysis to conduct research. Determination of a spectrum from RGB image data is a highly complex task because image color is formed by mixing colors. The inverse transformation of RGB color-to-spectrum does not have a unique solution. Consequently, this paper proposes an approach based on color deconvolution usage. RGB values are determined and set for each wavelength in the spectrum. Obtained values establish and settle the direction of the axis in the color space closest to the color of the respective wavelength, that is defined from color metamerism in RGB space. Accordingly, a new base of vectors is constructed along which deconvolution is performed. This procedure allows to define color components from the image fragment as a bin of a specific spectrum wavelength. The intensity of bins in the spectrum is found through the distribution of mean intensity for such images. The proposed approach makes it possible to obtain satisfactory results to determine the color of plants for monitoring tasks. Additionally, this method can also be employed as initial data to train a neural network for scrutinizing plant diseases in remote monitoring methods.

A fragment-based approach for computing the long-term visual evolution of historical maps

Cartography, as a strategic technology, is a historical marker. Maps are tightly connected to the cultural construction of the environment. The increasing availability of digital collections of historical map images provides an unprecedented opportunity to study large map corpora. Corpus linguistics has led to significant advances in understanding how languages change. Research on large map corpora could in turn significantly contribute to understanding cultural and historical changes. We develop a methodology for cartographic stylometry, with an approach inspired by structuralist linguistics, considering maps as visual language systems. As a case study, we focus on a corpus of 10,000 French and Swiss maps, published between 1600 and 1950. Our method is based on the fragmentation of the map image into elementary map units. A fully interpretable feature representation of these units is computed by contrasting maps from different, coherent cartographic series, based on a set of candidate visual features (texture, morphology, graphical load). The resulting representation effectively distinguishes between map series, enabling the elementary units to be grouped into types, whose distribution can be examined over 350 years. The results show that the analyzed maps underwent a steady abstraction process during the 17th and 18th centuries. The 19th century brought a lasting scission between small- and large-scale maps. Macroscopic trends are also highlighted, such as a surge in the production of fine lines, and an increase in map load, that reveal cultural fashion processes and shifts in mapping practices. This initial research demonstrates how cartographic stylometry can be used for exploratory research on visual languages and cultural evolution in large map corpora, opening an effective dialogue with the history of cartography. It also deepens the understanding of cartography by revealing macroscopic phenomena over the long term.

Application of a Three-Stage Procedure for Extracting a Small-Sized Anchor Object on a Noisy Image

The purpose of the article is to develop a three-stage procedure for identifying an anchor object in a noisy current image. This goal is achieved by determining the sampling threshold levels at which the greatest similarity of the compared images is ensured; development of a procedure for refining the maximum of the decision function by quantizing the current image. The solution to the first problem is based on the formation of a correlation field of radio brightness temperatures and the choice of a sampling threshold. It is proposed to use the cross-correlation coefficient as a criterion for the degree of image matching. The effectiveness of the procedure for selecting a fragment of a reference image is assessed based on the criterion of the probability of selecting a fragment of a reference image. It is shown that image noise can lead to a decrease in the probability of selecting a fragment of a reference image, down to 0.4. It is proposed to refine the maximum of the decision function based on the iteration method. The most significant results are the obtained dependences of the probability of choosing a fragment of the reference image on the threshold value and the signal-to-noise ratio, as well as the analytical relation for the asymptote of this dependence. The novelty of the work lies in the fact that the procedure for forming the decisive function has been further developed. This will significantly improve the operating efficiency of unmanned aerial vehicles, especially in conditions of interference.

Improvement of Machine Vision Video Signal Processing Algorithm for Higher Accuracy in Extended Object Speed Measurements

Machine vision systems are widely used for monitoring of the railway infrastructure condition. High reliability of machine vision systems allows using them not only for video recording but also for automation of technological processes. One of the important tasks in the process automation is measurement of an extended object speed (rail-road cars, electric locomotives and rolling stock). Some systems have a requirement to estimate speed in real time (braking of cars at a shunting yard). In this case the speed is calculated by two adjacent frames to ensure the minimum delay for measurement. However, the image of an extended object can be uninformative (image fragments are uniform in brightness), the weather can cause interference. It leads to high speed estimation error. This article describes the improvement for the existing real-time algorithm which allows for higher accuracy in speed measurements. The improvement implies selection of an informative image area corresponding to an extended object. A comparative analysis of the existing and improved algorithm showed a significant reduction in the estimation error in speed measurements.

An automatic segmentation algorithm for conglutinated bone fragments in 3D CT images of lower limb fractures based on morphology

An automatic segmentation algorithm for conglutinated bone fragments in 3D CT images of lower limb fractures based on morphology

Ideographic Dictionaries as desiderata of Author Lexicography

This article presents an analytical review of dictionaries created within the framework of ideographic author lexicography as a particular direction of general author lexicography and the most productive in terms of modelling artistic images of the world. An analysis of the scholarly literature dealing with the issues of this direction is also proposed. The aim of the study is to identify the typological features of dictionaries of the selected variety, to characterise the principles of compiling such dictionaries and to identify the approaches proposed in dictionary projects. The main method of the study is dictionary criticism, which includes the description of dictionaries, their analysis and scholarly evaluation. The interest in the ideographic form of the dictionary representation of the author’s language was born in the Russian lexicography at the turn of the twenty-first century. At present, the number of such dictionaries is small in relation to the demand and desirability of their use, which is explained by the general commitment of lexicographers to the traditional alphabetic form of the description of the author’s language. The analysis of the available dictionaries makes it possible to divide them, taking into account the different typological characteristics, mainly into monographic and summary dictionaries (representing the language of one or more authors). The dictionaries of the first group demonstrate the possibility of modelling fragments of the world view of a single author with the inclusion of different lexical arrangements: the entire vocabulary of the French poems of F. Tyutchev; occasional units of M. Saltykov-Shchedrin; figurative nominations in the poetry of A. Gorodnitsky. This makes it possible to predict the expansion of the composition of objects of interpretation in “one-author” ideographic dictionaries. At the same time, the form of a summary dictionary is relevant in the description of figurative means for recreating fragments of a general artistic image of the world. The article emphasises that in order to assess the degree of expression of the ideographic base in author lexicography, it is important to take into account dictionaries of other types that contain an ideographic component (for example, in the form of thematic word lists as part of frequency dictionaries of the writer’s language). The observations made during the analysis of scholarly publications on the subject of the study indicate possible vectors of the development of ideographic author lexicography. One of them relates to the use of the method of poetic ideography developed by scholars of the Ural Semantic School.

Research on Digital Image Record in Intangible Cultural Heritage Inheritance Innovation under the Background of Big Data

Abstract The purpose of this paper is to investigate the digital lens presentation effect and image dissemination effect of intangible cultural heritage. In order to test the lens boundary detection technology of ICH digital images, we focus on analyzing the number of image frames, the number of mutations and the number of gradual changes of ICH characters, space-time and events, and exploring the checking completeness, the checking accuracy and the comprehensive detection rate of this paper’s algorithm and the dual-threshold comparative method on different types of ICH image fragments. Combined with the descriptive analysis of valid samples, we explore the main types of transmission of NRM digital images and clarify the degree of media utilization and the ranking of transmission power of NRM digital images in the Yangtze River Delta region. Shanghai has the strongest online dissemination power of traditional theater, traditional fine arts and traditional arts and crafts, with 1284586, 310468, and 237818, respectively, while traditional music, Chinese opera and folklore have the weakest dissemination power in Shanghai. Traditional music makes up 18.73% of them. To ensure non-heritage culture’s inheritance, it is necessary to strengthen the use of digital imaging technology to achieve linkage communication between non-heritage cultures and build an overall communication system.

Репрезентація концепту „подорож” у паремійному словнику „Галицько-руські народні приповідки”, укладеному І. Франком

The article considers the peculiarities of the verbalization of the concept of travel, recorded in the paremia image of the world of one of the Ukrainian subcultures — the Galicians, interpreted and recorded by Ivan Franko in the dictionary Halytsko-ruski narodni prypovidky [Galician-Russian folktales]. The purpose of the study is to analyze the fragment of the paremic image of the world inherent in the Ukrainian linguocultura of Galicians, with the representation of the paremic verbalization of the concept of travel, recorded in the national consciousness of Galicians and interpreted by Ivan Franko in his unique collection of paremias, taking into account the theory of linguoculturology and linguoconceptology. The following primary research methods were used: the linguistic observation method, the continuous sampling method, the descriptive method, the comparative method, conceptual analysis, the interpretive method, the method of interpreting dictionary definitions, the technique of abstraction. The object of the study was the unique dictionaryof paremias Halytsko-ruski narodni prypovidky.