Surface roughness evaluation of electrical discharge machined surfaces using wavelet transform of speckle line images J. Mahashar Ali a,⇑ , H. Siddhi Jailani a , M. Murugan b a Department of Mechanical Engineering, B S Abdur Rahman Crescent Institute of Science and Technology, Chennai 600 048, India b School of Mechanical Engineering, Vellore Institute of Technology, VIT, Vellore Campus, 632 014, India article info Article history: Received 16 June 2019 Received in revised form 6 August 2019 Accepted 3 September 2019 Available online 10 September 2019 Keywords: Surface roughness Image processing Machine vision system Speckle line images Bi-orthogonal wavelet transform abstract The surfaces produced by electrical discharge machine (EDM) were illuminated using a redline laser source and the speckle line images of the specimens were captured using a CMOS camera. Signal vectors were generated from the speckle line images which are 1  2592 matrices, represented the grey scale intensity of the speckle line images. The image signal vectors were then decomposed using a wavelet transform by key local intensity variation method. The fourth and the fifth levels of the six-level decom- position of bi-orthogonal wavelet were expected to bear the details about the surface roughness. The root mean square (RMS) and variance of the 4 th and the 5 th level decomposition were obtained and were com- pared with the surface roughness parameters roughness average (R a ), arithmetic mean slope (R da ), and root mean square slope (R dq ). RMS and variance of the 4 th level decomposed signals were found to cor- relate well with surface roughness parameters. Ó 2019 Elsevier Ltd. All rights reserved. 1. Introduction Surface roughness is one of the most important parameters to measure and control the quality in manufacturing. The conven- tional method of evaluating the surface roughness using stylus instrument is intrusive and off line and hence it is not suitable for online evaluation. So, there is an increasing need for a reliable, non-contact method of surface roughness evaluation. For the past decade, there is a lot of advancement in the image processing tech- niques in iris and finger print recognition. This has provided the basis for developing image based surface roughness measuring techniques. Specific lay patterns are produced while machining the metal surfaces with processes like turning, planning and milling. But the electrical discharge machining components surface texture is random in nature, which is beneficial in many applications. Fur- thermore, by taking multiple skimming passes, EDM finish quality can become almost mirror-like. In any machining surfaces, three prominent features can be identified: roughness, waviness between the peaks or valleys and form feature. Stylus instrument is widely used to measure these surface variations with higher reli- ability. The main limitation of this process is that the stylus tip radius prevents it from reaching the real bottoms of the valley, thereby filtering the trace of the profile. Also substantial time is consumed in this offline measurement process. Since it is intrusive and touch type, it is not suitable for online measurement. Hence there is a real need for a system of reliable high-speed and non- contact method for surface measurement. Even though many tech- niques are available including the optical technique, the reliability was not attained in the real life manufacturing environments. In this study, an attempt is made to develop a vision based surface classification method for surface roughness measurement of elec- trical discharge machined (EDMed) surfaces. 2. Literature Technological shifts on surface metrology have been docu- mented very well by Jane Jiang et al., [1]. The Nano and micro roughness are formed by fluctuations in the surface characterized by peaks and valleys of varying amplitudes and spacings [2]. 2.1. Roughness parameters The surface texture study is quiet complex. Different machining processes leave different texture. So the characterization of this texture is much significance in engineering. For that a number of surface parameters have to be studied. It can be calculated in either 2D or 3D forms. The surface roughness parameters [3,4] are given https://doi.org/10.1016/j.measurement.2019.107029 0263-2241/Ó 2019 Elsevier Ltd. All rights reserved. ⇑ Corresponding author at: 7, Vallaba Agraharam Street, Triplicane, Chennai, Tamilnadu 600005, India. E-mail address: mahashar@crescent.education (J. Mahashar Ali). Measurement 149 (2020) 107029 Contents lists available at ScienceDirect Measurement journal homepage: www.elsevier.com/locate/measurement