Unauthorized Person Detector using UAV-Patrolling Drone.

In this project, an approach for real-time implementation of Un-Authorized Person Detection using YOLO Algorithm is presented. In real-time system where computational time is very big parameter and it effects the efficiency of the system, Latest version of YOLO implementation on an UAV-Autonomous Dr

Project Title

Unauthorized Person Detector using UAV-Patrolling Drone.

Project Area of Specialization

Artificial Intelligence

Project Summary

In this project, an approach for real-time implementation of Un-Authorized Person Detection using YOLO Algorithm is presented. In real-time system where computational time is very big parameter and it effects the efficiency of the system, Latest version of YOLO implementation on an UAV-Autonomous Drone for Un-Authorized Person Detection is a challenging task. We’ll develop an approach which can be implemented in real-time high FPS (Frames Per Seconds) surveillance videos captured by the autonomous drone to detect human violent activities that is happening because of any Un-authorized person using the latest YOLO Algorithm Version based on Computer Vision(AI). Although there is no such project is found in the above research yet, but a similar project was Made in Germany by A. Srivastava, T. Badal, A. Garg, A. Vidyarthi, and R. Singh, [5] that uses SVM (Support Vector Machines) Classifier for Detection and YOLO is more than 20x faster than the SVM and at least as accurate. So, this project will detect Un-Authorized person & Violent activities real fast as compared to the project they’ve made in Germany.

Project Objectives

The main objective of our project is to detect those violent activities that are happening in concerts or in any large gatherings such as political conferences & fetes or in stadiums of which the security guards or CCTVs are unaware with the help of surveillance drones.The no. of cases of violations in public events such as musical concerts, large gatherings has been recorded numerous times and it is very difficult to detect them manually for the security guards or for a random CCTV Camera since their field of view or target area is limited that is why we need the help of Machine Learning (AI) based Autonomous Drone to solve this problem to smartly detect any Unauthorized person.

Project Implementation Method

Model Approach: Agile Methodology is a people-focused, results-focused approach to software development that respects our rapidly changing world it is faster smaller methodology to the milestone. Agile approach has been used to complete this project successfully in a good environment.

YOLO Algorithm: YOLO Stands for “You Only Look Once”, YOLO is a regression algorithm that falls under the class of real-time object detection methods with a multitude of computer vision applications. This algorithm uses a single bounding box regression to identify elements like height, width, center, and object classes. It cornered the market because of its accuracy, demonstrated speed, and ability to detect objects in a single run, surpassing Faster R-CNN, RetinaNet, and Single-Shot MultiBox Detector (SSD). The R-CNN family was too slow. It took longer to find the proposed region for the bounding box, train a model, detect and classify regions, and then check for refined outputs in separate steps. In many tasks, extreme levels of accuracy (as the ones provided by CNNs) are not imperative, so it is reasonable to rely on less accurate but faster-to-train methods. Hence, YOLO’s unprecedented emergence. First, it improves the detection time given that it predicts objects in real-time.

Benefits of the Project

The world is experiencing a rise in both reported and unreported violations nowadays. As a solution to this growing menace, video surveillance may fill the gap of covering un-noticed actions which lead to violence, while also ensuring a safe and secure lifestyle. So far there is no public dataset for violent activity classification using drone surveillance among the fields that gave incorporated this technology includes police work, video categorization, biometrics and human-computer interaction. This project will help in detection real-time violence in musical concerts, political conferences & fete, large gatherings, that is happening due to any Un-authorized person. Although YOLO v5 has been released but for the stable version YOLO v4 (You Only Look Once) Algorithm will be used for detection of any un-authorized person from the real-time video capture by the drone.

Technical Details of Final Deliverable

Details of Final Deliverables

• GUI Based Software Intergration:
  • The software will have a complete dashboard for detecting unauthorised persons and real-time drone surveillance output.
  • The Software can be connected with any drone that is campatible with wifi connectivity or programmable.
• DJI Tello Drone:
  • This drone will completely in control/sync with the software.
  • The is a UAV-Drone-(Auto-pilot + Autonomous).
  • The flight time will depend on the battery specification chosen by the customer/client. 

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Security

Other Industries

Education , IT

Core Technology

Artificial Intelligence(AI)

Other Technologies

Internet of Things (IoT), Augmented & Virtual Reality

Sustainable Development Goals

Quality Education, Industry, Innovation and Infrastructure, Life on Land, Peace and Justice Strong Institutions

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com