Fractal geometry is an important field of mathematics and computer science. Fractal images have a complex structure and can be used in various fields, such as graphics, computer vision and design. In carpet design, fractal structures are used to create patterns and decorations. However, the creation of high-quality fractal images of carpets requires optimization of color parameters. In this article, we will propose two methods for optimizing the color parameters of carpet fractal images: a genetic algorithm and a simplex optimization method. In addition, we propose to use a neural network to classify the colors of fractal structures in carpet design.
23.04.2023 View more DownloadВ работе описаны стратегические цели Искусственного интеллекта и основные задачи научно-технической деятельности в этой области. Обозначены проблемы, актуальные для развития основных направлений и форм его деятельности. Предлагаются подходы к их решению, основанные на новом технологическом укладе, и обсуждается вопросы, важные для успешного развития данной научно-практической дисциплины в целом.
23.04.2023 View more DownloadВ статье рассматривается процесс проектирования нейросетевых методов решения задач в базах знаний интеллектуальных систем. Предложена модель версионирования нейросетевых методов решения задач, описанных на специализированном языке представления нейросетевых методов решения задач.
23.04.2023 View more DownloadMatnni normallashtirish text-to-speech tizimining muhim tarkibiy qismi bo‘lib, matnni normallashtirishdagi qiyinchilik nostandart so‘zlarni (non-standart words) ajratishdir. Ushbu maqola keng miqyosli o‘zbek tili nutqi asosida nostandart so‘zlarlarning taksonomiyasini ishlab chiqadi va ikki bosqichli nostandart so‘zlarlarni aniqlashtirish strategiyasini taklif qiladi, dastlabki tasniflash uchun cheklangan davlat avtomatlari va pastki sinflarni aniqlash uchun maksimal entropiya tasniflagichlari. yuqoridagi nostandart so‘zlar taksonomiyasiga asoslanib, ikki bosqichli yondashuv ochiq testda 98,53% aniqlikka erishadi. Tajribalar shuni ko‘rsatadiki, nostandart so‘zlar taksonomiyasining yuqori boshlang‘ich ishlashini ta'minlaydi va maksimal entropiya tasniflagichlari sezilarli yaxshilanishlarga erishadi.
23.04.2023 View more DownloadThe subject of study in image processing has advanced significantly in recent years because to the strong rise in interest in image super-resolution. The quality of images can be improved using a variety of techniques. In this research, the operating principles and image processing stages of one multi frame picture super-resolution approach were discussed. A sequence of low-resolution (LR) photos are combined to create a high-resolution (HR) image using the multi-frame image super-resolution (SR) technique. Recent years have seen a rise in interest in the issue of multi-frame SR reconstruction as it is necessary or desired in many actual applications. This group of algorithms often builds the association between the recorded LR pictures and the unidentified reconstructed HR image estimations using a linear observation model[1].
23.04.2023 View more DownloadAs a result of the daily increase in the number of resources in digital libraries and scientific information databases, significant difficulties arise in the process of finding the necessary information for users. For this reason, there is a need to create tools designed to extract resources corresponding to the interests and needs of the user from a large amount of data. This article explores the principles, methods and steps of creating a recommender system electronic library resources based on the information needs of users.
23.04.2023 View more DownloadThrough speech signals, it is possible to determine the identity of the speaker, the text of the speech, and the physical and mental state of the person. Speaker recognition offers effective solutions for today's biometric security systems. The 2 directions of speaker recognition are speaker identification and speaker verification, which consist of feature extraction, speaker model development, and character matching verification.
23.04.2023 View more DownloadIn this work, tools that allow saving time resources in the process of training a neural network based on artificial intelligence and information about their application are given. Here we consider how to speed up neural network training using the Intel Math Kernel Library (MKL) and present the comparative results obtained during the research.
23.04.2023 View more DownloadUshbu maqolada tizimida smart texnologiyalarni qo`llash bo`yicha fikr yuritilgan. Smart texnologiyalarni tutgan o`rni qanday ekanligi ko`rsatib o`tilgan.
23.04.2023 View more Download