MISSING CHILD IDENTIFICATION USING DEEP LEARNING

Abstract

The issue of missing children has become a critical
social concern, particularly in countries like India
where thousands of cases are reported annually and
a significant number remain unresolved. Traditional
identification methods rely heavily on manual
processes, which are time-consuming, inefficient,
and often inaccurate when dealing with large
datasets and cross-regional cases. This project
proposes an intelligent Missing Child Identification
System using deep learning techniques to improve
the speed and accuracy of locating missing
children. The system utilizes Convolutional Neural
Networks (CNN) for extracting high-level facial
features and integrates machine learning classifiers
such as Support Vector Machines (SVM) or KNearest
Neighbors (KNN) for effective matching. A
centralized web-based platform is developed where
authorities and the public can upload images of
missing or found children. The uploaded images
undergo preprocessing, feature extraction, and
comparison with a stored database of missing child
records. The system is designed to handle
challenges such as variations in lighting, pose, age
progression, and image quality. By enabling public
participation and automating facial recognition, the
system enhances collaboration between citizens
and law enforcement agencies. The proposed
solution significantly reduces identification time,
increases matching accuracy, and improves
recovery rates. Overall, the system offers a
scalable, cost-effective, and reliable approach to
addressing missing child cases using artificial
intelligence and computer vision technologies.