{"title":"Generation and Study of the Synthetic Brain Electron Microscopy Dataset for Segmentation Purpose","authors":"N. Sokolov, E. Vasiliev, A. Getmanskaya","doi":"10.20948/graphicon-2022-706-714","DOIUrl":"https://doi.org/10.20948/graphicon-2022-706-714","url":null,"abstract":"Advanced microscopy technologies such as electron microscopy have opened up a new field of vision for biomedical researchers. The use of artificial intelligence methods for processing EM data is largely difficult due to the small amount of annotated data at the training stage. Therefore, we add synthetic images to an annotated real EM dataset or use a fully synthetic training dataset. In this work, we present an algorithm for the synthesis of 6 types of organelles. Based on the EPFL dataset, a training set of 860 real fragments 256x256 (ORG) and 6000 synthetic ones (SYN), as well as their combination (MIX), were generated. An experiment of training models for segmentation into 5 and 6 classes showed that, despite the imperfection of synthetic data, for an axon poorly represented in the training data set, the use of a synthetic data set improves the Dice metric from 0.3 on the original dataset to 0.8 on the mixed and synthetic datasets. The synthetic data strategy gives annotations for free, but shifts the effort to producing sufficiently realistic images.","PeriodicalId":299055,"journal":{"name":"Proceedings of the 32nd International Conference on Computer Graphics and Vision","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125619926","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Usage of Scenario Based Modeling Automation in Optics CAD Systems","authors":"M. Kopylov","doi":"10.20948/graphicon-2022-55-63","DOIUrl":"https://doi.org/10.20948/graphicon-2022-55-63","url":null,"abstract":"This article describes approaches to the use of scenarios in the automation of applied tasks within the framework of working with a software complex of optical modeling and photorealistic computer graphics. Optical modeling complex should include support for scripting, since conventional tools, such as those provided using a graphical user interface, are usually not enough to perform a variety of tasks that arise in practice. A brief overview of existing solutions is made. Methods are proposed that ensure effective bringing of scenario support into an existing optical modeling system. These methods provide full access from scripts to system modules, including computational modules, various simulator modules, etc. Both trivial scripts containing ordinary command sequences and more complex ones using the capabilities of high-level scripting languages such as Python are supported. Separately, a special high-level API is considered which is used for the interaction between of scripts written in the Python language and the optical modeling system. This API can also be used to extend the capabilities of the system with new parametric objects. The specialized software modules integrated into the optical modeling system, such as the batch interpreter, script interpreter and editor, extension class editor, are considered in detail. Examples of scenario-based modeling automation are given.","PeriodicalId":299055,"journal":{"name":"Proceedings of the 32nd International Conference on Computer Graphics and Vision","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126153810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Nikita Djeffrievich Lockshin, A. Khvostikov, A. Krylov
{"title":"Augmenting Histological Images with Adversarial Attacks","authors":"Nikita Djeffrievich Lockshin, A. Khvostikov, A. Krylov","doi":"10.20948/graphicon-2022-637-647","DOIUrl":"https://doi.org/10.20948/graphicon-2022-637-647","url":null,"abstract":"Neural networks have shown to be vulnerable against adversarial attacks - images with carefully crafted adversarial noise that is imperceptible to the human eye. In medical imaging tasks this can be a major threat for making predictions based on deep neural network solutions. In this paper we propose a pipeline for augmenting a small histological image dataset using State-of-the-Art data generation methods and demonstrate an increase in accuracy of a neural classifier trained on the augmented dataset when faced with adversarial images. When trained on the non-augmented dataset, the neural network achieves an accuracy of 55.24 on the test set with added adversarial noise, and an accuracy of 97.40 on the same test set when trained on the augmented dataset.","PeriodicalId":299055,"journal":{"name":"Proceedings of the 32nd International Conference on Computer Graphics and Vision","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126814159","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Competency-based Approach in Expert-heuristic Evaluation of Graphical User Interfaces","authors":"U. Khaleeva","doi":"10.20948/graphicon-2022-1175-1185","DOIUrl":"https://doi.org/10.20948/graphicon-2022-1175-1185","url":null,"abstract":"This paper is a continuation of a cycle of research aimed at forming a new method for evaluating graphical user interfaces (GUI) using an expert-heuristic method. A combination of expert and heuristic approach is proposed, to detect a wide range of UI/UX problems, ensure the competence of the evaluation and reduce the level of incredulity to the expert. This paper illustrates the possibility of implementing a competency-based approach to form heuristic criteria. The key idea behind the application of such a method, the principle of scoring and the experiments conducted to confirm this hypothesis are described in detail. A practical implementation of the competency-based assessment technique for expert-heuristic assessment of UI/UX is presented. The analysis of the experiment is performed, the key advantages and disadvantages of the considered approach are revealed. Conclusions are made about the applicability of the method for evaluating interfaces of different types.","PeriodicalId":299055,"journal":{"name":"Proceedings of the 32nd International Conference on Computer Graphics and Vision","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123524910","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Noise Level Estimation in Images Using Haar Wavelets","authors":"A. Pronkin","doi":"10.20948/graphicon-2022-442-448","DOIUrl":"https://doi.org/10.20948/graphicon-2022-442-448","url":null,"abstract":"The paper investigates the possibility and expediency of using the Haar wavelet transform in the problem of estimating the level of discrete Gaussian noise in an image. An algorithm is proposed that uses Haar wavelets to obtain an estimate of the variance of discrete Gaussian noise in a digital image. To reduce the influence of image fragments with a large proportion of high-frequency oscillations of the useful signal, the image is divided into blocks, followed by the selection of blocks with a minimum dispersion. The proposed algorithm is compared with a method based on the use of difference operators for estimating the noise level. This method gives fairly accurate noise variance estimates and has low computational complexity. The results of estimating the variance of the noise of different intensity superimposed on the image by compared methods are presented. Based on the theoretical provisions and the results of experimental studies, it is concluded that the proposed algorithm has the best accuracy in estimating the noise level at lower computational costs.","PeriodicalId":299055,"journal":{"name":"Proceedings of the 32nd International Conference on Computer Graphics and Vision","volume":"509 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127602962","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Andreev, A. Bondarev, N. A. Bondareva, A.V. Bondarenko
{"title":"Applying Stereo Animations to Autostereoscopic Monitors in Various Fields of Research","authors":"S. Andreev, A. Bondarev, N. A. Bondareva, A.V. Bondarenko","doi":"10.20948/graphicon-2022-431-440","DOIUrl":"https://doi.org/10.20948/graphicon-2022-431-440","url":null,"abstract":"The construction of stereo images on autostereoscopic monitors, which do not require the viewer to use special glasses, is becoming more common in various fields of research. One of the important tasks is the construction of complex stereo images that combine the main image and the necessary accompanying elements in one frame. In this direction, a large number of studies have been carried out to organize the automatic construction of complex stereo images. This report considers the possibilities of implementing such a construction in various areas of human activity.","PeriodicalId":299055,"journal":{"name":"Proceedings of the 32nd International Conference on Computer Graphics and Vision","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126259126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Visualization Subsystem of the Software Package for Searching People and Vehicles Using Unmanned Aerial Vehicles","authors":"D. Bulatitskiy, A. Selifontov","doi":"10.20948/graphicon-2022-1102-1114","DOIUrl":"https://doi.org/10.20948/graphicon-2022-1102-1114","url":null,"abstract":"Unmanned aerial vehicles (UAVs) can be used to improve the efficiency of searching for missing people and vehicles in the natural environment. However, the use of UAVs alone may not be enough. To get a tangible effect, the authors of the article are developing a software package that includes a neural network core for processing images from UAVs, a web interface for the coordinator of the search operation and a mobile application for members of search groups working on the ground. The article substantiates the importance of the visualization subsystem for the software package being developed. The data output for the coordinator of the search operation and the participants of the search groups working on the ground are systematized. The following types of data are highlighted: information about the search area and the search targets; information about the image sets from flights and about image analysis; information about the participants of the search operation; information about the progress of the search operation; data about the search operation for a member of the search group; data on the results of search operations. Approaches to visualization of the listed data are described, interface elements for their visualization are selected and developed. Improvements to the software package are proposed and implemented considering the requirements for visualization.","PeriodicalId":299055,"journal":{"name":"Proceedings of the 32nd International Conference on Computer Graphics and Vision","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126381872","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Iuliia Tarasova, V. V. Andreev, A. Chechin, Denis V. Toskin
{"title":"Providing Decision Support in Career Guidance Through the use of Machine Vision","authors":"Iuliia Tarasova, V. V. Andreev, A. Chechin, Denis V. Toskin","doi":"10.20948/graphicon-2022-1027-1038","DOIUrl":"https://doi.org/10.20948/graphicon-2022-1027-1038","url":null,"abstract":"This paper is devoted to the review of the results of the development and implementation of the ColorUnique Pro career guidance software package in the career guidance activities of higher educational institutions. The review begins with the initial formulation of the problem, justification of the use of neural networks as the basis of one of the classifiers, consideration of the results of experiments and then – the introduction of a software package. At the end of the article, prospects for further research are described, such as the creation of a three-dimensional map of types and subtypes and the further identification of new subtypes. The possibility of using additional methods of analysis, detection and classification is also being considered in order to study the influence on the definition of ISA of such features of the images obtained as «demonstrative» and «true» structures, as well as the presence of a «background» that does not contain characteristic elements. In addition to neural networks, the authors also used the «sliding» window image processing method, as a result, the software package includes two classifiers that analyze images separately, however, in the future, the results of the analysis of both classifiers are compared by an expert, since some subtypes can only be determined by joint interpretation.","PeriodicalId":299055,"journal":{"name":"Proceedings of the 32nd International Conference on Computer Graphics and Vision","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130024779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Optical-information Method of Detection of Unmanned Aircraft by a Robotic Optical-electronic System","authors":"V. Smolin, I. Yakimenko, D.S. Rasskaza","doi":"10.20948/graphicon-2022-548-558","DOIUrl":"https://doi.org/10.20948/graphicon-2022-548-558","url":null,"abstract":"The informativeness of the operating spectral ranges of optoelectronic detection systems for unmanned aircraft is considered. The conducted studies have shown that the spectral range of 3 – 5 microns, due to the absence of powerful radiation sources in unmanned aircraft, has significantly lower radiation power compared to the spectral range of 8 – 13 microns. As an alternative to the range of 3 – 5 microns, a range of 1.2 - 1.7 microns is proposed. The proposed near–infrared range can complement the use of a range of 8 - 13 microns in daytime conditions, which, along with the development of new detection methods based on the background principle, will increase the detection efficiency of unmanned aircraft. A contrast method has been developed based on the background detection principle. A distinctive feature of the contrast method is the splitting of the frame into segments. The segment size corresponds to the size of the inhomogeneities of the atmosphere. A contrast algorithm for detecting unmanned aircraft is considered. The probability of detecting small-sized objects was calculated using the developed method. The developed contrast method is compared with existing methods. The advantage of the developed contrast method in comparison with the investigated ones is revealed. The detection range of unmanned aircraft has increased by 10%.","PeriodicalId":299055,"journal":{"name":"Proceedings of the 32nd International Conference on Computer Graphics and Vision","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131122072","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The Problems of Analyzing the Dependence of the Sizes of Structural Components of Samples Synthesized by the 3DMP Method on the Strategy of Wave Thermal Deformation Hardening","authors":"A. Kirichek, S. Fedonina","doi":"10.20948/graphicon-2022-1020-1026","DOIUrl":"https://doi.org/10.20948/graphicon-2022-1020-1026","url":null,"abstract":"The problems of analysis of metallographic images and the way to solve them using modern software for the analysis of metallographic images are described. In this paper, we consider the problem of analyzing the microstructure of parts obtained as a result of alternating surfacing operations by the 3DMP method and wave thermal deformation hardening (HTSH), based on the expansion of the technological capabilities of the previously known wave work hardening (WSH). The analysis of the characteristic dimensions of the structural components of the synthesized samples, hardened according to different strategies of HTFS, is carried out in the SIAMS software package in the \"Multi-phase analysis\" section. A brief explanation of the features of using the software for a particular case is given. As a result of metallographic studies using the SIAMS software of the microstructure of parts obtained as a result of alternating surfacing operations by the 3DMP method and HTDU, the characteristic sizes of particle colonies and fine particles of crushed dendrites in each deposited layer were determined, and the nature of the distribution of structural components over the cross section of the samples in the direction of surfacing was revealed.","PeriodicalId":299055,"journal":{"name":"Proceedings of the 32nd International Conference on Computer Graphics and Vision","volume":"28 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132869551","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}