The detection of weeds at the stages of cultivation is very important for detecting and preventing plant diseases and eliminating significant crop losses, and traditional methods of performing this process require large costs and human resources, in addition to exposing workers to the risk of contamination with harmful chemicals. To solve the above tasks, also in order to save herbicides and pesticides, to obtain environmentally friendly products, a program for detecting agricultural pests using the classical K-Nearest Neighbors, Random Forest and Decision Tree algorithms, as well as YOLOv5 neural network, is proposed. After analyzing the geographical areas of the country, from the images of the collected weeds, a proprietary database with more than 1000 images for each class was formed. A brief review of the researchers' scientific papers describing the methods they developed for identifying, classifying and discriminating weeds based on machine learning algorithms, convolutional neural networks and deep learning algorithms is given. As a result of the research, a weed detection system based on the YOLOv5 architecture was developed and quality estimates of the above algorithms were obtained. According to the results of the assessment, the accuracy of weed detection by the K-Nearest Neighbors, Random Forest and Decision Tree classifiers was 83.3 %, 87.5 %, and 80 %. Due to the fact that the images of weeds of each species differ in resolution and level of illumination, the results of the neural network have corresponding indicators in the intervals of 0.82–0.92 for each class. Quantitative results obtained on real data demonstrate that the proposed approach can provide good results in classifying low-resolution images of weeds.

Sign languages are an extremely important means of communication in many cases, especially for deaf and hard of hearing people. But the same gesture can convey different meanings in different countries, so many different sign languages have been developed all over the world. In this study, a convolutional neural network (CNN) model was developed based on an ensemble method containing the ResNet-50 and VGG-19 architectures, which will be able to classify the Kazakh sign language (KSL) consisting of 42 Kazakh alphabet signs (classes). A proprietary data set of 57,708 images for 42 signs of the KSL has been formed. The ensemble model was compared with ResNet-50 and VGG-19 by evaluation metrics such as accuracy, precision, recall, f1-measure, and loss function. The recognition accuracy of the ensemble method reached 95.7%, exceeding the performance of ResNet-50 and VGG-19. The developed method was also tested on test data, where 35 out of 42 gestures were recognized completely correctly. The reliability of the proposed approach and the classification results obtained by using preprocessing methods and data augmentation techniques to expand the data set was confirmed by a computational experiment.

Sign languages are characterized by complex and subtle hand movements, which are challenging for computer vision systems to accurately recognize. This study suggests an innovative deep learning pipeline specifically designed for reliable gesture recognition of Kazakh Sign Language. This approach combines key point detection using the YOLO model, optical flow estimation, and a bidirectional long short-term memory (BiLSTM) network. At the initial stage, a dataset is generated using MediaPipe, which is then used to train the YOLO model in order to accurately identify key hand points. After training, the YOLO model extracts key points and bounding boxes from video recordings of gestures, creating consistent representations of movements. To improve the recognition of dynamic gestures, the optical flow is calculated in an area covering 10% of the area around key points, which allows the dynamics of movements to be captured and provides additional time characteristics. The BiLSTM network is trained on multimodal input that combines data on keypoints, bounding boxes, and optical flow, resulting in improved gesture classification accuracy. The experimental results demonstrate that the proposed approach is superior to traditional methods based solely on key points, especially in recognizing fast and complex gestures. The proposed structure promotes the development of sign language recognition technologies, especially for poorly studied languages such as Kazakh, paving the way to more inclusive and effective communication tools.

Unpredictable strings are sequences of data with complex and erratic behavior, which makes them an object of interest in various scientific fields. Unpredictable strings related to chaos theory was investigated using a genetic algorithm. This paper presents a new genetic algorithm for converting large binary sequences into their periodic form. The MakePeriod method is also presented, which is aimed at optimizing the search for such periodic sequences, which significantly reduces the number of generations to achieve the result of the problem under consideration. The analysis of the deviation of a nonperiodic sequence from its considered periodic transformation was carried out, and methods of crossover and mutation were investigated. The proposed algorithm and its associated conclusions can be applied to processing large sequences and different values of the period, and also emphasize the importance of choosing the right methods of crossover and mutation when applying genetic algorithms to this task.

The optimization capabilities of Kolmogorov–Arnold Networks (KANs) remain largely unexplored, which has limited their practical use in video anomaly recognition compared to conventional 3D-CNNs. To address this gap, we introduce a novel hybrid optimization framework that combines a Minimax Ant System (MMAS) for hyperparameter selection with a modified DARTS strategy for adaptive tuning of the 3D KAN architecture. Unlike existing approaches, our method simultaneously optimizes both learning dynamics and architectural configurations, enabling KANs to better exploit their expressive power in spatiotemporal feature extraction. Applied to a three-class video dataset, the proposed approach improved model accuracy to 87%, surpassing the performance of a standard 3D-CNN by 6%.