NUNurgul Uzakkyzy

Nurgul Uzakkyzy

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Nurgul Uzakkyzy has a total of 38 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Combined analysis of the importance of factors in agricultural process management tasks

    by Gulzira Abdikerimova, Moldir Yessenova, Aizhan Zharkimbekova, Zhanar Beldeubayeva, Aigulim Bayegizova, Nurgul Uzakkyzy, Ainagul Alimagambetova, Gulden Murzabekova

    The article presents a combined approach for analyzing the significance of factors in the agro-industrial sector using Shapley additive explanations (SHAP), simple combination, and principal component analysis (PCA)+combination methods. The study addresses the pressing need for efficient agricultural resource management under constrained and changing climatic conditions. The proposed methodology evaluates the impact of various factors on key performance indicators such as yield, income, and operating costs. SHAP analysis identified critical determinants, with "Land Area (ha)" contributing significantly to "Market Capacity" (59.5%) and "Sales Revenue" (57.2%), highlighting the importance of production scale. The simple combination method, integrating gradient boosting (GB), mutual information (MI), and recursive feature elimination (RFE) with Lasso, revealed a more balanced factor distribution, assigning 14.5% to "Land Area" and 12.8% and 10.7% to “Seed Use” and “Fertilizer Cost,” respectively. The PCA+combination method emphasized global trends, identifying "Yield per Hectare" (22.5%) and "Field Size" (11.5%) as key contributors to variance. This integrative approach captures localized effects and global interdependencies, offering comprehensive data interpretations. The findings are instrumental in optimizing resource management and strategic planning and enhancing agricultural production efficiency.

  • ОПРЕДЕЛЕНИЕ ВЕКТОРА ПРИЗНАКОВ ДЛЯ ТОЧНОГО ВЫЯВЛЕНИЯ РАКА МОЛОЧНОЙ ЖЕЛЕЗЫ НА ИЗОБРАЖЕНИЯХ МАММОГРАММЫ

    by Д.и. Усипбекова, М.б. Есенова, Н. Узаккызы, Р. Муратхан, П. Шмидт

    В данной статье представлен гибридный метод классификации рака молочной железы на основе маммографических изображений в формате цифровой визуализации и коммуникаций в медицине (DICOM). Метод основан на объединении методов глубокого обучения и текстурных и геометрических характеристик изображений. Целью исследования является интеграция текстурных и геометрических характеристик в архитектуру нейронной сети для повышения точности классификации и локализации аномальных областей на медицинских изображениях. Актуальность исследования обусловлена ростом заболеваемости раком и необходимостью повышения точности диагностики в условиях ограниченных ресурсов медицинских учреждений. Предлагаемый метод основан на модифицированной архитектуре сверточной нейронной сети на основе более быстрых регионов (Faster R-CNN), которая включает в себя дополнительные блоки извлечения признаков, такие как эксцентриситет, метрики текстуры, энтропия и средняя интенсивность. Этап предварительной обработки изображения включает в себя преобразования таблицы поиска (LUT), нормализацию и выравнивание гистограммы. Эти изменения позволяют улучшить ключевые структурные особенности изображения. Эффективность метода была дополнительно повышена за счет интеграции текстурных и геометрических характеристик, что позволило сократить общую стоимость на 30%, а также улучшить качество классификации и локализации.

  • Open Access

    SING COMPUTER VISION ALGORITHMS TO IDENTIFY MOVING OBJECTS

    by G. M. Baenova, K. S. Agadilova, Sh. Zh. Seilov, N. Uzakkyzy, Н. Ұзаққызы

    There are no universal algorithms for solving problems of recognizing moving objects by video surveillance systems. However, for different systems and in the case of different situations, only some specific algorithm is optimal, allowing for object recognition. This article analyzes the stability of intelligent algorithms that affect the quality of speech recognition and considers an integrated approach that integrates object detection, classification of people, and recognition of gender differences. The accumulated experience in the field of pattern recognition has allowed us to achieve high results in the creation of various devices and systems in medicine, in the industrial sector, in information processing systems and video surveillance. However, computer vision technologies and optical recognition of dynamic objects continue to be an extremely difficult part of scientific research due to the variety of video cameras and devices. As well as a wide range of applications, primarily for security purposes in crowded places, disorder detection, etc. This study presents the main tasks for developing a software system using computer vision and deep learning algorithms to identify and classify people in video streams, determine their number and determine their gender.

  • Open Access

    Image noise reduction by deep learning methods

    by Nurgul Uzakkyzy, Aisulu Ismailova, Talgatbek Ayazbaev, Zhanar Beldeubayeva, Shynar Kodanova, Balbupe Utenova, Aizhan Satybaldiyeva, Mira Kaldarova

    Image noise reduction is an important task in the field of computer vision and image processing. Traditional noise filtering methods may be limited by their ability to preserve image details. The purpose of this work is to study and apply deep learning methods to reduce noise in images. The main tasks of noise reduction in images are the removal of Gaussian noise, salt and pepper noise, noise of lines and stripes, noise caused by compression, and noise caused by equipment defects. In this paper, such noises as the removal of raindrops, dust, and traces of snow on the images were considered. In the work, complex patterns and high noise density were studied. A deep learning algorithm, such as the decomposition method with and without preprocessing, and their effectiveness in applying noise reduction are considered. It is expected that the results of the study will confirm the effectiveness of deep learning methods in reducing noise in images. This may lead to the development of more accurate and versatile image processing methods capable of preserving details and improving the visual quality of images in various fields, including medicine, photography, and video.


  • Open Access

    Application of informative textural Law’s masks methods for processing space images

    by Moldir Yessenova, Gulzira Abdikerimova, Gulden Murzabekova, Kakabayev Nurbol, Natalya Glazyrina, Saltanat Adikanova, Nurgul Uzakkyzy, Zhanna B. Sadirmekova, Rozamgul Niyazova

    Image processing systems are currently used to solve many applied problems. The article is devoted to the identification of negative factors affecting the growth of grain in different periods of harvesting, using a program implemented in the MATLAB software environment, based on aerial photographs. The program is based on the Law’s textural mask method and successive clustering. This paper presents the algorithm of the program and shows the results of image processing by highlighting the uniformity of the image. To solve the problem, the spectral luminance coefficient (SBC), normalized difference vegetation index (NDVI), Law’s textural mask method, and clustering are used. This approach is general and has great potential for identifying objects and territories with different boundary properties on controlled aerial photographs using groups of images of the same surface taken at different vegetation periods. That is, the applicability of sets of Laws texture masks with original image enhancement for the analysis of experimental data on the identification of pest outbreaks is being investigated.


  • Open Access

    Assessment of plant disease detection by deep learning

    by Akan Alpyssov, Nurgul Uzakkyzy, Ayazbaev Talgatbek, Raushan Moldasheva, Gulmira Bekmagambetova, Mnyaura Yessekeyeva, Dossym Kenzhaliev, Assel Yerzhan, Ailanysh Tolstoy

    Plant disease and pest detection machines were originally used in agriculture and have, to some extent, replaced traditional visual identification. Plant diseases and pests are important determinants of plant productivity and quality. Plant diseases and pests can be identified using digital image processing. According to the difference in the structure of the network, this study presents research on the detection of plant diseases and pests based on three aspects of the classification network, detection network, and segmentation network in recent years, and summarizes the advantages and disadvantages of each method. A common data set is introduced and the results of existing studies are compared. This study discusses possible problems in the practical application of plant disease and pest detection based on deep learning. Conventional image processing algorithms or manual descriptive design and classifiers are often used for traditional computer vision-based plant disease and pest detection. This method usually uses various characteristics of plant diseases and pests to create an image layout and selects a useful light source and shooting angle to produce evenly lit images. The purpose of this work is to identify a group of pests and diseases of domestic and garden plants using a mobile application and display the final result on the screen of a mobile device. In this work, data from 38 different classes were used, including diseased and healthy leaf images of 13 plants from plantVillage. In the experiment, Inception v3 tends to consistently improve accuracy with an increasing number of epochs with no sign of overfitting and performance degradation. Keras with Theano backend used to teach architectures


  • About the results of the processing route radarogram Haar wavelets and daubechies

    by N. Uzakkyzy, K. Iskakov, S. Boranbayev

    Application of GPR in restoring the structure of the underground environment can be reduced to the solution of ill-posed problems for hyperbolic and parabolic types of equations. Many researchers describe algorithms for this kind of problems. At present, methods of interpretation georadarograms can be improved by new achievement of the inverse problems theory. The bottleneck of application of GPR is the complexity of interpreting data required at the present stage to attract highly qualified professionals. Forming of own methods of interpreting GPR data contributes to preservation of confidential geophysical information. The main area of application of the wavelet transforms is the analysis and processing of signals and functions that are nonstationary in time or inhomogeneous in space, when the results of the analysis should contain not only the frequency response of the signal (signal energy distribution by frequency components), but also information about local coordinates, on which some groups of frequency components are expressed or on which rapid changes in the frequency components of the signal take place [1]. In comparison with the decomposition of signals into Fourier series, wavelets can represent local features of signals with much higher accuracy. A discrete wavelet transform can be used to easily and quickly remove noise from a noisy signal. If we leave only a limited number of the highest coefficients of the spectrum of the discrete wavelet transform, and carry out the inverse wavelet transform, we can get a signal cleared of noise. A copper cable buried at a depth of 122 cm will be used as a research object. The result of the georadar survey is a set of single tracks (signals), registered by the receiving antenna at each position of the radar. The radar scan of the cable scan was obtained by the GPR `LOZA' (Figure 1). The trace of this radargram is processed by the Wavelet Toolbox, and by the System of numerical and mathematical modeling MatLab [2].