Dr. Erica Walker has a diverse background in visual communications which includes feature film production, web design & development, print buying, marketing, and graphic design for print. As a faculty member in the Department of Graphic Communications in the College of Business at Clemson University, Walker teaches courses in Photography, Video, Web Development, and Entrepreneurship. Walker is interested in many areas of research including software application, curriculum development, the entrepreneurial mindset, and artificial intelligence applications in color management for brands.

Dye Sublimation: An Image Quality Comparison Between Precoated and Self-Coated Non-traditional Substrates

Dr. Amanda Bridges and Dr. Erica Walker, Clemson University

Dye sublimation, also referred to as heat transfer printing, is frequently used to transfer designs onto plastics, polymers, and textiles and has also been used to print on metal packaging (Turner,1998). The process involves drafting the design, printing it onto transfer paper using special sublimation inks, preparing the substrate for printing, and applying the heat press to the substrate with the affixed transfer paper. Once the sublimation occurs (solid ink becomes gas), the material becomes permeated with the gas, the paper carrier is removed, and the image adheres to the substrate (Digital Technology Group, n.d.). Some advantages of dye sublimation printing are full color with a single pass unlike traditional screen printing, flexography, and lithography, high quality prints, print durability, versatility, and ease of replication and use (Apo, 2020).

In order for ink to adhere during the transfer process, the substrate must contain polyester or a polymer coating. These non-paper substrates, often referred to as sublimation blanks, can be purchased precoated or coating can be applied manually to any uncoated, non-plastic substrate. There are many precoated substrates available however, for more non-traditional surfaces such as wood or stone tile, self-coating may be required. Sublimation coatings are available in both liquid and aerosol forms with liquid being most common for hard, inflexible substrates. Building on a previous study examining the durability of color when exposed to environmental conditions such as humidity and UV full-spectrum light (Walker & Bridges, 2022), this study will focus on different brands of liquid coatings and examine application processes to ensure even coverage across substrates.

This study will examine precoated substrates as well as three additional self-coated products to determine which substrate and coating provides the best image transfer across different substrates. Previous research has revealed that some coatings increase challenges with burning, paper sticking, and consistent ink transfer (Engwall, 2022). The substrates included in this study are wood, glass, metal, and tile. A different sublimation coating will be used for each self-coated substrate. A color target will be printed on all substrates and a spectrophotometer will be used to record LAB and density readings. Once all color readings are gathered, a comparison of each substrate and coating will be made. For example, we will compare wood to wood, tile to tile, and so on across the different coatings and the precoated blanks. One primary goal of this research is to examine which coating, including the precoated substrates, provides the best color fidelity, range, and density.

In the dye sublimation process, four main factors impact the quality of the reproduction: ink quantity, temperature, pressure, and time (Wu & Bai, 2011). For this study we will use Epson T49M UltraChrome DS inks and DS Transfer Multiuse paper for all the samples and determine the best settings for temperature, pressure, and time to produce acceptable image transfer and the widest color gamut.

On the Game Day Sideline: A Case Study of Web-based AI Color Correction for Social Media Content

Dr. Erica Walker, Clemson University

Achieving consistent brand color is challenging in a fast-paced athletics game day environment. When consumers look at branded content across many different devices and channels, they expect to see their favorite team with accurate brand color representation (Budelmann et al.,2010; Chang & Lin, 2010). However, there are challenges on a game day. Content creators work in small teams on the sideline during events trying to capture the most memorable moments, edit, and post both photographs and video clips to multiple social media channels in as close to real time as possible. Often, the social media content team includes student interns with limited experience and content is created on several camera models from multiple manufacturers. For example, a standard kit for Clemson Athletics includes: Five different camera bodies: SonyA7III, SonyA7SII, SonyA7SIII, Canon 5D Mark IV, and Sony FX3 and three Sony lenses 24-70 mm, 70-200 mm, and 100-400 mm and two Canon lenses 24-70 mm and 100-200 mm. Each iteration of these camera and lens combinations produces different color representation in the resulting content (Golabkesh & Walker, 2022).

Previous studies conducted at Clemson University revealed that approximately half of the fans participating in the study recognized when brand colors are off specification on the jumbotron, and even more commented on the inaccuracies when they were asked specifically about the color representation (Conti & Walker, 2019). ColorNet is a patented artificial intelligence algorithm that is capable of addressing this issue by detecting and correcting a specified brand color pixel-by-pixel in a live video feed (Walker, E.B., Smith, D.H., Lineberger, J.P., Mayer, M.,Mayes, E., & Sanborne,A., 2020; Walker, E.B., Smith, D.H., Mayes, E., Lineberger, J.P., Mayer,M., & Sanborne,A., 2020). ColorNet was initially developed by a team of researchers and undergraduate students to ensure brand color consistency for video displayed on the jumbotron at live sporting events but with further adjustments, it could be applied in other contexts.

A recent study applied ColorNet in a social media context to see if it could also improve color consistency on real-world social media content across channels, platforms, and content types finding a statistically relevant improvement in brand color representation (Smith & Walker,2022). That investigation analyzed brand color and skin tones on content that had been posted on Twitter and Instagram over the past ten years.

This study builds upon previous work to propose the development and implementation of a web-based ColorNet application that could be used in a real world, game day situation with Clemson Athletics content creators as a case study. We will begin by investigating the level of color tolerance for brand color with the content creation team to understand a baseline of visual preferences at the creator level.This will help determine the level of need and value placed on the adoption of this tool into their workflow. In addition, the application will need easy, front end usability to address the fast-paced expectations in that environment and an optimized back end that delivers speedy and accurate results even in slow internet circumstances like those commonly found during an athletic event. Mixed methods analysis will include individual feedback from content creators using the app in their process and analysis of the brand color consistency across the resulting social media posts to measure the resulting ColorNet adjustments.