Process control within a coal processing plant requires rapid and reliable intervention to the operation. Ash content has paramount importance for the identification of a particular lignite's quality. Among the methods used for on-line content determination, efforts covering the ash content estimation at the processed particle size ranges and ensuring fast responses are more favorable. The aim of this study is to investigate the applicability of the imaging to estimate the ash content of a particular lignite sample. Experimental study was powered by filtered and unfiltered imaging in visible and near-infrared ranges, determination of the reflectance values distribution of individual particles, and proposing the best linear model to define the proximate ash content by using stepwise multiple linear regression (SMLR) followed by a statistical validation. Results showed that ash content could be estimated with a determination coefficient up to 87% by using the best linear SMLR correlation formed with reflectance values of individual particles.