Rock surface roughness measurement using CSI ...

Home

Search

Collections

Journals

About

Contact us

My IOPscience

Rock surface roughness measurement using CSI technique and analysis of surface characterization by qualitative and quantitative results

This content has been downloaded from IOPscience. Please scroll down to see the full text. 2016 IOP Conf. Ser.: Earth Environ. Sci. 29 012028 (http://iopscience.iop.org/1755-1315/29/1/012028) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 192.99.241.141 This content was downloaded on 20/09/2016 at 16:39

Please note that terms and conditions apply.

You may also be interested in: Characteristics of laser scanning confocal microscopes for surface texture measurements Jariya Buajarern, Chu-Shik Kang and Jae Wan Kim Contrast source inversion technique for the reconstruction of 3D inhomogeneous materials loaded in a rectangular waveguide Emre Klç, Funda Akleman and Ali Yapar Quantification of phosphorus metabolites in human calf muscle and soft-tissue tumours from localized MR spectra acquired using surface coils V L Doyle, G S Payne, D J Collins et al.

International Symposium on Geophysical Issues IOP Conf. Series: Earth and Environmental Science 29 (2016) 012028

IOP Publishing doi:10.1088/1755-1315/29/1/012028

Rock surface roughness measurement using CSI technique and analysis of surface characterization by qualitative and quantitative results Husneni Mukhtar1, Paul Montgomery1, Gianto1 and K. Susanto2 1

ICube Laboratory, University of Strasbourg-CNRS, 23 rue du Loess, 67037 Strasbourg, France 2 Geophysics, Universitas Padjadjaran, Jl. Raya Jatinangor-Sumedang km.21, Indonesia [email protected] Abstract. In order to develop image processing that is widely used in geo-processing and analysis, we introduce an alternative technique for the characterization of rock samples. The technique that we have used for characterizing inhomogeneous surfaces is based on Coherence Scanning Interferometry (CSI). An optical probe is first used to scan over the depth of the surface roughness of the sample. Then, to analyse the measured fringe data, we use the Five Sample Adaptive method to obtain quantitative results of the surface shape. To analyse the surface roughness parameters, ‫ܪ‬௠௠ and ܴ௤ , a new window resizing analysis technique is employed. The results of the morphology and surface roughness analysis show micron and nano-scale information which is characteristic of each rock type and its history. These could be used for mineral identification and studies in rock movement on different surfaces. Image processing is thus used to define the physical parameters of the rock surface.

1. Introduction The surface analysis of rocks is necessary in order to be able to define their physical parameters, such as for instance the surface roughness that can be measured by microscopy to give information not only about the rock surface texture but also on its porosity. Various techniques that have been used to obtain images of rock surfaces are conventional optical microscopy, electron microscopy such as scanning electron microscopy (SEM), atomic force microscopy (AFM), etc. The results of image measurements using conventional optical microscopy is on the micro-metre scale and for SEM it is up to the nano-scale. CSI, or white light scanning interferometry (WSLI), or Coherence Probe Microscopy (CPM) as it also known, is an increasingly popular method for measuring the roughness and shape of microscopic surfaces. The technique consists of a fast, non-contacting method for measuring surfaces that is increasingly important as a means of examining not only material surfaces but also components such as miniature optical elements, micro-fluidic devices and micro-electro-mechanical systems. It combines the axial resolution of an interferometer and the lateral resolution of a high-magnification

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. Published under licence by IOP Publishing Ltd 1



microscope; commercial instruments are now an integral partt of most reseearch laboraatories [1optical m 4]. Most reaal surfaces arre smooth suurfaces whichh have higheer self-affinitty. They cann be treated as a fractals under well-defined w s spatial limitss, so that witthin these lim mits specificc statistical properties p uppon scale change aare preserved d by the surfface morphoology. In thiss case, they can c be auto--scaled [5,6]. Fractals can be characterized c d by self-affi finity, an exaample of theese in naturee being treess, leaves, cloouds, and rocks. S Self-affinity occurs whenn the fractalls are not exact e copies of themselvves but therre is still preservaation of the general g shapee with more than t one trannsformation and a the needd for a scale reduction r for morpphology invaariance [5,7]. According to Peitgen et e al. [5,8], one o of the moost common methods that is ussed to calculate the fractaal dimensionn is the box counting c metthod or what we call in thhis paper, the winddow resizing technique. 2. Coheerence Scann ning Interfeerometry (CSI) Techniq que CSI is ann important optical o technnique that is now n widely used u in the measurement m t of surface rooughness and micrroscopic surfface shape. CSI C has the advantages a o being rapiid, non-destrructive and applicable of a to manyy different ty ypes of surffaces. The technique t m makes use off a series of white lighht fringes superimpposed on an image of thee sample on the camera target, t the ceentral fringe along the opptical axis correspoonding to thee position of the surfacee. The fringges are scannned over thee whole depth of the sample ssurface by modifying m thee distance beetween the obbjective and the sample. A series of images i is acquiredd with a cameera at regulaar intervals to o give a stackk of XYZ im mages and ussing signal prrocessing along thee z-axis the peak p of the fringe f enveloope is determ mined at eachh pixel and thhus the corresponding height off the surface at each poinnt in the imagge [9]. The systtem developed at ICube is shown inn figure 1, based b on a modified m Leittz-Linnik miicroscope equippedd with two 50× 5 objectivves and an in ncandescent lamp (centeered at a wavvelength of 580 nm), giving a lateral opticcal resolutionn of 0.42 μm with a CCD D camera Bassler-AVA 10000, and usinng a piezo m PI) for verrtical scanninng. The compputer specificcation is Proocessor Intel® ® Xeon® scanner (PIFOC from M 8G. CPU 2.333 GHz RAM

Figuree 1. CSI instrrument developed at ICuube [10]

3. Analyysis Method ds

3.1. Fivve sample addaptive (FSA)) The FSA A method (fr fringe visibiliity method proposed p by Larkin [11]]) is used to determine the fringe envelopee and the su urface heightt. This meth hod detects the t envelopee of fringe signals s by using u five

2



a the opttical axis. Foor each samppling positioon on the envvelope, the aamplitude samplingg positions along value is calculated from f the locaal position and a four neigghbouring saampling posiitions. The calculated c envelopee is given byy: Ͷଶ ‫݊݅ݏ‬ସ ɔ ൌ ሺ ଶ െ ସ ሻଶ െ ሺ ଵ െ ଷ ሻሺ ଷ െ ହ ሻ

(1)

where is the amplitude value on o one samp pling positionn; ɔ is phasee shift due too scanning step; s ଷ is the locall position; ଵ , ଶ , ସ , ହ aree the neighbo oring sampliing positionss. For valuues of ɔ closse to 90o, the equation (1)) can be writtten as: ଵ

ሻ ଶ ൌ ሾሺ ଶ െ ସ ሻଶ ሺ ଵ െ ଷ ሻሺ ଷ െ ହ ሻሿ

(2)

ସ

The ampplitude valuee of the enveelope can be calculatedd by only tw wo multiplicaations and onne square root opeeration. This gives the addvantage in time computation. Howeever, the FSA algorithm m requires the valuees of ɔ to bee close to 90o, for accuratte measuremeent. 3.2. Winndow resizing g technique In order to analyse the t roughnesss data quanttitatively, wee used the window w resiziing techniquue to give the rougghness ampliitude parameters, ‫ܪ‬௠௠ or the peakk-valley roug ghness and ܴ௤ or the rooot mean square (R RMS) value of the roughhness as a function of window size. The roughneess calculatioon is then conducteed on the cell. Thee number of squaress depend on the vaalue of ߜ , where ߜ ൌ ͳǡ ʹǡ ͵ǡ ‫ ڮ‬ǡ ݉‫݀݋‬ሺሺ݊Τʹሻ wherre ݊ is a pix xel size in 1D. 1 This is called c a geom metrical seriies and is written as: a windoow size ൌ ʹఋାଵ squares

(3)

The cell has the shappe of a square, where ߲‫ݔ‬ ‫ ݔ‬ൌ ߲‫ݕ‬. Winndow resizing g calculates the roughneess values at a giveen cell size ovver the wholle of the imagge arithmeticcally.

Figurre 2. Compoosition of one cell consisting of four squares s for ࢾ ൌ ૚ We use tthe formula in i equation (4) ( to calculaate ‫ܪ‬௠௠ and equation (6)) to calculatee ܴ௤ [12] ‫ܪ‬௠௠ ሺߜ ሻ ൌ ሺ௜

ଵ ఋሻሺ௝೘ೌೣ ିଶఋሻ ೘ೌೣ ିଶఋ

௜೘ೌೣ ିఋ ௝೘ೌ ೌೣ ିఋ σ௜ୀଵାఋ σ௝ୀଵଵାఋ ሺ‫ݖ‬௠௔௫ െ ‫ݖ‬௠௜௡ ሻ௜ǡ௝

3

(4)



where ሺ‫ݖ‬௠௔௫ ሻ݅ǡ ݆ ൌ ‫ݔܽܯ‬ሺ‫ݖ‬௡ǡ௠ ሻǡ ݊ ‫ א‬ሾ݅ െ ߜǡ ݅ ൅ ߜሿǡ ݉ ‫ א‬ሾ݆ െ ߜǡ ݆ ൅ ߜሿ ሺ‫ݖ‬௠௜௡ ሻ݅ǡ ݆ ൌ ‫݊݅ܯ‬ሺ‫ݖ‬௡ǡ௠ ሻǡ ݊ ‫ א‬ሾ݅ െ ߜǡ ݅ ൅ ߜሿǡ ݉ ‫ א‬ሾ݆ െ ߜǡ ݆ ൅ ߜሿ ܴ௤ ሺߜሻ ൌ ‫ܪ‬௥௠௦ ሺߜሻ ൌ ሺ௜

ଵ ೘ೌೣ ିଶఋሻሺ௝೘ೌೣ ିଶఋሻ

(5)

௝೘ೌೣ ିఋ ೘ೌೣ ିఋ σ௜௜ୀଵାఋ σ௝ୀଵାఋ ഥሻଶ ‫ۄ‬௜ǡ௝ ට‫ۃ‬ሺ‫ ݖ‬െ ‫ݖ‬

(6)

and ଵ

ଶ

௝ାఋ ሺ‫ݖ‬ ഥሻ௜ǡ௝ ൌ ‫ۄݖۃ‬௜ǡ௝ ൌ ሺଵାଶఋሻమ σଵାఋ ௡ୀ௜ିఋ σ௠ୀ௝ିఋ ‫ݖ‬௡ǡ௠

‫ۃ‬ሺ‫ ݖ‬െ ‫ݖ‬ ഥ൯ ഥሻଶ ‫ۄ‬௜ǡ௝ ൌ ሺଵାଶఋሻమ σ௡ୀ௜ିఋ σ௠ୀ௝ିఋ ൫‫ݖ‬௡ǡ௠ െ ‫ݖ‬

(7)

With: ଵ

(8)

4. Results : Surface Characterization Two types of limestone, marble limestone and hematite limestone, were used as samples in order to test the CSI method. These stones have very fine and smooth surface because of their self-affinity character. A coating treatment process was not necessary for each sample as in the case of SEM measurement, which means that using the CSI method, the samples are not damaged. Results of CSI roughness measurement are presented in size dimension of 1024×1024 [pixel size dx = 0.13 Pm]. 4.1. Results of sample marble limestone Figure 3 shows a gray scale height image in (a) and a 3D altitude distribution in (b). A line profile A-B from (a) is shown in (c).

4


(a)


(b)

(c) Figure 33. Marble lim mestone rougghness meassured by CSII: (a) gray leevel height image i (b) 3D D altitude distributtion (c) line profile p along line A to B from (a) The calcculation resullt of the rougghness distribbution accordding to wind dow size is sh hown in figuure 4 based onn the image in figure 3(c)).

(a) (b) Figure 4. Result off window reesizing analyysis for maarble limestoone surface rock (a) peak-valley roughnesss (b) RMS roughness r 4.2. Ressult of samplee hematite liimestone Figure 5 shows a graay scale heighht image in (a) ( and a 3D altitude distrribution in (b b). A line proofile A-B from (a) is shown in (c).

5


(a)


(b)

(c) Figure 5. 5 Hematite limestone measured m by CSI (a) grayy level imagee (b) altitudee distributionn (c) line profile A-B A from (a) This imaage shows fin ne texture, and a is smooth her than the sample in fig gure 3. The line profile presented p in figuree 5 shows litttle microstructure. The caalculation ressult of the rooughness disttribution acccording to window size is show wn in figure 6 based on thhe image in fiigure 5(a).

(b) (a) Figure 6. 6 Result off window reesizing analy ysis for hem matite limestoone surface rock (a) peak-valley roughnesss (b) RMS roughness r

Conclussion The C CSI techniquue can be used to quanttitatively meeasure the suurface roughhness of rockk sample surfaces. Image proccessing can thus t be usedd to define thhe physical parameters p of the rock suurface. In work, the anallysis of the results of the images from m the CSI tecchnique coulld be used foor mineral future w

6



identification and to study the link between surface roughness and the movement of rocks on different types of surfaces in the study of landslides. References [1]

Lee B S and Strand T C 1990 Profilometry with a coherence scanning microscope. Appl Opt. Vol.29, 3784–3788.

[2]

De Groot P and Deck L 1995 Surface profiling by analysis of white-light interferograms in the spatial frequency domain. J. Modern Opt. Vol.42, 389–401.

[3]

Sheppard C J R, Roy M and Sharma M D 2004 Image formation in low-coherence and confocal interference microscopes. Appl. Opt. Vol.43, 1493–1502.

[4]

J M Coupland and J Lobera 2009 Measurement of steep surfaces using white light interferometry J Strain, Vol.46, 69-78.

[5]

M Prado, L C Lima and R A Simão 2012 Scale laws for AFM image evaluation: potentialities and applications. Current Microscopy Contributions to Advances in Science and Technology, Microscopy Series No.5 Vol.2 (A. Méndez-Vilas, Ed.), Spain: Formatex Research Center C/Zurbarán 1,2o-Officina 1 06002 Badajoz.

[6]

Mandelbrot BB 1983 The fractal geometry of nature. 1st ed. San Francisco, Freeman and Company.

[7]

Mandelbrot BB 2001 Gaussian Self-Affinity and Fractals. 1st ed. New York: Springer.

[8]

Peitgen H et al 1992 Fractals for the classroom, part 1: Introduction to fractals and chaos. 1st ed., New York: Springer.

[9]

Montgomery P, Montaner D, Salzenstein F, Serio B. & Pfeiffer P 2013 Implementation of a fringe visibility based algorithm in coherence scanning interferometry for surface roughness measurement. SPIE, Vol. 8788, 124-126.

[10] E Pecheva, P Montgomery, D Montaner, L Pramatarova 2007, White light scanning interferometry adapted for large area optical analysis of thick and rough hydroxyapatite layers, Langmuir Vol.23, 3912–3918. [11] Kieran G Larkin 1996 Efficient nonlinear algorithm for envelope detection in white light interferometry. J. Opt. Soc. Am. A/ Vol. 13, No. 4, 832-843. [12] KL Apedo et.al 2015 Cement paste surface roughness analysis using coherence scanning interferometry and confocal microscopy. Materials Characterization Vol.100, 108–119.

7