Faculty of Engineering Universitas Brawijaya (FTUB) has added a new Doctorate in Mechanical Engineering Materials and Manufacturing, Dr. Raden Arief Setyawan, ST., M.T.
The Doctor who is also a Lecturer in Electrical Engineering received his Doctorate degree after successfully defending his dissertation entitled Surface Roughness Prediction with Stereo Gray Level Co-Occurance Matrix Algorithm.
With promoters consisting of Prof. Dr. Ir. Rudy Soeoko, M.Eng. Sc., Prof. Dr. eng. Moch. Agus Choiron, S.T, M.T., and Dr. eng. Panca Mudjirahardjo, S.T., M.T.
The open examination of his dissertation was held in the Main Meeting Room, 2nd floor, Engine 1 building, on Tuesday, July 12, 2022.
That morning also attended to witness this open exam, UB’s Vice Chancellor for Academic Affairs Prof. Dr. Aulanni’am drh DES, UB University Secretary Dr. Ir Setyono Yudo Tyasmoro MS, FTUB Dean Prof. Ir Hadi Suyono ST MT PhD IPU ASEAN Eng., along with the Vice Deans, Head and Secretary of the Department as well as Electrical Engineering lecturers, and all invitees.
In his presentation, he explained that surface roughness is one of the important parameters in determining the quality of production.
A common method for measuring surface roughness is with a diamond stylus that follows the contours of the surface to produce data that varies according to the height of the surface profile.
This process produces accurate measurements but is not suitable for high-volume applications as it requires direct physical contact. In addition, the contact measurement method cannot avoid scratches on the specimen surface.
The development of digital camera technology makes it possible to obtain surface images with various digital image qualities.
“In this study, a new method for measuring surface roughness is proposed using a stereo camera with a stereo gray level co-occurrence matrix algorithm,” he said.
The object to be measured, he continued, is imaged using two cameras that have been designed telecentrically.
To perform the test, the standard test object of roughness ISR – CS130 is used. The two images obtained are then pre-processed to reduce the noise that occurs.
The camera image was divided into 10 pieces to become a sample data set. Furthermore, the stereo matching process is carried out to obtain the same point between the left and right images to obtain the disparity value using the SIFT algorithm.
This disparity value is converted into the distance to produce a surface depth value.
The image obtained is then processed using the GLCM stereo algorithm to compare the texture pattern values between the two images. After getting the GLCM matrix, it is calculated with the initial image to obtain a 3-dimensional image.
The result of this process is then the image cutting process is carried out at several points on the y – axis to obtain surface contours which will be converted into surface roughness values.
To perform calibration, the highest and lowest values on the surface contour are converted to reference values measured by means of a stylus.
Based on the test results, it is found that the method offered has been able to produce a surface roughness value with an error rate of 17% compared to the results of measurements using a stylus.
“The measurement process requires an average computation time of 2.23 seconds to generate the measured value. In the experiment, it was shown that the effect of light could be minimized by using the offered algorithm,” he concluded.
The final examination session for this dissertation was carried out by the Examiner Board chaired by Prof. Dr.Eng., Anindito Purnowidodo, S.T., M.Eng. While sitting on the Board of Examiners, among others, Prof. Dr. Ir. Pratikto, MMT., Dr. Ir. Achmad As’ad Sonief, M.T., and Guest Examiner Prof. Dr. Eng Ir Abraham Lomi MSEE.