Deep Learning-Based Detection of Shoplifting Behavior: using 3DCNN and LRCN

Chinthulla, Anish (2025) Deep Learning-Based Detection of Shoplifting Behavior: using 3DCNN and LRCN. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (2MB) | Preview
Preview	PDF (Configuration Manual) Download (801kB) | Preview

Abstract

This research explores the effectiveness of deep learning models in detecting pre-crime behavior in surveillance footage, specifically targeting shoplifting scenarios. Using 94 videos (50 shoplifting and 44 normal) from the UCF-Crime dataset, the study employs the Pre-Crime Behavior (PCB) methodology to segment video clips into four distinct behavior classes: Normal, Suspicious, Theft and Post-Theft. Two architectures, Long-Term Recurrent Convolutional Network (LRCN) and 3D Convolutional Neural Network (3DCNN) were trained on input sequences of 120 grayscale frames (64×64 resolution). Three experiments were conducted: multi-class classification, binary classification, and data augmented binary classification to address class imbalance. In the best performing setup, LRCN achieved 90.15% accuracy and an F1-score of 0.90, outperforming 3DCNN which reached 83.33% accuracy and an F1- score of 0.84. The results underscore the benefits of combining temporal modeling with balanced training data for early and accurate detection of suspicious activity. These findings support the development of intelligent surveillance systems capable of identifying abnormal behavior before crimes occur.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Niculescu, Hamilton UNSPECIFIED
Subjects:	H Social Sciences > HV Social pathology. Social and public welfare > Criminology > Crimes and Offences Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science H Social Sciences > HD Industries. Land use. Labor > Specific Industries > Retail Industry
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	18 Nov 2025 17:24
Last Modified:	18 Nov 2025 17:24
URI:	https://norma.ncirl.ie/id/eprint/8941

Actions (login required)

View Item