<![CDATA[Product sensory specialists find themselves adrift, lacking clear guidance on how to instruct sensory panelists in the art of evaluating product color and conducting color sensory tests designed to mitigate any color biases that might skew perceptions. To address this gap, Image Processing (IP) and Computer Vision Systems (CVS) were proposed to deliver objective assessment of color quality through the versatile RGB color space. In this study, a 3-level, 4-factor central composite response surface methodology (RSM) design was employed to study how various extrusion cooking factors influence RGB color profile of extruded cornmeal. The process examined four key factors with their levels: screw speed (SS: 100-120rpm), barrel temperature (BT: 170-190℃), feed rate (FR: 40-60rpm), and moisture content (MC: 20-25%). For each trial run, extrudate samples were randomly selected and digitally captured using HD camera app of an Android phone, meticulously set to ISO 200, a resolution of 2160×2160, and a zoom of 4.0X for optimal clarity. IP and CVS tools in MATLAB 2023a application software were used for RGB color space data extraction. Regression equations were developed for each response as a function of the process factors. Results showed increasing SS increases R, G, and B intensities. However, increasing BT decreases B and R intensities, whereas decreasing FR did not affect R intensity. Key statistical metrics such as R-squared and Adequate precision, provided insight into the models’ performance: the R, G, and B R-squares/Adequate Precision were (0.82, 0.73, and 0.72) / (5.51, 3.70, and 3.17) and were found satisfactory.]]>
Copyrights © 2026
