Maize breeders improve germplasm for grain traits that enhance industrial and food products. The maize kernel is composed of endosperm, embryo, pedicel and pericarp. The different weight proportions of these components and of vitreous vs. soft endosperm, which affect grain characteristics and their potential use, are commonly estimated subjectively. Our objective was to quantify the relative proportion of these kernel components in maize commercial hybrids grown in different environments using digital image processing. Grain samples of six hybrids were collected by hand at harvest in replicated plots at five environments. A total of seven dried kernels per sample were cut following a perpendicular line through half of the embryo. Digital images were taking to internal grain surfaces with a high-resolution digital camera. Images were then processed with the Image Processing Toolbox of MATLAB. Proportions of each grain components were estimated with a pixel-counting algorithm based on a red-green-blue and hue-saturation-value color systems that allowed clear separation of hard endosperm and embryo. The estimates obtained using this technique were highly correlated with the quantification estimated using texture by dissection. Therefore, this technique could assist breeders and maize processors in quickly assess the relative amounts of maize grain components.