With the growing emphasis of industrial automation in manufacturing

vision techniques play an important role in many applications. Since different surfaces have different textures

the techniques of texture analysis can be used for the recognition of surfaces. In this paper

a novel non-contacted approach to measure the roughness of machined surfaces based on texture analysis techniques is presented. When using Gabor filters

It is more complex to classify multiple textural images than to distinguish the texture between two images. According to other related paper and our experiments

the surface of a measured specimen can be classified coarsely according to its gray-level variance. Then

the roughness of the surface can be detected using Gabor filters. We present the method of designing the filters and the experiments show better results as well. The approach can detect the surface roughness automatically and quickly. It is invariant to rotation

and has fewer classifiers. Furthermore the cost of the device for implementing the approach is low and the parameters can be set easily. If the system is connected with the control system of a machine

we can realize real-time close looped control of the machining procedure.