Andrii Podorozhniak | National Technical University "Kharkiv Polytechnic Institute" (original) (raw)

Papers by Andrii Podorozhniak

Advanced Information Systems, 2024

The subject of study is analysis of performance of artificial intelligence systems with different... more The subject of study is analysis of performance of artificial intelligence systems with different architectures for reforestation detection. The goal is to implement, train and evaluate system with different models for deforestation and reforestation detection. The tasks are to study problems and potential solutions in forestry for reforestation detection and present own solution. As part of model comparison, results are presented for different artificial neural network architectures with different encoders. For training and testing purpose custom dataset was created, which includes different areas of territory of Ukraine within different timestamps. Main research methods are literature analysis, experiment and case study. As a result of analysis of modern artificial intelligence methods, machine learning, deep learning and convolutional neural networks, high-precision algorithms U-Net and LinkNet were chosen for system implementation. Conclusions. The studied problem was stated formally and broken down in smaller steps; possible solutions were studied and proposed solution was described in details. Necessary mathematical background for analysis of the performance was provided. As part of the development, accurate deforestation/reforestation module was created. All analysis results were listed and a comparison of the studied algorithms was presented.

Neural networks are intensively developed and used in all spheres of human activity in the modern... more Neural networks are intensively developed and used in all spheres of human activity in the modern world. Their use to determine the fire hazardous forest areas can begin to solve the problem of preventing wildfires. In recent years, wildfires have acquired enormous proportions. Wildfires are difficult to control and, if they occur, require a large amount of resources to eliminate them. The paper is devoted to solve the problem of identifying fire hazardous forest areas. The Camp Fire (California, USA) areas are considered. The purpose of the paper is to research the possibility of using convolutional neural networks for the detection fire hazardous forest areas using multispectral images obtained from Landsat 8. The tasks of research are finding the territories where the largest fires occurred in recent time; analyzing economic and ecologic losses from wildfires; receiving and processing multispectral images of wildfire areas from satellite Landsat 8; calculation of spectral indices (NDVI, NDWI, PSRI); developing convolutional neural network and analyzing results. The object of the research is the process of detecting fire hazardous forest areas using convolutional neural network. The subject of the research is the process of recognition multispectral images using deep learning neural network. The scientific novelty of the research is the recognition method of multispectral images by using convolutional neural network has been improved. The theory of deep learning neural networks, the theory of recognition multispectral images and mathematical statistics methods are used. The spectral indices for allocating the object under research (green vegetation, humidity, dry carbon) were calculated. It is obtained that the classification accuracy for a convolutional neural network on the test data is 94.27%. K e ywor d s : deep learning; convolutional neural networks; multispectral images; spectral indices; fire hazardous forest areas.

Advanced Information Systems

In the current era, numerous social networks and messaging platforms have become integral parts o... more In the current era, numerous social networks and messaging platforms have become integral parts of our lives, particularly in relation to work activities, due to the prevailing COVID-19 pandemic and russian war in Ukraine. Amidst this backdrop, the issue of spam and spammers has become more pertinent than ever, with a continuous rise in the incidence of spam within work-related text streams. Spam refers to textual content that is extraneous to a specific text stream, while a spammer denotes an individual who disseminates unsolicited messages for personal gain. The proposed article is devoted to address this scientific and practical challenge of identifying spammers and detecting spam messages within the textual context of any social network or messenger. This endeavor encompasses the utilization of diverse spam detection algorithms and approaches for spammer identification. Four algorithms were implemented, namely a naive Bayesian classifier, Support-vector machine, multilayer perce...

Automation of technological and business processes

Системи управління, навігації та зв’язку. Збірник наукових праць

Національний технічний університет "Харківський політехнічний університет", Харків, Україна НЕЙРО... more Національний технічний університет "Харківський політехнічний університет", Харків, Україна НЕЙРОМЕРЕЖЕВА СИСТЕМА РОЗПІЗНАВАННЯ АВТОНОМЕРА Ан от а ц і я. Предметом дослідження є нейромережева система ідентифікації автомобільних номерів на зображеннях, отриманих за допомогою відеорегіструючих засобів. Метою роботи є забезпечення процесу розпізнавання номерних знаків транспортних засобів в широких межах зміни кутів спостереження і рівнів освітленості. Завдання-дослідження нейромережевої системи розпізнавання автономерів на зображеннях, отриманих за допомогою засобів відеофіксації в широких межах зміни кутів спостереження і рівнів освітленості. Аналіз проблем методів та алгоритмів автоматизованого розпізнавання номерів автомобілів показав, що найбільш перспективно використовувати нейромережеві алгоритми, які підлаштовуються до зміни умов спостереження засобів контролю дорожнього руху. Рішення завдання розпізнавання автомобільних номерів можна представити у вигляді ряду підзадач, що включають в себе первинну обробку зображення, виявлення області номера на зображенні, сегментацію символів і розпізнавання символів. Висновки: запропонована нейромережева система розпізнавання автономера, що дозволяє здійснювати пошук текстових областей під довільним кутом в різних умовах освітленості. Система дозволяє забезпечити розпізнавання автомобільних номерів в широких межах зміни відстані до автомобіля, кутів спостереження і рівнів освітленості. К лю чов і с лов а : вимірювальна система, температурні вимірювання, мікропроцесорний вимірювач, обробка даних.

Sučasnì ìnformacìjnì sistemi, Dec 23, 2019

Among all types of cancer, breast cancer is the most common. In 2017 breast cancer was the fourth... more Among all types of cancer, breast cancer is the most common. In 2017 breast cancer was the fourth rate for death reasons in Ukraine. The paper is devoted to the automatization of histopathological analysis, which can improve the process of cancer stage diagnosis. The purpose of the paper is to research the ability to use convolutional neural networks for classifying biopsy images for cancer diagnosis. The tasks of research are: analyzing cancer statistics in Europe and Ukraine; analyzing usage of Machine Learning in cancer prognosis and diagnosis tasks; preprocessing of BreCaHAD dataset images; developing a convolutional neural network and analyzing results; the building of heatmap. The object of the research is the process of detecting tumors in microscopic biopsy images using Convolutional Neural Network. The subject of the research is the process of classifying healthy and cancerous cells using deep learning neural networks. The scientific novelty of the research is using ConvNet trained on the BreCaHAD dataset for histopathological analysis. The theory of deep learning neural networks and mathematical statistics methods are used. In result it is obtained that the classification accuracy for a convolutional neural network on the test data is 0.935, ConvNet was effectively used for heatmap building. K e ywor d s : deep learning; convolutional neural networks; breast cancer; histopathological analysis; biopsy images; BreCaHAD.

Advanced Information Systems, 2023

Automation of technological and business – processes, 2023

Automatic license plate recognition (ALPR) systems can be found in many different applications. I... more Automatic license plate recognition (ALPR) systems can be found in many different applications. If a person has a driving license, the person probably had already seen smart road camera after a speed limit sign or on a crossroad. Every year number of cars on roads growth very fast. It is also obvious that such systems can be used out-of-road situations. For instance, this type of systems can be used for automatic access control on private property or smart parking, or even log system that are being used literally everywhere. Because of popularity of ALPR systems, there are two main goals, which are being pursued by researches: speed and accuracy of recognition. Speed of the detection is important for real-time systems. Accuracy is important for every system. The more accurate a system is, the more reliable it is. For example, car accident detection systems should be as accurate as possible in order to be used, because no one wants to get billed with the wrongdoing, that wasn't committed by them. The purpose of the study is to develop high precision automatic license plate detection system with number extraction possibilities. In order to achieve the goal many different modern solutions and technologies were studied and solution is presented. The main technology of the project is artificial intelligence system and, more specifically, convolutional neural network. As the main algorithm Mask R-CNN is used for license plate detection. To present reasonable research, the system was tested on different hardware (CPU, GPU, Raspberry PI 4) and on different datasets.
Анотація. Системи автоматичного розпізнавання номерних знаків (АРНЗ) можна знайти в різних сферах людської діяльності. Якщо у людини є водійське посвідчення, то, напевно, вона вже бачила розумну дорожню камеру за знаком обмеження швидкості або на перехресті. З кожним роком кількість автомобілів на дорогах дуже стрімко зростає. Також очевидно, що такі системи можна використовувати поза межами доріг. Наприклад, цей тип систем можна використовувати для автоматичного контролю доступу на приватну власність або розумну парковку, або навіть для систем ведення обліку, які використовується буквально всюди. Через популярність систем АРНЗ існує дві основні цілі, які переслідують дослідження: швидкість і точність розпізнавання. Швидкість виявлення важлива для систем реального часу. Точність важлива для кожної системи. Чим точніша система, тим вона надійніша. Наприклад, системи виявлення автомобільних аварій мають бути максимально точними, щоб їх можна було використовувати, оскільки ніхто не хоче отримати рахунок за порушення, скоєні не ним. Метою дослідження є розробка високоточної автоматичної системи виявлення номерних знаків з можливістю вилучення номера. Для досягнення мети було вивчено багато різних сучасних рішень і технологій, а також представлено рішення проблеми. Основною технологією проекту є система штучного інтелекту, а точніше-згорткова нейронна мережа. В якості основного алгоритму для виявлення номерних знаків використовується Mask R-CNN. Щоб представити обґрунтоване дослідження, систему протестували в різних середовищах (CPU, GPU, Raspberry PI 4) і на різних наборах даних.

Journal of information and organizational sciences

Deep learning approaches are applied for a wide variety of problems, they are being used in the r... more Deep learning approaches are applied for a wide variety of problems, they are being used in the remote sensing field of study and showed high performance. Recent studies have demonstrated the efficiency of using spectral indexes in classification problems, because of accuracy and F1 score increasing in comparison with the usage of only RGB channels. The paper studies the problem of classification satellite images on the EuroSAT dataset using the proposed convolutional neural network. In the research set of the most used spectral indexes have been selected and calculated on the EuroSAT dataset. Then, a novel comparative analysis of spectral indexes was carried out. It has been established that the most significant set of indexes (NDVI, NDWI, GNDVI) increased classification accuracy from 64.72% to 84.19% and F1 score from 63.89% to 84.05%. The biggest improvement was obtained for River, Highway and PermanentCrop classes.

Radioelectronic and Computer Systems, 2021

The system of automatic license plate recognition (ALPR) is a combination of software and hardwar... more The system of automatic license plate recognition (ALPR) is a combination of software and hardware technologies implementing ALPR algorithms. It seems to be easy to achieve the goal but recognition of license plate requires many difficult solutions to some non-trivial tasks. If the license plate is oriented horizontally, uniformly lighted, has a clean surface, clearly distinguishable characters, then it'll be not too difficult to recognize such a license plate. However, the reality is much worse. The lighting of each part of the plate isn't equal; the picture from the camera is noisy. Besides, the license plate can have a big angle relative to the camera and be dirty. These obstacles make it difficult to recognize the license plate characters and determine their location on the image. For instance, the accuracy of recognition is much worse on large camera angles. To solve these problems, the developers of automatic license plate recognition systems use a different approach to processing and analysis of images. The work shows an automatic license plate recognition system, which increases the recognition accuracy at large camera angles. The system is based on the technology of recognition of images with the use of highly accurate convolutional neural networks. The proposed system improves stages of normalization and segmentation of an image of the license plate, taking on large camera angles. The goal of improvements is to increase of accuracy of recognition. On the stage of normalization, before histogram equalization, the affine transformation of the image is performed. For the process of segmentation and recognition, Mask R-CNN is used. As the main segment-search algorithm, selective search is chosen. The combined loss function is used to fasten the process of training and classification of the network. The additional module to the convolutional neural network is added for solving the interclass segmentation. The input for this module is generated feature tensor. The output is segmented data for semantic processing. The developed system was compared to well-known systems (SeeAuto.USA and Nomeroff.Net). The invented system got better results on large camera shooting angles.

Neurocomputing, 2020

Most widely used modern artificial neural networks are based on the connectionist paradigm of bui... more Most widely used modern artificial neural networks are based on the connectionist paradigm of building and learning. The authors propose an alternative detector approach. The basis of this approach is the original architecture of the neural network, as well as a new procedure for its learning. The developed neural network is called the detector neural network. This network consists of two layers of neurons. The neurons of the first layer are called neurons-pre-detectors and they do not learn. They are designed to highlight the structural elements of recognizable images, as well as to determine their measured parameters. The types of structural elements and their parameters are set a priori and depend on the type and complexity of recognizable images. Neurons of the second layer can be trained. They recognize individual complex images. These neurons are called neurons-detectors (ND). The model of the ND is significantly different from all known models of neurons and has important fea...

Advanced Information Systems, 2019

У сучасному світі нейронні мережі інтенсивно розвиваються і використовуються в усіх сферах людськ... more У сучасному світі нейронні мережі інтенсивно розвиваються і використовуються в усіх сферах людської діяльності. Їх застосування для визначення пожежонебезпечності лісових територій може розпочати вирішення проблеми попередження лісових пожеж. Лісові пожежі важко контролюються та, у разі виникнення, вимагають великої кількості ресурсів для їх усунення. Робота присвячена вирішенню задачі визначення пожежонебезпечності лісових територій. Розглядається територія пожежі «Camp Fire», що сталася у Каліфорнії (США). Метою роботи є дослідження можливості застосування згорткових нейронних мереж для визначення пожежонебезпечності лісових територій на основі мультиспектральних зображень, отриманих з супутника Landsat 8. Поставлена мета передбачає вирішення таких завдань : огляд територій, на яких відбулися наймасштабніші лісові пожежі за останній час, аналіз економічних та екологічних збитків від лісових пожеж; отримання та обробка мультиспектральних зображень території пожежі з супутника Lands...

CEUR Workshop Proceedings, COLINS-2021, 5th International Conference on Computational Linguistics and Intelligent System, 2021

There are many social media and messengers in use today, because of the situation with the corona... more There are many social media and messengers in use today, because of the situation with the corona virus pandemic the social media have become an integral part of our daily lives, including work activities. However, there is a lot of unnecessary information that comes to users in large quantities, so the problem of dealing with spam messages on social networks and messengers is now very relevant. By spam we mean any messages that a particular user (person, company, etc.) considers unnecessary in a particular text stream. The project is dedicated to solving the scientific problem of detecting spam messages in the text context of any social network or messenger using anti-spam bot that is based on various spam detection algorithms. Four algorithms were implemented and investigated: an algorithm using naive Bayesian classifier, support vector method, multilayer perceptron neural network and convolutional neural network. The main idea is to develop a complex spam detection algorithm for ...

Breast cancer is the most common in the world and its rates could be increased from 2 million in ... more Breast cancer is the most common in the world and its rates could be increased from 2 million in 2018 to 3 million in 2040, and the death rate from 600 thousand to nearly 1 million per year. Histopathological analysis is used for diagnosis of almost all cancer types. Nowadays histopathological tissue analysis and evaluating the microscopic appearance of a biopsied tissue sample are provided by a pathologist. The paper is devoted to the problem of histopathological analysis automatization using a region-based convolutional neural network (RCNN). The purpose of the research is to automatizate the tumor nuclei detection in the histopathological images, because detection can be used as qualitative and quantitative analysis. In the research breast cancer histopathological annotation and diagnosis dataset is used (BreCaHAD). The classification accuracy for SVM classifier, which uses features, extracted by CNN, is 0.96. The object detection heatmap was built. It is obtained that the averag...

The subject of the study in the article is the neural network method of object recognition on mul... more The subject of the study in the article is the neural network method of object recognition on multispectral Earth remote sensing (ERS). The goal providing automatic recognition of objects illegal exploitation of natural resources in multispectral ERS images. The task is formulation of the method of intellectual processing of ERS data, which implements automatic recognition of objects of illegal use of natural resources on multispectral ERS images by using a convolutional neural network.. Analysis of the problems of methods and algorithms for processing multispectral aerospace images has shown that it is most promising to use flexible algorithms that adapt to changing conditions for observing search objects. One of the most promising technologies of the implementation of such algorithms is the use of neural networks. The selection of convolutional neural networks for solving the recognition problem is related to the ability of these networks, under the condition of correct training, ...

2019 IEEE 2nd Ukraine Conference on Electrical and Computer Engineering (UKRCON), 2019

The scientific and practical problem of detecting fire hazardous forest areas by using deep learn... more The scientific and practical problem of detecting fire hazardous forest areas by using deep learning artificial neural networks applied to Camp Fire (California, USA) is considered in the paper. The theory of deep learning neural networks, the theory of recognition multispectral images and mathematical statistics methods are used. A novel solution of the multispectral images recognition method by using capsule and convolutional neural networks applied to Camp Fire area is presented. A comparative analysis of convolutional and capsule neural networks is conducted.

CEUR Workshop Proceedings, vol. 2870, 5th International Conference on Computational Linguistics and Intelligent Systems, Volume I: Main Conference, COLINS 2021, 2021

2021 IEEE 3rd Ukraine Conference on Electrical and Computer Engineering (UKRCON)

Advanced Information Systems, Sep 30, 2021

The subject of study in the article is artificial intelligence methods that can be used for recog... more The subject of study in the article is artificial intelligence methods that can be used for recognition of specific areas of the earth's surface in multispectral images provided by Earth remote sensing systems (ERS). The goal is to automate data analysis for recognizing areas affected by fire on multispectral remote sensing images. The task is to study and formulate a method for processing multispectral data, which makes it possible to automate the process of operational recognition of areas of burned-out areas in images, for the development of an eco-monitoring software system using artificial intelligence tools such as deep learning and neural networks. As a result of the analysis of modern methods of processing multispectral data, an investigation of the supervised learning strategy was chosen. The choice of the described method for solving an applied problem is based on the high flexibility of these method, as well as, provided that there is a sufficient amount of used training input data and correct training strategies, the possibility of analyzing heterogeneous multispectral data with ensuring high accuracy of results for each individual sample. Conclusions: the application of methodologies for intelligent processing of multispectral images has been investigated and substantiated. The theoretical foundations of the construction of neural networks are considered, the applied area of application is selected. An architectural model of a software product is analyzed and proposed, taking into account its scalability, the model of software system is developed and the results of its work are shown. The obtained results show the efficiency of proposed system and prospects of the proposed algorithms, which is a reason for further research and improvement of the used algorithms, with their possible use in industrial and enterprise eco-monitoring systems. K e y w or d s : eco-monitoring system; Earth remote sensing; multispectral image; image processing; neural network.

Handbook of Research on Artificial Intelligence Applications in the Aviation and Aerospace Industries

This chapter uses deep learning neural networks for processing of aerospace system multispectral ... more This chapter uses deep learning neural networks for processing of aerospace system multispectral images. Convolutional and Capsule Neural Network were used for processing multispectral images from satellite Landsat 8, previously processed using spectral indices NDVI, NDWI, PSRI. The authors' approach was applied to wildfire Camp Fire (California, USA). The deep learning neural networks are used to solve the problem of detecting fire hazardous forest areas. Comparison of Convolutional and Capsule Neural Network results was done. The theory of neural networks of deep learning, the theory of recognition of multispectral images, methods of mathematical statistics were used.

Advanced Information Systems, 2024

Advanced Information Systems

Automation of technological and business processes

Системи управління, навігації та зв’язку. Збірник наукових праць

Sučasnì ìnformacìjnì sistemi, Dec 23, 2019

Advanced Information Systems, 2023

Automation of technological and business – processes, 2023

Journal of information and organizational sciences

Radioelectronic and Computer Systems, 2021

Neurocomputing, 2020

Advanced Information Systems, 2019

CEUR Workshop Proceedings, COLINS-2021, 5th International Conference on Computational Linguistics and Intelligent System, 2021

2019 IEEE 2nd Ukraine Conference on Electrical and Computer Engineering (UKRCON), 2019

CEUR Workshop Proceedings, vol. 2870, 5th International Conference on Computational Linguistics and Intelligent Systems, Volume I: Main Conference, COLINS 2021, 2021

2021 IEEE 3rd Ukraine Conference on Electrical and Computer Engineering (UKRCON)

Advanced Information Systems, Sep 30, 2021

Handbook of Research on Artificial Intelligence Applications in the Aviation and Aerospace Industries

IEEE 2015 13th International Conference The experience of designing and application of CAD systems in microelectronics (CADSM) , 2015

There is a description of modern satellite-based Earth remote sensing systems in the paper. It sh... more There is a description of modern satellite-based Earth remote sensing systems in the paper. It shows the contradiction between the tendency to increase the resolution, increasing the number of spectral channels of onboard equipment against the limited on-board computational resources and communication channels using modern technologies. A solution in the form of intellectualization visual data processing is proposed by using neural network technology in the Earth remote sensing systems.