The given article considers the development of a personal identification technique based on the mechanism of scanning and analyzing such biometric parameter as a vein pattern of the palm for automation access control systems. A number of problems characteristic of the existing approaches to solving the given problem have been formulated and the operation analysis of the main ones has been carried out. A mechanism for reading a vein pattern of the palm, as well as three methods for further analysis of the referred biometrics and personal identification: a method based on a categorical classification, a method based on a binary classification, and a combined method have been developed. The resulting architecture of the neural network for the categorical classification of the vein pattern has been built and a method for calculating the number of the model parameters depending on the number of the registered subjects has been obtained. Based on the results of the research, experimental measurements of the system operation accuracy have been made while implementing the mentioned methods. The system based on a binary classification has demonstrated the highest accuracy; however applying a combined approach allows improving the obtained result.

No abstract

The given article presents the study of the training process of a composite heterogeneous neural network with deep architecture. A brief description of the architecture of the analyzed neural network system has been given. The key nodes of computational load distribution on each of the layers of the neural network have been tracked. The monitoring results have been analyzed and the following conclusions have been made: the most resource-intensive part of the system during training is the LSTM network. The results of the study of structural features of the neural network hidden layers have been presented. Graphs have been constructed and there has been carried out the study of a statistical distribution of weights on each unique layer of the architecture: on a fully connected layer, on the first hidden layer of the subnet-encoder, on the second hidden layer of the subnet-encoder, on the first and second fully connected output layers of the neural network. Based on the research results, a qualitative assessment of the effectiveness and accuracy of the entire neural network system has been given.

No abstract

No abstract