AN OPTICAL METHOD AND NEURAL-NETWORK FOR SURFACE-ROUGHNESS MEASUREMENT

The measurement of surface roughness using the stylus equipment has se veral disadvantages. A non-contact optical method becomes demanding fo r measuring the surface roughness of the engineering metals with impro ved accuracy. One of the candidates for the optical method is the use of a laser source in which case the reflected laser light intensity fr om the surface may represent the surface roughness of the illuminated alien. Consequently, a relation can be developed between the reflected laser beam intensity and the surface roughness of the metals. The pre sent study examines the measurement of the surface roughness of the st ainless-steel samples using a He-Ne laser beam. In the measurement, a Gaussian curve parameter of a Gaussian function approximating the peak of the reflected intensity is measured with a fast response photodete ctor. To achieve this, an experimental setup is designed and realized. In the experimental apparatus, fiber-optic cables are used to collect the reflected beam from the surface. The output of the fiber-optic sy stem is fed to a backpropagation neural network to classify the result ing surface profile and predict the surface roughness value. The resul ts obtained from the present study is, then, compared with the stylus measurement results. It is found that the resolution of the surface te xture improves considerably in the case of optical method and the neur al network developed for this purpose can classify the surface texture according to the control charts developed mathematically. (C) 1998 El sevier Science Ltd.