Thesis Archive

A Customer Profiling System Using Image Processing For Stock Management In Apparel Stores (2013)

DSP-01-1213-02

Calderon, Arianne Cupino, Einer John Dimaapi, Sandra Joyce Saret, Lady Sheila

Abstract:

-Customer profiling is a marketing strategy that helps in identifying and knowing the customers' needs and demands. Customer profiling is based on building common definitions that groups of people will fit into. Through customer profiling, business concepts can be projected, and be assessed if it will meet the market's demands. Also, identifying the target market through its demographics, and buying and spending habits, business owners will be able to project and implement their marketing and advertising efforts more effectively. This paper aims to develop a customer profiling system using image processing to suggest a stock management strategy for an apparel store. LabVIEW was used as the platform for the implementation of the customer profiling system. The Vision Builder for Automated Inspection (VBAI), an image processing toolbox of LabVIEW, was used to implement the image processing. The system that was developed operates offline. For this study, the experimental and reliability tests were done in a controlled environment. Two parameters were put into consideration – the gender identification and the shirt size classification. One hundred test subjects participated in this study, having forty females and sixty males. Considering the said parameters independently, the gender identification, and the shirt size classification have demonstrated a system reliability of 94% and 95% respectively. The results showed that the system worked at a system reliability of 94% for both gender identification and shirt size classification.

Correspondence:

Edwin Sybingco
edwin.sybingco@dlsu.edu.ph