The moisture content of grain is one of the most important factors determining quality, grade, and price and affecting the physical appearance such as the morphology and color of grain kernels. The objective of this study was to evaluate the correlation between moisture content and image characteristics of corn kernels. A simple and inexpensive machine vision system was set up to obtain images and weights of individual kernels of four varieties of corn. The corn kernels were dried from 32% to 12% (wet basis) moisture at low temperature. A total of 23 morphological and 24 color features were extracted from the individual kernel images. Four optimized feature sets (morphological, color, morphological-color, and weight-morphological-color) were used to fit moisture content using a stepwise linear regression. The coefficient of determination (R²) of the linear regression improved considerably with the combination of morphology, color, and weight. The best results, with an R² of greater than 0.96, were achieved using the optimized weight-morphological-color set for all corn varieties. The results indicate that the machine vision technique proposed in this study has the potential for rapid measurement of grain moisture content in the industry.