A novel method of measuring and controlling the primary layer quality in a mineral wool production process is presented. The measurement method is based on image acquisition and processing of the mineral wool primary layer structure. As a statistical estimator of the primary layer quality, a ratio of root mean square (RMS) and average intensity in a selected region of the image was used. A regression model of the mineral wool production process control is presented. The final product local density of mineral wool as a function of the primary layer statistical estimators is analysed. The key process parameters are also taken into consideration. The regression model was derived using dimensional analysis and a selection of characteristic numbers. Results give good agreement with a regression coefficient higher than 0.92.