Measurement Of Surface Roughness Using Image Processing Technique

Luei, Hong Keat (2010) Measurement Of Surface Roughness Using Image Processing Technique. Masters thesis, UTeM.

	PDF (24 Pages) Measurement_Of_Surface_Roughness_Using_Image_Processing_Technique_-_24_Pages.pdf - Submitted Version Download (4MB)
	PDF (Full Text) Measurement_Of_Surface_Roughness_Using_Image_Processing_Technique_-_Full_Taxt.pdf - Submitted Version Restricted to Registered users only Download (37MB)

Abstract

Measurement of surface roughness after completion of a machining process is a common procedure undertaken in order to analyse the quality of the end product so that the desired results are determined. Similar process will be repeated if the desired surface roughness quality is not achieved. Currently, contact method is applied and is commonly used where a stylus is drag across the surface of the work piece to obtain its profile. In this method, the work piece needs to be removed and placed on flat surface before measured. Therefore additional time required for setup, calibration, work piece removal and putting is back into machine holder. Time means cost in production. In order to reduce or eliminate non value added time, an alternative technique of measuring or determine surface roughness is required. Machining parameters like the cutting speed and feed will be set for rough cut and finishing cut for both material mild steel and aluminium. For mild steel rough cut speed will be 27000 rpm and feed will be 0.2 mm, while for finishing cut speed will be 30000 rpm and feed is 0.05mm. For aluminium rough cut speed is 62500 rpm with feed 0.5 and finishing cut speed 95000 rpm with feed 0.15mm. This research started with the application of vision camera in capturing work piece's surface images and using image processing technique through MATLAB computing software to calculate the Grey Level Co-occurrence Matrices (GLCM) features such as contrast, entropy, energy, homogeneity and correlations. Collected data will. then be analysing using data mining to extract the data patterns for both material aluminium and mild steel surface. For aluminium, only 4 GLCM features are computed contributing to influence surface roughness that is homogeneity, energy, contrast and entropy. Whereas mild steel all 5 features were computed affect the roughness value. With the aid of latest software, information can be easily extracted and be analyse for research purposes and also be used in manufacturing, therefore this research will lead into development of automated intelligent machine tools.

Item Type:	Thesis (Masters)
Subjects:	T Technology > T Technology (General) T Technology > TA Engineering (General). Civil engineering (General)
Divisions:	Library > Tesis > FKP
Depositing User:	Mr. Thaqif Mohd Isa
Date Deposited:	05 Apr 2013 08:47
Last Modified:	28 May 2015 03:47
URI:	http://eprints.utem.edu.my/id/eprint/7091
Statistic Details:	View Download Statistic

Actions (login required)

View Item