A novel approach to estimate the Surface finish of machined Components using computer vision And soft computing techniques

Abstract

Researchers hitherto used Computer Vision and digital image newlineprocessing for grabbing images of machined surfaces improving their quality newlineby pre processing and then analysed them for evaluation of surface finish newlinewith reasonable success In the conventional mechanical stylus method used newlinefor roughness evaluation many of the fundamental requirements need to be newlinetaken care during measurements which include alignment of components with newlinethe stylus pick up movement tracing length filter cut off length etc newlineInspection of parts by machine vision requires accuracy high speed newlinereliability and consistency Also it is required that images are to be rotated newlinesheared and scaled while preserving image integrity Practical use of newlineComputer Vision for surface roughness estimation faces many challenges as newlineonly image is used for evaluation rather than the components Existing newlineschemes are less suited for real time machine vision applications due to the newlinegreat computational burden involved in processing a large image For newlineexample an operation such as rotation and scaling involves four newlinemultiplications and four additions per pixel which is going to be newlinecomputationally complex This problem is getting worse when the size of the newlineinput image increases which poses a common problem for today s machine newlinevision system newline newline