This paper describes a practical system that was developed to identify material surfaces in order to obtain its absorption coefficient values without using complicated and time-consuming procedures and expensive physical equipment to get the measurements. The system is divided into two subsystems which are subsystem 1, by extracting material surface images by using the Gray Level Co-occurrence Matrix (GLCM) to produce four Haralick coefficients, which is used by subsystem 2 that uses Generalized Regression Neural Network (GRNN) to classify the surfaces. Four types of material surfaces are captured and a total number of 280 images used as the input. It is concluded that the system for identification of material surfaces using GLCM and GRNN was successfully developed.