Development of an Automated Dye Color Mixing System with Computer Vision

Abstract

Objective: The objective of this study is to investigate the development of a small-scale tintometric system prototype based on computer vision, with the aim of offering a cost-effective solution tailored for small and medium-sized enterprises, promoting raw material savings and higher reliability in paint mixtures.   Theoretical Framework: Grounded on industrial automation in the paint sector, computer vision applied to color analysis (RGB/CMYK), artificial intelligence, and low-cost embedded systems (ESP32).   Method: Experimental technological development. A belt conveyor was constructed with dosing pumps and sensors controlled by an ESP32 microcontroller connected via serial interface to a Jupyter Notebook. Data collection involved physical validation of digital inputs and visual feedback. Results and Discussion: Results showed technical feasibility and high precision in RGB/CMYK color analysis, successfully replicating conventional machine functions while reducing human error and waste.   Research Implications: Promotes technological modernization, chemical waste reduction, and automated AI-driven quality control for small and medium paint manufacturers. Integrates accessible computer vision and cheap embedded hardware (ESP32) to establish a closed-loop quality control system in an industry traditionally dependent on subjective human inspection.   Originality/Value: This study contributes to the literature by integrating low-cost computer vision and standard embedded systems (ESP32) to create a closed-loop quality control framework for an industry that has historically relied heavily on subjective human visual inspection. The relevance and value of this research are evidenced by providing an affordable and highly efficient alternative that lowers implementation barriers for small and medium-sized businesses, standardizing the process and increasing final product reliability.