Microscopy-Assisted Digital Image Analysis with Trainable Weka Segmentation (TWS) for Emulsion Droplet Size Determination

Abstract

The size distribution of droplets in emulsions is very important for adjusting the effects of many indices on their quality. In addition to other methods for the determination of the size distribution of droplets, the usage of machine learning during microscopic analyses can enhance the reliability of the measurements and decrease the measurement cost at the same time. Considering its role in emulsion characteristics, in this study, the droplet size distributions of emulsions prepared with different oil/water phase ratios and homogenization times were measured with both a microscopy-assisted digital image analysis technique and a well-known laser diffraction method. The relationships between the droplet size and the physical properties of emulsions (turbidity and viscosity) were also investigated. The results showed that microscopic measurements yielded slightly higher values for the D(90), D[3,2], and D[4,3] of emulsions compared to the laser diffraction method for all oil/water phase ratios. When using this method, the droplet size had a meaningful correlation with the turbidity and viscosity values of emulsions at different oil/water phase ratios. From this point of view, the usage of the optical microscopy method with machine learning can be useful for the determination of the size distribution in emulsions.