The objective of this work is to develop approaches to automating inspection procedure at airports. The article presents the deficiencies of the existing inspection system, concluding in the negative impact of the human factor. It is proposed to use convolutional neural networks for automatic x-ray image analysis of passenger baggage. The paper presents the results of the convolutional neural network with various input data and architecture within limited computing resources. In a view to further development, this study can contribute to the development of specialized software to help aviation security screeners through partial automation of their work.

The article considers the use of neural networks to solve the problem of recognizing dangerous and safe objects carried in the luggage of airport passengers. A comparative analysis is performed to define the accuracy achieved on the test sample for different convolutional neural networks. It also explores the influence of various regularizations on the accuracy of a two-class classification. The increased probability of correct recognition is achieved due to augmentation, reset weights and saturation of the network. The method of transfer training is used to increase the efficiency of the recognizer. In this case, a study was carried out for the transfer of various neural networks.

No abstract

The article proposes a unified principle of permutation decoding (PD), which applies to redundant systematic block codes. This method allows using correction capabilities of such codes, but in the classical interpretation, the method requires cumbersome matrix calculations, which does not allow using the positive properties of the error correction method. The computing process is excessively complex. Therefore, to reduce the negative effect in the PD system, it is proposed to use the cognitive principle of data processing at the channel level, which significantly reduces the decoder complexity and ensures PD application in the air passenger biometric data systems.

In this paper, an analysis of world experience was conducted and it was concluded that one of the ways to improve the efficiency of aviation security in the Russian Federation is to use modern network training complexes. A new approach to assessing the competence of aviation security screeners was proposed and tested, allowing taking into account the parameters of oculomotor activity and heart rate variability of test aviation security screeners, and differing from the existing approaches by using fuzzy classification models. According to the results of an experimental study, three different models were synthesized. The results of the comparison showed that the Sugeno model, trained using the ANFIS-algorithm, is more accurate than the Mamdani model and the linear regression model depends on the competence assessment of aviation security screeners. It described ways of addressing the important task of obtaining more precise relevant digital data in network training complexes using noise-resistant coding tools. It presented a model of a permutation decoder of a non-binary redundant code based on a lexicographic cognitive map. This model of a redundant code decoder uses cognitive data processing methods for completing permutation decoding procedures in order to protect remote control commands from the influence of destructive factors on the control process.