Vision based prediction of surface roughness for end milling

Abstract

Abstract Measurement of surface roughness helps to assess the machined component's functionality. In the past three decades, several scientists have contributed to the computation of surface roughness. This research article deals with two distinct methods for prediction of surface roughness employing the surface profilometer and machine vision for AISI 1040 steel specimens prepared by varying cutting parameters of end milling viz. feed rates, speed and cutting depth. Using a surface profilometer, the surface roughness parameters are evaluated. At the other hand, the texture features were extracted using a Gray Level Co-occurrence Matrix Algorithm (GLCM) and a computer vision system. Correlations are formed among characteristics of machined surface and the texture feature such as contrast, entropy, energy, and homogeneity. The comparable findings revealed a maximum relative error of −8% using contrast and energy, − 11% using entropy and −10% using homogeneity.