Police Surveillance System for Missing Persons

Abstract

The rapid progress in facial recognition technology has broadened its use across multiple sectors, with law enforcement and public safety being among the primary areas of impact. This study presents an innovative framework for identifying criminals and missing persons by integrating two cutting-edge technologies: FaceNet and MTCNN. FaceNet, a deep learning-based model, produces high-dimensional facial embeddings that capture unique facial features consistently across various conditions, while MTCNN performs real-time face detection, isolating facial regions accurately to improve identification precision. The combined application of FaceNet and MTCNN addresses common challenges in facial identification, such as changes in lighting, pose, and expression, providing law enforcement with a robust tool to expedite investigations and locate missing individuals. Through testing on diverse datasets, this study assesses the system\'s effectiveness, focusing on practical applicability and examining ethical concerns, privacy protections, and potential societal impacts. This research contributes to the ongoing discussion on using advanced technologies responsibly to enhance public safety and support law enforcement efforts.

Introduction

Facial recognition technology is transforming industries, especially in public safety and law enforcement. This research presents a system that combines FaceNet and MTCNN to enhance the identification of criminals and missing persons with high accuracy and real-time performance.

Key Technologies

• MTCNN (Multi-task Cascaded Convolutional Networks):
  Used for real-time face detection, isolating facial regions across diverse conditions (angles, lighting, expressions).

• FaceNet:
  Generates high-dimensional facial embeddings that uniquely and consistently represent faces, aiding in accurate matching even under varied conditions.

System Workflow

1. Face Detection: MTCNN isolates facial regions from input images.

2. Feature Extraction: FaceNet produces unique facial vectors (embeddings).

3. Matching: These embeddings are compared against a database using similarity scores.

4. Ethical Safeguards: Secure data storage, access control, and anonymization are integrated to ensure privacy and responsible use.

Benefits

1. High Identification Accuracy: Effective under variable lighting, poses, and facial expressions.

2. Real-time Performance: Speeds up police investigations and helps locate missing persons promptly.

3. Privacy-focused Design: Ethical implementation ensures public trust through secure and responsible data handling.

Educational Applications

• AI provides real-time feedback to students, helping identify strengths and weaknesses.

• Enables personalized learning, but challenges include data privacy, accuracy of AI-generated content, and teacher training.

System Implementation

• Interface: Features a welcome page with navigation to real-time monitoring and a search database.

• Form Entry: Allows input of missing person details.

• Detection & Alert: On identifying a match, the system alerts the user and displays the person's name, age, photo, and last known data.

Evaluation & Results

• Tested across multiple datasets for accuracy, speed, and false-positive rates.

• The system demonstrated strong performance in identifying individuals and ensuring ethical data use.

Conclusion

In summary, combining MTCNN and FaceNet in facial recognition offers a robust solution for identifying criminals and missing persons. By merging real-time face detection with precise feature extraction, this system effectively overcomes challenges related to lighting, angles, and expressions, providing a more dependable tool for identification. Experimental results indicate its effectiveness in various situations, making it a valuable resource for law enforcement. Moreover, ethical guidelines and privacy protections are integrated into the system, promoting responsible usage. This research underscores the important role of advanced technology in boosting public safety and enabling timely, accurate investigations, all while upholding ethical principles and building public trust.

References

[1] Petra G. R. D., “Introduction to Human Age Estimation using Face Images,” Research Papers, Faculty of Materials Science And Technology in Trnava Slovak University of Technology in Bratislava, Slovak University of Technology, 2013. [2] G. Mahalingam and K. Ricanek, “LBP-based Periocular Recognition on Challenging Face datasets,” EURASIP Journal on Image and Video Processing, 2013 [3] P. Thukral, et al., “A Hierarchical Approach For Human Age Estimation,” IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 1529-1532, 2012. [4] M. Bereta, et al., “Local descriptors in application to the aging problem in face recognition,” IEEE Transactions on Pattern Recognition, vol. 46, no. 10, pp. 2634-2646, 2013.

Copyright

Copyright © 2025 Nayana C P, Harshitha P. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.