Modern Video Surveillance Systems Research Articles

Implementation and optimisation of intelligent police systems based on artificial intelligence

The rapid development of artificial intelligence provides new opportunities for law enforcement agencies. Nowadays, the developed countries of the world are increasingly using surveillance cameras to monitor public safety, detect criminals and suspicious objects. The facial identification systems on the market have tremendous potential to help law enforcement agencies. Facial recognition software helps to identify missing persons and criminals whose faces are caught on CCTV cameras. The use of artificial intelligence in such systems accelerates their operation, which, in turn, facilitates the quick search for suspects and their rapid apprehension. Modern video surveillance systems can help counter terrorist attacks by tracking and identifying people and suspicious objects. On the other hand, the issue of personal data protection and privacy when using CCTV cameras to identify people's faces is increasingly being discussed. The obvious solution to this problem is to regulate it at the legislative level, in particular, to introduce guidelines aimed at ensuring transparency and accountability of the use of facial recognition software. For a more objective understanding of the circumstances which should be regulated by law, the author conducts a study of modern technical solutions in the field of facial identification with integrated artificial intelligence, their features and possibilities of use in the work of the National Police of Ukraine, and also identifies the steps which outline the sequence of actions during objective facial identification of people and ensure the high quality of this process and the reliability of its results.

Choice of architecture of video surveillance systems of State Border Guard of Ukraine based on cloud and fog technologies

Modern video surveillance systems have evolved from closed-circuit television systems to the level of complex systems that operate as part of complex integrated systems and solve the tasks of not only recording events in the sectors of view of video surveillance cameras but also analysing the received video information. The increase in the amount of information circulating in modern systems requires the creation of new concepts. In recent years, the attention of scientists has been drawn to distributed information processing technologies. The concept of cloud computing is developing rapidly, and the basic ideas that were laid down in its construction can no longer solve the problems faced by the technology. This has led to the emergence of so-called post-cloud architectures that extend and complement the capabilities of cloud computing. These architectures include Mist, Edge, Fog, etc. The purpose of the article is to study video surveillance systems based on the concept of cloud and fog computing and criteria for assessing their effectiveness. The research was conducted using the methods of analysis and generalisation, modelling, and experimentation. This approach made it possible to conduct a comparative study of video surveillance systems built on two architectures. The results obtained indicate that fog computing technology has the advantage of reducing latency, minimising the need for repeated requests to the cloud by performing calculations at the cloud edge. In a model based on the cloud computing concept, services that use cloud resources lead to an increase in network load. At the same time, fog computing technology allows you to relieve the network load by performing part of the computation by fog nodes. The results of the experimental study show the advantages of fog computing for networks that are sensitive to delays. However, if we consider a video surveillance system with the main task of recording events in the sectors of view of CCTV cameras and the ability to view video information in real time, it is obvious that fog nodes will not provide long-term storage of video information, and delays will not be critical. In the case of a video surveillance system with video analytics functions, fog nodes will be able to perform part of the video analytics algorithms, thus unloading the cloud. Therefore, an urgent task is to study the effectiveness of building video surveillance systems with video analytics functions based on fog architecture.

Facial Emotion Recognition for Photo and Video Surveillance Based on Machine Learning and Visual Analytics

Modern video surveillance systems mainly rely on human operators to monitor and interpret the behavior of individuals in real time, which may lead to severe delays in responding to an emergency. Therefore, there is a need for continued research into the designing of interpretable and more transparent emotion recognition models that can effectively detect emotions in safety video surveillance systems. This study proposes a novel technique incorporating a straightforward model for detecting sudden changes in a person’s emotional state using low-resolution photos and video frames from surveillance cameras. The proposed technique includes a method of the geometric interpretation of facial areas to extract features of facial expression, the method of hyperplane classification for identifying emotional states in the feature vector space, and the principles of visual analytics and “human in the loop” to obtain transparent and interpretable classifiers. The experimental testing using the developed software prototype validates the scientific claims of the proposed technique. Its implementation improves the reliability of abnormal behavior detection via facial expressions by 0.91–2.20%, depending on different emotions and environmental conditions. Moreover, it decreases the error probability in identifying sudden emotional shifts by 0.23–2.21% compared to existing counterparts. Future research will aim to improve the approach quantitatively and address the limitations discussed in this paper.

A NEURAL NETWORK MODEL FOR RECOGNIZING HUMAN POSES BASED ON INFORMATION OBTAINED FROM VIDEO IMAGES

Nowadays, there are several ways of recognizing and identifying a person, and these keep enhancing day by day. However, cases of falsification of these methods can also be observed. Modern video surveillance systems keep on developing, and satellite systems enable us to photograph everything that happens in a certain area and analyze effectively the data obtained using neural networks. Video images help to monitor motions of people, detect illegal entries into prohibited areas, as well as to identify remaining criminals by means of photos taken from cameras and control exit and/or entry of wanted criminals to foreign countries by acquiring other biometric data. This system serves to catch criminals at airports, railway stations, seaports; it automatically counts numbers of people in a line or in a crowd and analyzes the nature of their movements, which lessens the amount of subjective human intervention and reduces time required for data processing. Moreover, neural network-based pose estimation system is now being used widely in sports, and this can distinguish between athletes and change their behavior based on their personal characteristics.

AXM-Net: Implicit Cross-Modal Feature Alignment for Person Re-identification

Cross-modal person re-identification (Re-ID) is critical for modern video surveillance systems. The key challenge is to align cross-modality representations conforming to semantic information present for a person and ignore background information. This work presents a novel convolutional neural network (CNN) based architecture designed to learn semantically aligned cross-modal visual and textual representations. The underlying building block, named AXM-Block, is a unified multi-layer network that dynamically exploits the multi-scale knowledge from both modalities and re-calibrates each modality according to shared semantics. To complement the convolutional design, contextual attention is applied in the text branch to manipulate long-term dependencies. Moreover, we propose a unique design to enhance visual part-based feature coherence and locality information. Our framework is novel in its ability to implicitly learn aligned semantics between modalities during the feature learning stage. The unified feature learning effectively utilizes textual data as a super-annotation signal for visual representation learning and automatically rejects irrelevant information. The entire AXM-Net is trained end-to-end on CUHK-PEDES data. We report results on two tasks, person search and cross-modal Re-ID. The AXM-Net outperforms the current state-of-the-art (SOTA) methods and achieves 64.44% Rank@1 on the CUHK-PEDES test set. It also outperforms by >10% for cross-viewpoint text-to-image Re-ID scenarios on CrossRe-ID and CUHK-SYSU datasets.

Сравнение алгоритмов распознавания объектов

Despite the increasing popularity of process automation, modern video surveillance systems still require constant human involvement to establish the fact of dangerous situations. But at present, systems are becoming more complex, this leads to an increase in threats and it is no longer possible for the operator to keep track of all emerging threats. In addition, in the field of video surveillance, tasks have been added that a person can no longer control just by watching video cameras. In this connection, you need to automate the process. Methods that provide maximum detection stability for small object movements, zoom changes, turning the object at a small angle, and changing lighting are based on describing the image at specific points. A special point is a point that has a number of key features that distinguish it from many other points in the image. Special points are the main characteristics of the object in the video surveillance system. The best object recognition algorithms based on this principle are the SURF and SIFT algorithms. These algorithms search for the direct occurrence of the reference image in relation to the observed one. The article discusses algorithms for detecting objects in an image based on the description of the image by special points. A comparison of SIFT and SURF algorithms, the analysis highlighted particular points in the recognition of each object, error analysis AI Node in identifying objects in the video stream.

Video Surveillance System with Two Channels and Information Transmission

CCTV with recognition is widely used today. They are designed to provide information about the current state of a protected object (process) by collecting, processing, archiving, storing, displaying and analyzing this information. Modern video surveillance systems seek the outside world through the exchange of data via the Internet and allow not so much to observe the object and record the information received, but lately can process it with the use of image recognition algorithms and automatically affect the object management systems.The purpose of this work is to analyze the features of construction and operation of video surveillance systems, to consider various options for data transfer and the implementation of face recognition. Offer advanced CCTV options, such as additional face identification and emergency communications features.New CCTV systems are being built using IP cameras and related software. Smartphones and tablets, laptops, traditional computers and cloud technologies are also widely used as video viewing devices. Video surveillance is an important part of the security system, where IP cameras, recognition features and information transmission modules are now widely used in security systems.It is advisable to use an IP camera when there is constant Wi-Fi, which allows the information to be transmitted continuously. Surveillance cameras and human identification are important in that they can collect and store individuals' faces.The diversity of people's faces is enormous, which is why the system has a low probability of recognition. The requirements for the design and construction of such systems are incredibly high.Considering all the advantages and disadvantages of different types of data transmission modules, it follows that GSM is the best option for transmitting notifications, as its range is almost limitless. And the video stream can be transmitted both via GSM and via Wi-Fi.As a result of the analysis of modern autonomous CCTV systems, a variant of the system with additional functions is proposed. The system has two communication channels for better communication and identification.This system can be used in offices, establishments, and in ordinary homes. The proposed security system is available in facilities where an admission system is required. The system works as follows. For example, the room has a CCTV system and access is available to only three people. Their faces are already recorded in memory. When a person approaches the camera, the system identifies it and aligns it with the already stored information about the persons it has access to. If there is no match, a message is sent via the GSM module about the penetration of the object and a link to the IP address of the embedded server from which the video is broadcast from the premises.

A Distributed Tracking Algorithm for Counting People in Video

The problem of tracking people in a video stream with the aim of counting them is studied. Modern video surveillance systems, such as the Moscow video surveillance system, use hundreds of thousands of cameras. The use of modern methods developed for working on a single computer with an expensive graphical processor is economically inefficient for such large-scale systems. In this paper, a distributed tracking algorithm is proposed. It makes it possible to reduce the amount of computational resources due to detecting people in a sparse set of frames. The detection is performed on servers installed in a data center, while the video stream is processed by local camera computation nodes. The experimental evaluation showed that the proposed algorithm provides acceptable quality at the detection rate of 4/3 Hz.