An automated platform for measuring infant formula powder rehydration quality using a collaborative robot integrated with computer vision

• Published in: Journal of food engineering Vol. 383; p. 112229
• Main Authors: Mozafari, Behrad, O'Shea, Norah, Fenelon, Mark, Li, Runjing, Daly, David F.M., Villing, Rudi
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.12.2024
• Subjects: Automated quality inspection; Foam height; Image analysis; Particle detection; Powder rehydration; Sediment height; PSD; Automated quality inspection; CI; Sediment height; ROI; DS; Particle detection; Image analysis; TTS; CV; SOP; Powder rehydration; Foam height; IF; Cobot
• ISSN: 0260-8774
• DOI: 10.1016/j.jfoodeng.2024.112229

Current methods used for testing the rehydration quality of infant formula (IF) are mainly subjective. For a better understanding of rehydration, objective measurements are required. A computer vision (CV) system was synchronized with a collaborative robot (cobot) to automatically estimate foam height, sediment height, and the number of white particles after IF powder rehydration. Two different robotic agitations were used to prepare the mixtures in a commercially available baby bottle. To evaluate the platform, twenty-four stage-1 IF powders were rehydrated. Cobot-captured images were processed by CV algorithms and independently rated by eight participants. The participants' and platform's estimates of foam height, sediment height, and white particles score, respectively, showed agreements of 2.1 mm, 3.4 mm, and 1.7 scores, and correlation coefficients of 0.82, 0.77, and 0.68. The results show that the platform has the potential to enable objective rehydration tests and to monitor changes in visible foam and sediment over time. [Display omitted] •An automated platform for testing infant formula rehydration quality was developed.•Using computer vision, the platform estimated three rehydration attributes.•Platform estimates were compared to estimates from eight human participants.•Reference "white particles" images were digitally generated for participants.•The platform can monitor foam and sediment levels over time.