Special Issue(24-09):Deep Learning for 3D Object Recognition in Smart City Surveillance Systems Posted on 2024-08-02

Technology is the only way to deal with this predicted increase in demand, and privacy and urbanisation challenges are predicted to develop in order to maximise current resources. The goal of the smart city is to seamlessly integrate ICT (information and communication technology) with cutting-edge developments, such as networked homes and appliances. A smart city's efficient architecture and improved security improve the quality of life for its citizens. When carried out over an extended period of time, surveillance is a repetitive and tedious task that diminishes the effectiveness of natural defenders. Although numerous sites have adopted video surveillance for 3D object recognition, it is now necessary to continuously watch an individual camera operator across several cameras in order to monitor anomalous behaviour. This is a laborious operation. It is challenging to recognise various types of firearms and knives in real-time settings on multidimensional devices and to classify them apart from other video surveillance items.

The majority of detecting cameras are low-resource, computationally-capable sensors. Surveillance is a crucial component of smart cities, both to ensure public safety and to discourage criminal activity. Thus, frameworks for intelligent video surveillance (IVS) are becoming more and more well-known in safety applications overall. The most common methods for improving the quality of images and videos for smart city surveillance are provided, along with the key functions and statistics. Smart city applications include things like transportation, social security, and service-related smart monitoring. They also go over the most important recently developed deep learning methods for monitoring and managing smart cities. The rise of the smart city has led to an increasing demand for adaptable deep learning techniques that assess human behaviour and infrastructure usage to simulate social behaviours pertinent to flexible and ecological buildings. In order to better understand the way humans utilise facilities, particularly social structures, the cyber-physical system (CPS) frameworks that are used for tracking and automating facilities in smart cities can be expanded to detect people. In order to create a cyber-physical social system (CPSS) for smart cities, it uses convolutional neural network (CNN) topologies to automate 3D object recognition and topographical tracking of individuals using camera data.

In the present day, dynamic surveillance systems aim for high-speed streaming, and object recognition in real-time visual data has proven difficult to do within a tolerable latency. The special issue is categorised according to smart city design, methodologies, software, and historical data, and this allows the article to provide a thorough review of object recognition deep learning topologies.

List of Interested Topics

  • Deep Learning-Based IoT-Based Smart City Management Network
  • Leveraging 3D object recognition for Proactive Surveillance in Smart Cities
  • AI and deep learning-based object recognition for surveillance footage systems
  • An effective foundation for IoT-based smart city monitoring utilising object recognition
  • Real-time anomalous object recognition for smart city surveillance cameras
  • Utilising deep learning to implement region-based video monitoring in smart cities
  • Deep learning-driven computer vision for monitoring in its Assessment of cutting-edge techniques
  • Regarding public street contexts, quick item detection utilising dimensions based characteristics
  • Deep learning applications in smart cities: current state, classification, and obstacles
  • Spatiotemporal Deep Learning for Object Recognition of Anomalies in Intelligent and Smart Cities
  • An ideal deep learning-based intelligent surveillance system for object recognition

Guest Editor Information

Dr.Muhammad Asim Saleem 

Dr. Uzair Aslam Bhatti

Dr.Zahid Ullah

Important Dates

  • Deadline for submissions will be due on March 20, 2025
  • The preliminary notification will be due on Feb 15, 2025
  • Revision will be due on Apr 20, 2025
  • Final notification will be due on July 20, 2025

The final publication process will be followed based on journal direction