This article advances the research on the intelligent monitoring and control of helicopter turboshaft engines in onboard conditions. The proposed neural network system for anomaly prediction functions as a module within the helicopter turboshaft engine monitoring and control expert system. A SARIMAX-based preprocessor model was developed to determine autocorrelation and partial autocorrelation in training data, accounting for dynamic changes and external factors, achieving a prediction accuracy of up to 97.9%. A modified LSTM-based predictor model with Dropout and Dense layers predicted sensor data, with a tested error margin of 0.218% for predicting the TV3-117 aircraft engine gas temperature values before the compressor turbine during one minute of helicopter flight. A reconstructor model restored missing time series values and replaced outliers with synthetic values, achieving up to 98.73% accuracy. An anomaly detector model using the concept of dissonance successfully identified two anomalies: a sensor malfunction and a sharp temperature drop within two minutes of sensor activity, with type I and II errors below 1.12 and 1.01% and a detection time under 1.611 s. The system’s AUC-ROC value of 0.818 confirms its strong ability to differentiate between normal and anomalous data, ensuring reliable and accurate anomaly detection. The limitations involve the dependency on the quality of data from onboard sensors, affected by malfunctions or noise, with the LSTM network’s accuracy (up to 97.9%) varying with helicopter conditions, and the model’s high computational demand potentially limiting real-time use in resource-constrained environments.

The research aims to develop a neural network-based lost information restoration method when the complex nonlinear technical object (using the example of helicopter turboshaft engines) sensors fail during operation. The basis of the research is an auto-associative neural network (autoencoder), which makes it possible to restore lost information due to the sensor failure with an accuracy of more than 99%. An auto-associative neural network (autoencoder)-modified training method is proposed. It uses regularization coefficients that consist of the loss function to create a more stable and common model. It works well on the training sample of data and can produce good results on new data. Also, it reduces its overtraining risk when it adapts too much to the training data sample and loses its ability to generalize new data. This is especially important for small amounts of data or complex models. It has been determined based on the computational experiment results (the example of the TV3-117 turboshaft engine) that lost information restoration based on an auto-associative neural network provides a data restoring error of no more than 0.45% in the case of single failures and no more than 0.6% in case of double failures of the engine parameter registration sensor event.

Vectorization is happened by a wave method. The wave method is to analyze the path of the way to the spherical wave in the image. At every step is analyzed offset center of point's mass forming a new step wave relative to the previous position of the center of mass. Offered of the perfected wave tracing algorithm is to use a spherical wave on the combined raster image while creating skeleton. The feature of skeletezationon prints, where deposited security elements in the form guilloché, is that almost no curves intersect. This avoids skeletezation of connection segments.

Biometric recognition systems require the development of new technologies and need improvement. A new filter of Ateb-Gabor has been investigated. The filter is based on the use of a combination of Gabor filter and periodic Ateb-functions. This combination allows us to provide flexibility for control by choosing two parameters m and n, which is provided by the mathematical apparatus of the Ateb functions. Filtration was performed on the example of biometric images. It has been shown that Ateb-Gabor filtration gives better filtration results than ordinary Gabor filter. Experimental research has been carried out.

Background: Systems of the Internet of Things are actively implementing biometric systems. For fast and high-quality recognition in sensory biometric control and management systems, skeletonization methods are used at the stage of fingerprint recognition. The analysis of the known skeletonization methods of Zhang-Suen, Hilditch, Ateb-Gabor with the wave skeletonization method has been carried out and it shows a good time and qualitative recognition results. Methods: The methods of Zhang-Suen, Hildich and thinning algorithm based on Ateb-Gabor filtration, which form the skeletons of biometric fingerprint images, are considered. The proposed thinning algorithm based on Ateb-Gabor filtration showed better efficiency because it is based on the best type of filtering, which is both a combination of the classic Gabor function and the harmonic Ateb function. The combination of this type of filtration makes it possible to more accurately form the surroundings where the skeleton is formed. Results: Along with the known ones, a new Ateb-Gabor filtering algorithm with the wave skeletonization method has been developed, the recognition results of which have better quality, which allows to increase the recognition quality from 3 to 10%. Conclusion: The Zhang-Suen algorithm is a 2-way algorithm, so for each iteration, it performs two sets of checks during which pixels are removed from the image. Zhang-Suen's algorithm works on a plot of black pixels with eight neighbors. This means that the pixels found along the edges of the image are not analyzed. Hilditch thinning algorithm occurs in several passages, where the algorithm checks all pixels and decides whether to replace a pixel from black to white if certain conditions are satisfied. This Ateb-Gabor filtering will provide a better performance, as it allows to obtain more hollow shapes, organize a larger range of curves. Numerous experimental studies confirm the effectiveness of the proposed method.

In this paper we propose a method of cryptographic protection of information which is based on encrypting blocks. The threats that mostly influence the loss of information are analyzed, therefore, one of the potential threats to information in information systems should be considered accidental or deliberate destructive actions of staff. Block cipher, which is based on certain key, is built. The essence of this method is to replace the original text information in a file to an encrypted offset by a nonlinear law based on tabulated value of hyperbolic Ateb-functions. The method of playback of encrypted information without loss is developed. Experiments confirming cipher strength are carried out.