MATLAB Image Processing as a Viable Tool to Study Low Surface Roughness

The purpose of this paper is to present the results of a study comparing an old technique for measuring low surface roughness with a new technique of data acquisition and processing that is potentially cheaper, quicker and more automated. It offers the promise of in-process quality monitoring of surface finish.Since the late 1800s, researchers have investigated the light scattering effects of surface asperities and have developed many interferometry techniques to quantify this phenomenon. Through the use of interferometry, the surface roughness of objects can be very accurately measured and compared. Unlike contact measurement such as profilometers, interferometry is nonintrusive and can take surface measurements at very wide ranges of scale. The drawbacks to this method are the high costs and complexity of data acquisition and analysis equipment. This study attempts to eliminate these drawbacks by developing a single built-in MATLAB function, to simplify data analysis, and a very economically priced digital microscope (less than $200), for data acquisition. This is done by comparing the results of various polishing compounds on the basis of the polished surface results obtained from MATLAB’s IMHIST function to the results of stylus profilometry methods. The study with the MATLAB method is also to be compared to 3D microscopy with a Keyence microscope.With surface roughness being a key component in many manufacturing and tribology applications, the apparent need for accurate, reliable and economical measuring systems is prevalent. However, interferometry is not a cheap or simple process. “Over the last few years, advances in image processing techniques have provided a basis for developing image-based surface roughness measuring techniques” [1]. One popular image processing technique is through the use of MATLAB’s Image Processing Toolbox. This includes an array of functions that can be used to quantify and compare textures of a surface. Some of these include standard deviation, entropy, and histograms of images for further analysis. “These statistics can characterize the texture of an image because they provide information about the local variability of the intensity values of pixels in an image. For example, in areas with smooth texture, the range of values in the neighborhood around a pixel will be a small value; in areas of rough texture, the range will be larger. Similarly, calculating the standard deviation of pixels in a neighborhood can indicate the degree of variability of pixel values in that region” [2].By combining the practices of interferometry with the processing techniques of MATLAB, this fairly new method of roughness measurement proved itself as a very viable and inexpensive technique. This technique should prove to be a very viable means of interferometry at an affordable cost.Copyright © 2015 by ASME