Brand Recognition AI
Posted on Sunday, July 23rd, 2023
Client | Matthew Jerabek |
Professor(s) | Mejdi Eraslan David Lindsay |
Program | Computer Engineering Technology- Computing Science |
Students | Kelsey Philips, Moses Santos, Shivam Patel, Nathan Fan, Dylan Hunter, Salah Griffith-Jones |
Project Description:
Our project displays the development of an Artificial Intelligence (AI) brand recognition software prototype, integrated by machine learning, to provide a means for Algonquin College staff to detect differences from the Algonquin Branding Visual Identity Standard guideline in digital media during production of new official Algonquin College reports and products to ensure brand unity. Branding is an institution’s way to signify to the masses which products belong to them. An institution adheres to strict branding guidelines to ensure customers can instantly recognize a brand and the company that it came from. As brands are designed and produced by people, potential mistakes that misalign with set guidelines may introduce and entirely miss the defining characteristics of their identity.
Our client, Matthew Jerabek, noticed a potential solution during the creation process of these brands through AI. Such an AI would be trained on a dataset following the Algonquin College Visual Identity Standard to allow the Algonquin marketing staff to detect discrepancies contributed by human error. Algonquin College is an institution for higher education located in Ottawa Ontario, which retains thousands of employees and has educated over 20,000 full time students in 2022 alone. An organization produces ample signage, letterheads, and merchandise both as a form of communication and outreach. Outside the walls of the college, most individuals are unable to recognize the signs of Algonquin College without a recognizable and consistent symbol.
This project took the team approximately seven months to complete. The team started with research in machine learning to familiarize ourselves with the tools and development process. The team put a large chunk of their efforts and commitments towards creating multiple models to identify Algonquin brand materials produced by the college to allow the marketing team to proofread their work. In the end, the team gained a ton of knowledge by learning new techniques and technologies as well as experience for how to develop full stack applications.