International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 02 | Feb 2024
p-ISSN: 2395-0072
www.irjet.net
Reinforcing Security: ML and Deep Learning Integration in Smart CCTV for Sensitive Zones Sagar Rajebhosale1, Rajeshwari Dandage2, Shravan Jadhav3, Anushka Jawalkar4, Adnan Mulla5, Anjali Yadav6 1Professor, Dept of Computer Engineering, Keystone School of Engineering, Maharashtra, India 2
Professor, Dept of Computer Engineering, Keystone School of Engineering, Maharashtra, India BE student, Dept of Computer Engineering, Keystone School of Engineering, Maharashtra, India 4 BE student, Dept of Computer Engineering, Keystone School of Engineering, Maharashtra, India 5 BE student, Dept of Computer Engineering, Keystone School of Engineering, Maharashtra, India 6 BE student, Dept of Computer Engineering, Keystone School of Engineering, Maharashtra, India ---------------------------------------------------------------------***-------------------------------------------------------------------responsive system capable of safeguarding various sectors Abstract - In response to heightened security concerns, 3
and improving public safety.
safeguarding sensitive locations is now a top priority. This project introduces an advanced surveillance system utilizing cutting edge technologies like AI, IoT, CC, and ML to ensure comprehensive situational awareness and bolster security measures. By deploying a network of IoT devices equipped with high resolution cameras and sensors, the system continuously gathers data. Leveraging Cloud Computing, this data undergoes efficient processing and secure storage, enabling scalability and real-time analysis. At its core, AI and ML algorithms are meticulously trained to detect anomalies, identify suspicious activities, and recognize potential threats. Through deep learning techniques, the system adapts to changing scenarios, reducing false alarms, and optimizing resource allocation for a more effective response. Emphasizing privacy, the project implements advanced encryption and compliance measures, ensuring that sensitive data remains confidential and accessible only to authorized personnel
2.SYSTEM ARCHITECTURE The system architecture begins with a live stream of CCTV video, capturing the real-time activities in each environment. This video stream is the input for the system, which then processes the incoming frames to extract valuable information. The first step in this process is pre-processing, where the raw images are refined and prepared for further analysis. The pre-processed images are then subjected to feature extraction, a crucial step in identifying unique characteristics that distinguish one face from another. This is where FaceNet, a face recognition system, comes into play. FaceNet utilizes advanced technologies like OpenCV, TensorFlow, and Convolutional Neural Networks (CNN) to analyze and extract distinctive features from each face in the images. The extracted features are then fed into a classifier, a component responsible for determining whether the face in the image is known or unknown. The system relies on a Face Database to compare the extracted features with previously stored information, enabling it to recognize familiar faces.
Keyword’s: Real-time Monitoring, Security Infrastructure, Intelligent Security, Threat Detection, Facial Recognition
1.INTRODUCTION
If the classifier successfully identifies a known person, the system generates an event record and stores relevant data. This data includes information about the recognized individual, creating a comprehensive record of events. The system also allows for the review of past data, providing users with insights into historical patterns and occurrences.
The Eye Spy project is a concept that harnesses the power of existing Closed-Circuit Television (CCTV) infrastructure to create an intelligent and proactive surveillance system. This solution involves enhancing traditional CCTV systems with advanced computer vision, machine learning, and automation capabilities to improve security, situational awareness, and efficiency. The primary objective of this project is to transform passive CCTV systems into proactive and intelligent surveillance networks, capable of identifying and responding to security threats and critical events in realtime. The Eye Spy project repurposes existing CCTV infrastructure, making it a cost-effective and efficient solution for improving security and situational awareness. By integrating artificial intelligence and automation, it transforms passive surveillance into an active and
© 2024, IRJET
|
Impact Factor value: 8.226
On the other hand, if the classifier fails to identify the person, indicating that the face is unknown, the system takes specific actions. It notifies the administrator about the presence of an unrecognized individual, generating an event record and storing pertinent data for future reference. The entire recognition system operates seamlessly, utilizing the capabilities of FaceNet, OpenCV, Tensorflow, and CNN to ensure efficient and accurate face recognition. This system not only enhances security by identifying known individuals
|
ISO 9001:2008 Certified Journal
|
Page 639