Multiple camera video surveillance for customer behaviour analysis in a retail store

Date of Publication

2015

Document Type

Master's Thesis

Degree Name

Master of Science in Electronics and Communications Engineering

College

Gokongwei College of Engineering

Department/Unit

Electronics and Communications Engineering

Thesis Adviser

Edwin Sybingco

Abstract/Summary

Traditional methods of determining in-store customer behaviour inside a retail store do not actually represent the customer's actual behaviour. Customer information like gender and knowledge of actual customer behaviour is important in making business decisions (called business intelligence). This research presents a cost-efficient business intelligence solution using video analytics by utilizing IP cameras, a Wi-Fi router, and a Core® 2 Quad based desktop computer. Since in-store security video was not available from an actual boutique store, video samples from a class room were used instead. The proposed solution was able to provide information on people count to within 20% of the ground truth. Male and female detection error rates using body silhouettes were 29.67% and 40.51% respectively. Movement tracking and distribution of people inside the room were achieved using heat maps by tracking blob centroid locations through time.

Abstract Format

html

Language

English

Format

Electronic

Accession Number

CDTG006369

Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

Physical Description

1 computer optical disc. 4 3/4 in.

Upload Full Text

wf_no

This document is currently not available here.

Share

COinS