Architecture For Video Surveillance Research Articles

Abnormal Activity Recognition with Residual Attention-based ConvLSTM Architecture for Video Surveillance

Abnormal Activity Recognition with Residual Attention-based ConvLSTM Architecture for Video Surveillance

Visual Surveillance for Human Fall Detection in Healthcare IoT

This article designs a visual surveillance framework for human fall detection. In order to solve the conventional issues in fall detection, such as unsatisfactory feature generalization, low recall rates, and large computational time, we design a model that incorporates the deep convolutional neural network and the aggregated heuristic visual features in detecting the occurrence of falls. First, the convolutional neural network (Openpose model) is utilized to extract human skeleton in the image. Second, the hand-crafted spatial features, such as the angle of human shank inclination, are aggregated to determine the fall presence. It should be noticed that our fall detection method has been integrated to healthcare Internet of Things (IoT) video surveillance architecture, which has multiple graphic processing unit groups to perform real-time monitoring and alarming for the elderly in need. The experimental results prove that our method is able to accurately distinguish fall and nonfall activities with a competitive false-alarm rate.

Toward an Online Continual Learning Architecture for Intrusion Detection of Video Surveillance

With recent advances in deep learning technologies, many commercialized video surveillance systems have adopted Artificial Intelligence (AI)-powered video analytics technologies as a way to make our life smarter and safer. Nevertheless, there is no robust architecture with an appropriate network model for commercial services considering both high accuracy and low computational cost. Existing deep learning technologies would not be enough to model and represent the dynamics of the real-world scene, so it is difficult to satisfy all environments using a generic model. Appropriate training data from false-alarm and/or missed cases can address this limitation but is rarely available due to legal issues relating to the privacy of personal data and the unpredictability of new incoming data. In this paper, we propose a novel end-to-end hybrid video surveillance architecture for reliable object detection, consisting of front-end and back-end intelligence. For the intelligent front-end, we propose a new object detector with a Multi-scale ResBlock scheme to consider the scalability and flexibility of the system. We are also developing a new domain adaptation method to replace the generic model with each camera&#x2019;s individual personal model by understanding real-time space and context information for intelligent back-end architecture. It is an iterative and continuous process in which new upcoming data and previous models are consistently engaged in a continuous improvement process. We conducted a series of experiments, including an interesting proof-of-concept tests called the <i>Chameleon</i> project, which demonstrated the high accuracy and versatility of the new architecture, while producing robust results that can be implemented in practice.

A New Intelligent Video Surveillance Architecture

Aimed at the old mode of traditional video surveillance for manual operation and forensics, and the fact that massive amounts of video data occupy a large amount of storage and transmission resources, an intelligent video surveillance architecture based on network and big data is proposed. The architecture includes three core modules: face real-time detecting, face intelligent coding, and face fast retrieval, which can automatically detect human faces, and then intelligently encode them and compare them with databases quickly. It greatly reduces the space occupied by video data storage and transmission, improves the intelligent level of video surveillance, and transforms “after-the-fact analysis” to “in-the-fact analysis” to meet modern criminal investigation requirements.

Scalable and flexible wireless distributed architecture for intelligent video surveillance systems

This paper presents a novel distributed intelligent video surveillance architecture based on Wireless Multimedia Sensor Networks (WMSNs). This architecture is part of a video surveillance project and has been built using the Robot Operating System (ROS). ROS allows to develop and connect (through wireless TCP-IP) several modules to process and manage multimedia data in an easy way. The real-time intelligent surveillance system has been trained for detecting, tracking and monitoring people and vehicles in an indoor-outdoor real environment. The test process shows the reliability of the developed system as a tool for the identification of security incidents. Besides, using wireless connections and a distributed architecture together, we have achieved a really flexible, easy to install and lower-maintenance system that supports many different devices. Thus, the proposed architecture can be applied in distributed locations such as smart cities.

Video surveillance system based on a scalable application-oriented architecture

Nowadays, colossal numbers of facilities all over the world are protected from various types of threats by video surveillance cameras. Video surveillance systems are in urgent and increasing demand to ensure people, things and places security. These networks represent a huge amount of video to be transmitted, viewed and analyzed. However, current surveillance methodologies become increasingly ineffective as the number of cameras grows. To remedy this issue, we propose appending a pre-analysis stage to the video surveillance system. This stage extracts, compresses and sends only the information of interest from the captured video data. Through this paper, scalability, which has always been treated as a processing tool in video surveillance, is handled otherwise to generate more efficient and adaptable video surveillance systems. The pre-analysis phase works in a scalable way. It narrows down the analyzed data by extracting just the useful information depending on the final desired application. The originality of this work lies in the combination of the pre-analysis and the scalability to generate a progressive scalable video surveillance architecture responding to the end user needed application. The principal contributions mentioned in this paper are: i) A framework for a scalable application-oriented architecture for video surveillance systems; ii) A region of interest simplification method aiming to filter in a spatio-temporal way the captured data; iii) A region of interest modeling technique which simplifies the moving objects' representation in the image plane. The evaluation step shows promising results and demonstrates the effectiveness of the suggested architecture.

A flexible and scalable component-based system architecture for video surveillance as a service, running on infrastructure as a service

There are many proposals for moving traditional video surveillance systems into the cloud, commonly known as Video Surveillance as a Service (VSaaS). Most systems use Hadoop technology for storing ...

Communication Mechanisms and Middleware for Distributed Video Surveillance

A new generation of advanced surveillance systems is being conceived as a collection of multisensor components such as video, audio, and mobile robots interacting in a cooperating manner to enhance situation awareness capabilities to assist surveillance personnel. The prominent issues that these systems face are the improvement of existing intelligent video surveillance systems, the inclusion of wireless networks, the use of low power sensors, the design architecture, the communication between different components, the fusion of data emerging from different type of sensors, the location of personnel (providers and consumers), and the scalability of the system. This paper focuses on the aspects pertaining to real-time distributed architecture and scalability. For example, to meet real-time requirements, these systems need to process data streams in concurrent environments, designed by taking into account scheduling and synchronization. This paper proposes a framework for the design of visual surveillance systems based on components derived from the principles of real-time networks/data-oriented requirements implementation scheme. It also proposes the implementation of these components using the well-known middleware technology common object request broker architecture. Results using this architecture for video surveillance are presented through an implemented prototype.