Intruder recognition in uncontrolled environments is a critical function of biometric-based security control systems (SCS), ensuring protection at facilities with large crowds. The accuracy of such systems can be improved by combining multiple biometric traits extracted from video imagery. However, processing video data involves several challenges, including viewpoint variation, occlusion, and selecting an appropriate feature fusion level. To resolve these challenges, a complete approach to intruder recognition is proposed. The approach is based on video preprocessing, the combination of two convolutional neural networks (CNNs), Dempster–Shafer theory, and the random forest method. These techniques fuse behavioral biometric features at the score level to classify video sequences. Gait and gestural behavior are selected as biometric modalities, as they can be captured without direct subject interaction. For the classification task, three classes of video data are defined: class 1 – person is not intruder, class 2 – a potential intruder, class 3 – intruder. The study also presents a general architecture for the proposed approach, along with a detailed description of its processing stages. The effectiveness of the approach is measured through experiments performed on the KTH dataset, which comprises six types of simple human actions performed by different subjects under different background conditions. Experimental results show that the proposed approach improves intruder recognition accuracy in uncontrolled environments, achieving an 87 % classification rate.
Electrical equipment (EE) is a key part of industrial electrical systems where unexpected mechanical failures in operation can cause serious consequences (disruption of the technological process, reduction in the quality and quantity of manufactured products and emergencies). For timely detection of such faults, as well as to ensure normal operation of the systems, it is required to conduct regular assessment of EE technical state using modern computer technologies under conditions of incomplete and fuzzy information. To solve this problem, we propose an approach using quantization and convolutional neural networks (CNNs) which differs from existing approaches by complex processing of thermograms obtained with a thermal imaging device; images with black-and-white and color graphs obtained from instruments or built based on statistical data. This approach provides an opportunity to improve the accuracy of classification of various EE malfunctions, reduce unscheduled equipment failures due to prompt decisionmaking regarding the EE technical state under conditions of incomplete and fuzzy information. The review of studies in this subject area by both Russian and foreign scientists reflects a number of successful experiments on the use of CNNs. The CNN developed to classify faults outputs a class number to which the current state of the equipment relates (class 1 – serviceable EE; class 2 – serviceable EE with small deviations). This paper considers a generalized scheme and algorithm of a complex approach to EE fault detection with their detailed description. The study results were obtained when diagnosing the asynchronous motor АИР63А4У1 and confirm the validity and objectivity of using the proposed approach
