Professional Documents
Culture Documents
Yannick Caulier
Department Process Integrated Inspection Systems
Fraunhofer Institute for Integrated Circuits IIS,
D-91058 Erlangen, Germany
yannick.caulier@iis.fraunhofer.de
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6642
9. J.P. Besl and J. Ramesh, “Three-dimensional object recognition,” ACM Comput. Surv. 17, 75–145 (1985).
10. Gerd Häulser, “Verfahren und Vorrichtung zur Ermittlung der Form oder der Abbildungseigenschaften von
spiegelnden oder transparenter Objekten,” Patent, (1999).
11. A. Williams, “Streifenmuster im spiegelbild,” Inspect Magazine, GIT Verlag GmbH & Co. KG, Darmstadt
(2008).
12. P. Marino, M.A. Dominguez, and M. Alonso, “Machine-vision based detection for sheet metal industries,” in The
25th Annual Conf. of the IEEE Industrial Electronics Society (IECON’1999), 3, 1330–1335 (1999).
13. I. Reindl, and P. O’Leary., “Instrumentation and measurement method for the inspection of peeled steel rods,” in
IEEE Conf. on Instrumentation and Measurement (IMTC’2007), (2007).
14. F. Pernkopf., “3d surface inspection using coupled hmms,” in Proc. of the 17th Int. Conf. on Pattern Recognition
(ICPR’2004), (2004).
15. M. Petz, and R. Tutsch, “Optical 3d measurement of reflecting free form surfaces,” (2002).
16. G. Delcroix, R. Seulin, B. Laalle, P. Gorria, and F. Merienne., “Study of the imaging conditions and processing
for the aspect control of specular surfaces,” Int. Society for Electronic Imaging 10, 196–202 (2001).
17. R. Seulin, F. Merienne, and P. Gorria, “Machine vision system for specular surface inspection: use of simula-
tion process as a tool for design and optimization,” in 5th Int. Conf. on Quality Control by Artificial Vision
(QCAV’2001), (2001).
18. S.K. Nayar, A.C. Sanderson, L.E. Weiss, and D.A. Simon, “Specular surface inspection using structured highlight
and gaussian images,” IEEE Trans. Rob. Autom. 6, 208–218 (1990).
19. F. Puente Leon, and J. Beyerer, “Active vision and sensor fusion for inspection of metallic surfaces,” in Intel-
ligent Robots and Computer Vision XVI: Algorithms, Techniques, Active Vision, and Materials Handling, D.P.
Casasent (ed.), Proc. SPIE 3208, 394–405, (1997).
20. R. Woodham, Y. Iwahori, and R. Barman, “Photometric stereo: Lambertian reflectance and light sources with
unknown direction and strength,” University of British Columbia, Vancouver, BC, Canada, 1991, (1991).
21. J.S. Weska, “A survey of threshold selection techniques,” Comput. Graph. Image Process. 7, 259–265 (1978).
22. Huang Zhi, and Rolf B. Johansson, “Interpretation and classification of fringe patterns,” in 11th Int. Conf. on
Image, Speech and Signal Analysis (IAPR’1992) 3, 105–108 (1992).
23. L. Lepistö, J. Rauhamaa, I. Kunttu, and A. Visa, “Fourier-based object description in defect image retrieval,”
Machine Vision Applications 17, 211–218 (2006).
24. Cem Ünsalan, “Pattern recognition methods for texture analysis case study: Steel surface classification,” Ph.D.
dissertation, University of Hacettepe, Turkey, (1998).
25. D.M. Tsai, and T.Y. Huang, “Automated surface inspection for statistical textures,” Image Vision Comput. 21,
307–323 (2003).
26. H. S. Soon, K. Qian, and A. Asundi, Fringe 2005: Fault detection from temporal unusualness in fringe patterns.
Stuttgart, Germany, (2005).
27. T.M. Cover, and P.E. Hart, “Nearest neighbor pattern classification,” IEEE Trans. Inf. Theory 13, 21–27 (1967).
28. R. Gutierrez-Osuna, “Pattern analysis for machine olfaction: A review,” IEEE Sens. J. 2, 189–202 (2002).
29. R. Kohavi, “A study of cross-validation and bootstrap for accuracy estimation and model selection,” in IJCAI,
1137–1145 (1995).
30. I. H. Witten, and E. Frank, Data mining: Practical machine learning tools and techniques, 2nd ed., ser. The
Morgan Kaufmann series in data management systems. Amsterdam: Morgan Kaufmann/Elsevier, (2008).
1. Introduction
One major goal of computer vision processes is the characterization of industrial (quality con-
trol) or medical (diagnosis) objects using automatic surface inspection methods. In other words,
this research field tackles the processing of different surfaces to be characterized, by means of
different types of illuminations and/or recording techniques, for the visual enhancement of de-
fective surface parts. Common and main requirements of proposed vision methods are their
ability to characterize the surface to be inspected and their rapidity in terms of inspection time.
Thus, effective lightings, and algorithms, but also efficient handling and software frameworks
have to be involved, in order to address more and more challenging tasks, like the recognition
of different types of surface defects in real-time.
In general, the visual enhancement of a certain type of defective surfaces is directly depen-
dent on the lighting and the recording technology. Typically, depth defects related to geometric
deformations of the surface, or textural defects synonymous of different surface roughness have
to be visually enhanced by means of a structural and a diffuse lighting. Different automatic in-
spection systems have been recently proposed, as the measurement of electronic devices by the
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6643
companies Aceris [1] or Comet [2], or the wafer inspection by the company Solvision [3].
With the same objective of increasing different defect types with one system, an alterna-
tive surface inspection procedure has been proposed in [4]. It has been demonstrated that the
characterization of a projected structured pattern serves the direct surfaces interpretation. The
adapted and textural feature-based content description method of the corresponding stripe im-
ages rely on the characterization of the depicted bright/dark structures [5]. The advantages of
such a method are manifold, especially in terms of real-time processing, “simple” algorithmic
procedure, and process simplification, enhancement of different defect types in one camera
shot.
However, in order to simplify the algorithmic processing of the stripe images, periodical and
vertical bright/dark structures have been considered. Hence, the described inspection task in [5]
can be applied to the same approaches, i.e. the inspection of specular and cylindrical surfaces
by means of an adapted illumination technique, or to further inspection tasks necessitating the
interpretation of similar vertical bright/dark structures. Such results are therefore only applica-
ble in case of the characterization of vertical patterns. There are two possible steps toward the
generalization of the proposed inspection method to further non-cylindrical free-form surfaces.
The first possible alternative could be to adapt the structured illumination to the inspected
surface shape so that a periodical vertical pattern is depicted in the recording sensor. This ap-
proach is difficult or even impossible to implement in case of free-formed surfaces, in particular
if these are highly specular which are more difficult to record: the camera does not observe the
surface itself, but the reflection of the light on it. This problem is addressed in detail in [6].
The second possible solution could be to consider the characterization of non-vertical and non-
periodical bright/dark structures which are produced when a light pattern is projected onto
free-form surfaces. This approach is tackled in this paper.
Hence, one major purpose of this paper is to define and to optimize a free-form stripe pat-
tern recognition process, in terms of retrieving the most relevant set of features that accurately
classifies the reference image sets. According to Raudys and Jain [7], the main steps defining a
typical pattern recognition system are the data collection, the pattern class formations, the char-
acteristic feature selections, and the classification algorithm specifications. With the proposed
free-form surface inspection task a successive optimization of these stepwise procedures will
be addressed.
At first, various reference stripe image sets defining the free-form surfaces to be character-
ized will be introduced. Each considered set of patterns will be classified in two formations,
corresponding to two distortion types. Then, two different stripe feature-based image content
description methods will be considered: a method specially adapted for the characterization
of such stripe patterns, and a general textural Fourier-based approach. The optimization of the
feature selection will be addressed by means of specific feature fusion and one optimal fea-
ture selection method. Finally, the determination of the optimal pattern recognition process is
achieved by means of the classification rate.
Hence, on the basis on the previous experiments related to the characterization of cylindrical
specular surfaces, the purposes of this paper are:
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6644
characterization.
The rest of this paper is organized as follows: The use of structured illumination for surface
quality control is introduced in Sec. 2. The two considered image content description methods,
namely the general Fourier and the adapted stripe approaches are presented in Sec. 4. Section 5
describes the involved proposed four steps procedure for the determination of optimal feature
subsets in case of free-form bright/dark structure characterization. Finally, a summary is given
in Sec. 6.
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6645
ΩA (Acceptable) ΩR,3D (Rejected) ΩR,2D (Rejected)
Geometry Grey level
changes changes
Fig. 1. Six examples taken from the reference initial set of images Φ00 made of 252 patterns.
This set has been used for the qualification of the industrial system for the cylindrical spec-
ular surfaces inspection. All the patterns have been classified into distinct classes: Accept-
able ΩA , rejected (non-acceptable) ΩR,3D , and ΩR,2D . All other 246 patterns depict similar
structures, i.e. correspond to similar geometry and/or grey level changes/perturbations.
Fig. 2. (a) Surface inspection principle: A camera C records the object surfaces to be
inspected S, illuminated by a lighting L. (b) Recordings of one free-form rough surface and
one free-form specular surface, both are illuminated by a structured light pattern.
The major difference with the considered images in [4], where regular periodical stripe struc-
tures are observed, is that, as both objects have a non-planar surface, and a conventional planar
structured lighting is used, the bright and dark stripes in the images of Fig. 2(b) are neither
vertical nor periodical. Rather, their geometries depend on the shape of the inspected objects.
As a consequence, with the purpose of generalizing the inspection principle to free-form
rough and specular surfaces, an extensive range of stripe structures have to be considered.
Such stripe geometries can be obtained by means of different surface shapes, structured light
positions or sensor types. Our task is not to enumerate all possible combinations of these com-
ponents and to compute the corresponding stripe geometries. This would hardly be possible.
Hence, it is preferable to focus our investigations on a restricted and predefined number of
non-vertical and non-periodical bright/dark stripe deformations.
Figure 3 depicts two examples of bright/dark geometries. Each geometry can be obtained
when a rough or a specular surface is inspected. The case of a linear moving object with speed
V recorded with a static line-scan camera is considered. One application example for a spherical
object is also shown.
For the purpose of clarity, only the case of surfaces recorded by means of line-scan sensors
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6646
Fig. 3. Left and middle: Two different examples of stripe deformations arising when (a1)
and (b1) free-formed rough or (a2) and (b2) free-formed specular objects are recorded.
These two examples show that depending on the surface geometry, similar stripe deforma-
tions can be observed for rough or specular objects. These two examples show the inspec-
tion of a surface S illuminated by a lighting L and recorded by a line-scan camera C during
its linear movement along V. Right: One possible bright/dark structure example in case of
a round (spherical) object. Upper image shows a sphere with 3 surface portions of different
sizes. Lower images show the corresponding image structures if a light pattern is projected.
has been considered in Fig. 3. However, this does not restrict the application of the proposed
surface inspection principle, as similar patterns can be obtained when matrix cameras are used.
On the basis of these two examples, it can be demonstrated that it is possible to obtain the
same bright/dark structures for specular and rough surfaces.
The underlying assumption is that the disturbances induced by the depicted bright/dark im-
age patterns are always distinguishable from the undisturbed pattern. Thus, as stripe structure
geometry and/or gray level is used for the defect localization and characterization, this means
that in the vicinity of a defective region, the “background” variations, i.e. of the surface geome-
try and/or the surface texture, are below a certain level. The illumination is considered as ideal
and projects a homogeneous bright/dark light structure on all the surfaces to be inspected.
Hence, the same reference stripe patterns can be used for the characterization of rough and
specular surfaces. The considered reference patterns used for retrieving the ideal free-form
stripe pattern recognition process, and the influence of such different bright/dark structures on
the classification rate, will be tackled in the next sections.
As it is not possible to consider all possible stripe disturbance induced by object geometry, for
the rest of the paper, the investigations will be “restricted” to the two types of stripe geometrical
deformations depicted in Fig. 3(a) and 3(b), the perspective and the geometrical ones. Indeed,
Fig. 3(c) shows an example of bright/dark stripe geometry obtained for a round object. It can
be seen how these disturbances encompass, i.e. can be described by the two considered.
Thus, perspective and geometrical disturbances will serve for the generalization of the pro-
posed inspection method based on the direct interpretation of “almost” free-form stripe pat-
terns, i.e. non-vertical and non-periodical ones, for the characterization of free-form surfaces.
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6647
The terminology “almost” is used, as it is assumed that the complex bright/dark structures to be
characterized, must permit the localization and description of all the defects situated completely
on the surface to be inspected.
The limiting factors for this statement is the bright/dark structures disturbance degree. Figure
3(c) clearly shows by means of a concrete example, that the recorded surface size must be
adapted to object geometry. Content description with the proposed method will be more robust
if it is applied to image (1) than to image (3). Thus, in case of spherical objects, it would be
preferable to increase the number of recordings in order to permit a robust inspection of the
whole object.
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6648
Fig. 4. Left: Reference patterns for the classification of free-form rough and specular sur-
faces. These image patterns correspond to three different surface shapes illuminated with
a regular periodical structured illumination: -0- for surfaces inducing no deformations, and
-1- and -2- for surfaces inducing perspective and cylindrical distortions. Φ00 corresponds
to patterns without distortion related to the shape of the object. These patterns have been
measured. Φ41 and Φ42 corresponds to patterns with a maximal perspective distortion of type
-1- and a maximal cylindrical distortion of type -2-. These patterns have been simulated by
transforming patterns Φ00 with perspective and cylindrical distortions. All the patterns have
a size of 64 × 64 pixel. Right: Bright/dark geometry and reflectance characteristics: Sur-
face: Period dL,P and coefficient ρS ; Defect: Size [dD,u × dD,v ] and coefficient ρD .
Defect minimal size and intensity is directly linked to the bright/dark structure deformations
and sensor sensitivity.
In order to be detected, each defect D must be at least as huge as the minimal depicted
bright/dark structure period and have a significant reflectance coefficient. Hence, if dD,u and
dD,v are the defect width and height, rC,u and rC,u the sensor resolution in u− and v− directions,
dL,P the projected light stripe period, ρD and ρS the reflectance coefficients of the defect D and
the neighboring surface S, following equation holds:
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6649
for optimal components spatial arrangements are not tackled here.
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6650
was to use the particularity of the spectral domain by selecting different frequency subbands,
which is equivalent to retaining certain levels of details and directions in the patterns to be
analyzed. Weska considers the radial and the angular spectral distributions, saying that the
former is sensitive to texture coarseness and the latter to texture directionality. He also uses the
distributions corresponding to the principal spectral image axes, the u- and v-directions.
The features are directly computed from amounts of values in the Fourier spectrum for differ-
ent spectral regions. [21] defines various radial, directional, horizontal and vertical frequency
regions. The assumption is here that the use of different spectral regions characterizing different
image frequencies, would be more appropriate for the description of almost free-form patterns,
corresponding to spatial frequency variations of bright/dark structures.
The length of each feature vector depends on the considered frequency regions. The vector
cF,r,θ ,v,u , which considers all possible regions has a maximal length of Nc = 33.
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6651
Pr1 ,r2 = ∑ | fˆ(κ , λ )|2 Pv1 ,v2 = ∑ | fˆ(κ , λ )|2
r12 ≤κ 2 +λ 2 <r22 0<κ <Mu ;v1 <λ <v2
0<κ < M2u ;0<λ < M2v
Fig. 5. Image characterization from amounts of values in the Fourier spectrum, according
to [21]. Four different spectral regions are considered: The radial Pr1 ,r2 , directional Pθ 1 θ 2 ,
horizontal Pv1 ,v2 and vertical Pu1 ,u2 spectral ones.
Table 1. Notation and names of the 20 considered adapted features of the stripe feature
vector cS . These features characterize the bright and the dark stripe depicted in a pattern.
cS (00), cS (01) Bright and dark stripes horizontal deviations
cS (02), cS (03) Bright and dark horizontal minimum distances
cS (04), cS (05) Bright and dark horizontal maximum distances
cS (06), cS (07) Bright and dark intensities
cS (08), cS (09) Bright and dark tangents
cS (10), cS (11) Bright and dark curvatures
cS (12), cS (13) Bright and dark lengths
cS (14), cS (15) Bright and dark shapes
cS (16), cS (17) Bright and dark straightness
cS (18), cS (19) Bright and dark number of elements
Hence, a total of 20 adapted features are computed for each stripe pattern F. The algorithmic
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6652
procedure to retrieve these feature images is described in [5]. The mathematical expression of
related feature vectors of lengths Nc are described in [5].
N
cm = 1
Nw ∑n=1
w
Om m ∈ {02; 08}
Fig. 6. Typical “projected” and “interferometric” bright/dark structures. The three upper
images correspond to ideal “projected” structures, the three lower ones to more complex
“interferometric” fringe structures. The first “vertical stripes” feature group was defined for
the characterization of the former, whereas the second “free-form” feature group was de-
veloped for description of the latter. For illustration purpose, the equations of one adapted
“vertical stripes” “minimum distance” feature c02 and one adapted “free-form stripe” “tan-
gent” feature c08 are listed. The results of corresponding operators are written for one cen-
tral blue marked pixel (B) sic . Both are the average results of operators O02 and O08 applied
to all bright stripes central pixel elements (B) sic of the considered image F.
The major difference between the two feature groups is that for the former group the as-
sumption is made that the stripe structures are vertical, i.e. that the result depends on the main
direction of the structures. For the latter group, the feature value is independent of the stripe
direction, with the assumption that it is more adapted for the characterization of non-vertical
and non-periodical structures.
The following three different feature vectors of lengths Nc will be used:
cS = {cS06 ; cS14 } Nc = 20
cS06 = cS ([00 : 05]): Nc = 06 (3)
cS14 = cS ([06 : 19]): Nc = 14
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6653
The considered feature vector cS encompasses two different feature groups or types, cS06 , cS14 ,
each defined for similar bright/dark image structure characterization tasks. Former group con-
sists of vertical bright/dark structures, whereas the latter of more free-form bright/dark struc-
tures. It therefore seems appropriate, to fuse these two feature groups into one describing vec-
tor, as the considered inspection task consists of describing and characterizing disturbed “pro-
jected” stripe patterns, whose disturbance degrees are in between the disturbance degrees of the
considered bright/dark image groups, see Fig. 4. Thus, we are convinced that such a fused fea-
ture vector should lead to optimal image classification rates, as non-vertical and non-periodical
bright/dark structures must be interpreted.
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6654
ture groups in case of problem generalization, i.e. free-form surface quality control. The third
section addresses the evaluation of FSS methods by considering the further eight reference
databases defined for problem generalization to free-form surfaces. Finally, the last and fourth
section is dedicated to the evaluation, in terms of types and number, of the previously selected
features.
Evaluation criterion for all these investigations is the classification rate R , expressed in per-
cent, which corresponds to the amount of correctly classified patterns for the 3 considered
classes, ΩA , ΩR,3D , and ΩR,2D . These 3 distinct pattern classes and the annotated data set
Φ00 were considered, as these correspond to the qualification requirements of the reference
industrial system [4].
Thus, changing this predefined pattern distinction and/or the reference data set, would have
a direct impact on the results. In case of other applications, the number n of distinct pattern
classes, but also the number ni , i = {0, .., n − 1} of reference pattern for each considered class
i, must be determined in accordance. Typical industrial applications for example consider 2-
classes problems n = 2 with the same proportion of reference pattern ni ∼ = n j ∀i = j, {i, j} =
{0, .., n − 1}.
However, in case of the considered inspection task, other different approaches would have
been possible, such as the consideration of two consecutive 2-classes procedures. It would have
been possible to first classify all good and all bad patterns, ΩA , {ΩR,3D ; ΩR,2D }, and then to
classify all 3D- and 2D-bad ones, ΩR,3D , ΩR,2D .
Table 2. Rates R of correctly classified patterns for image set Φ00 with Fourier’s textural
features and stripe adapted features by means of a 1-NN classifier.
cS 89.3 0.26
In Sec. 3.1 the assumption was made that some frequency subbands could be more represen-
tative of the stripe patterns to be characterized. This is clearly observable in case of the results
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6655
Rates for increasing distortions of types -1- and -2- for feature vectors cFr,θ ,v,u cFθ and cS .
Fig. 7. The detection rates were computed for different image sets and correspond to in-
creasing distortions of type -1- and of type -2-. Left to right values: detection rates for
image set Φ00 to image sets Φ41 and Φ42 .
listed in Table 2. A high discrepancy in the classification results is observable concerning the
Fourier-based approach. Best classification rates of 92.4% are obtained when only the 10 direc-
tional Fourier features cFθ are used. With the adapted features, the best rate could be obtained
when all the 20 features are used. It is however noticeable that the 14 “free-form” features
outperform the 6 “adapted” ones.
These first results show that from the 33 Fourier features and the 20 stripe features, the feature
group made of the 10 directional Fourier features is particulary relevant in terms of stripe pattern
characterization. Thus, further investigations are dedicated to the characterization of free-form
surfaces using the 33 Fourier, the 10 directional Fourier, and the 20 adapted features sets and
groups.
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6656
Rates for increasing distortions of type -1- and -2- for feature vectors cFr,θ ,v,u S , cFθ S , and
1−NN cF S .
θ
Fig. 8. The detection rates were computed for different image sets and correspond to in-
creasing distortions of type -1- and of type -2-. Left to right values: detection rates for
image set Φ00 to image sets Φ41 and Φ42 .
The next step now consists of attempting to improve these results in terms of increasing the
classification rates and decreasing the number of necessary features, by fusing and selecting the
involved features.
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6657
rates, but leads to similar classification results when the combined 10 directional Fourier and
20 adapted stripes are used.
Thus, the last investigation is dedicated to a more detailed depiction of the considered FSS
method, in order to determine the relevant features.
Table 3. Selected features when a wrapper 1-NN approach is used, for increasing distortion
of type -1-. The maximum number of times a feature can be selected is 10. The variables
Nc,sub on the left give the total number of selected features after the 10 runs. The 10 time,
9 time and 8 time selected features are marked with ∗∗∗ , ∗∗ and ∗ . Results for all relevant
features are marked in bold.
Feature set Φ00 Φ41 Φ42 Φ43 Φ44
Nc,sub 90 90 95 107 108
cS (00) 0 0 0 0 0
cS (01) 1 1 0 0 3
cS (02) 0 0 0 0 0
cS (03) 0 0 1 2 3
cS (04) 3 3 2 8∗ 2
cS (05) 4 4 1 3 3
cS (06) 2 2 4 1 0
cS (07) 9∗∗ 9∗∗ 3 8∗ 3
cS (08) 6 6 6 3 3
cS (09) 7 7 4 6 4
cS (10) 7 7 6 7 5
cS (11) 0 0 3 5 4
cS (12) 6 6 3 3 5
cS (13) 2 2 6 8∗ 10∗∗∗
cS (14) 0 0 1 0 7
cS (15) 0 0 3 2 7
cS (16) 9∗∗ 9∗∗ 6 6 9∗∗
cS (17) 2 2 5 10∗∗∗ 6
cS (18) 1 1 5 4 6
cS (19) 6 6 8∗ 10∗∗∗ 9∗∗
cFθ (0) 0 0 2 1 0
cFθ (1) 0 0 2 1 1
cFθ (2) 5 5 3 4 1
cFθ (3) 4 4 5 2 3
cFθ (4) 4 4 4 1 3
cFθ (5) 10∗∗∗ 10∗∗∗ 10∗∗∗ 10∗∗∗ 10∗∗∗
cFθ (6) 0 0 0 1 0
cFθ (7) 1 1 2 1 0
cFθ (8) 1 1 0 0 1
cFθ (9) 0 0 0 0 0
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6658
Table 4. Selected features when a wrapper 1-NN approach is used, for increasing distortion
of type -2-. The maximum number of times a feature can be selected is 10. The variables
Nc,sub on the left give the total number of selected features after the 10 runs. The 10 time,
9 time and 8 time selected features are marked with ∗∗∗ , ∗∗ and ∗ . Results for all relevant
features are marked in bold.
Feature set Φ00 Φ41 Φ42 Φ43 Φ44
Nc,sub 90 99 107 116 107
cS (00) 0 0 1 0 0
cS (01) 1 1 0 1 0
cS (02) 0 0 2 0 0
cS (03) 0 0 3 2 3
cS (04) 3 3 4 5 3
cS (05) 4 2 1 3 5
cS (06) 2 4 1 5 8∗
cS (07) 9∗∗ 8∗ 7 6 5
cS (08) 6 6 6 6 6
cS (09) 7 5 4 4 2
cS (10) 7 4 6 6 1
cS (11) 0 0 1 1 0
cS (12) 6 6 2 6 1
cS (13) 2 5 5 8∗ 4
cS (14) 0 4 4 4 3
cS (15) 0 5 0 2 6
cS (16) 9∗∗ 4 6 3 4
cS (17) 2 1 4 5 8∗
cS (18) 1 3 0 4 6
cS (19) 6 7 7 7 4
cFθ (0) 0 0 0 0 1
cFθ (1) 0 1 3 3 2
cFθ (2) 5 7 4 5 5
cFθ (3) 4 2 5 2 3
cFθ (4) 4 5 9∗∗ 6 6
cFθ (5) 10∗∗∗ 10∗∗∗ 10∗∗∗ 10∗∗∗ 10∗∗∗
cFθ (6) 0 2 4 0 0
cFθ (7) 1 2 2 5 4
cFθ (8) 1 1 5 3 1
cFθ (9) 0 1 1 4 6
An important parameter is the variable Nc,sub , which is the total number of selected features
after the 10 runs of the 10-Fold cross-validation. As 10 is the maximum number of times a
feature can be selected, Nc,sub /10 is the average measure of feature relevance. For both tables,
increasing the distortion of the bright/dark structures, leads to an increase of the necessary
relevant features.
A general remark for both tables concerns the types and the number of selected features,
which are approximately the same. It appears that approximately seven features, i.e. only a
fourth of the initial 30 ones, are relevant. Most of the selected features are adapted ones, whereas
mainly the directional 90o Fourier features have a strong relevance.
It is also noticeable, that feature relevance is related to the bright/dark structure distortion
degree. As an example, in case of both tables, the importance of feature cS13 is proportional to
the distortion degree, whereas the contrary is observed for feature cS07 .
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6659
6. Summary
In this paper, a general structured-illumination-based method for the characterization and inter-
pretation of free-form and rough surfaces is proposed. Such a procedure was initially defined
for the inspection of cylindrical specular industrial objects.
Starting from the reference image set defined for the qualification of the industrial process,
eight additional image sets, corresponding to various types of free-form rough and specular
shapes recorded with a structured illumination could be defined. All these image sets are used
to search for the most appropriate pattern recognition process, in terms of retrieving the most
adequate subset of features.
In order to address such an inspection task within a general approach, two different image
content description methods, a Fourier-based approach and an adapted stripe-based technique,
were considered. These methods necessitate the computation of a huge amount of 33 and 20
features, which signifies high computational costs. Hence, in order to propose a competitive
solution adapted to real-time processes, extensive investigations were done to retrieve only the
most relevant features that accurately classify the reference image sets.
A four steps feature evaluation, grouping, fusion, and selection procedure was taken into
consideration. At first, each feature group is evaluated and compared individually. Then, the
influence of various feature combination and selection techniques on the detection accuracy
was evaluated. Finally, it has been demonstrated that feature grouping leads to an increase
of at least 2 % of the classification rates, and that on average approximately a fourth of the
initial features are relevant for free-form surfaces characterization by means of a structured
illumination.
Acknowledgment
The author would like to thank the Bavarian Research Foundation BFS (Bayerische
Forschungsstiftung) for its financial support.
#120418 - $15.00 USD Received 25 Nov 2009; revised 25 Jan 2010; accepted 26 Jan 2010; published 16 Mar 2010
(C) 2010 OSA 29 March 2010 / Vol. 18, No. 7 / OPTICS EXPRESS 6660