Your region and language were selected automatically. You can change this at any time in the menu on the site.
Americas | English

2D/3D Surface Texture Measurement

Our visual perception is influenced by color, gloss and the surface structure. Our visual rating takes all three parameters into consideration and makes an overall judgement. Up to now, grain or surface structure could only be judged visually or with high sophisticated microscopes. This has changed with the new spectro2profiler, a pioneering technology combining color, gloss, 2D reflectivity and 3D topography in a robust, portable tool with a short measuring time.


1. Color measurement as you see it

The spectro2profiler uses a circumferential illumination at 45° from six directions and 0° viewing. The proven, innovative BYK LED technology guarantees an outstanding performance: Short-term, long-term and temperature stability are controlled with highest possible accuracy. The extra-large measurement spot with homogenous illumination guarantees highly repeatable and representative readings. All together highest accuracy and inter-instrument agreement are ensured and allow use of digital standards – the key for global color management.

Set-Up_45°c0_1180x580_RGB.jpg

Figure 1 Setup of 45°c:0 color measurement

2. Conventional gloss meter to measure reflectivity and 60° gloss

For historical reasons, the spectro2profiler has a 60° glossmeter built in. Reflectivity and gloss are based on the interaction of light with the physical properties of the sample surface. The intensity is dependent on the material and the angle of illumination. The measurement results of a conventional glossmeter are related to the amount of reflected light from a black glass standard with a defined refractive index. Today's measuring instruments are very precise and widely used in industry, but they hold some weak points in the measurement of structured surfaces. Cast shadows and areas that are invisible to the measurement detector can falsify the measurement result.

Cast_Shadows_1180x580_RGB.jpg

Figure 2 Cast shadows when using traditional 60° gloss measurement

3. Spatially resolved 2D reflectivity measurement

Moreover, the perception of gloss does not only depend on specular gloss but also on the observed contrast between specular highlights and diffusely reflecting surface areas. (1)  A conventional gloss meter is not capable to capture more complex reflective behaviour such as spatially distributed reflections e.g., high reflecting hills next to low reflecting valleys which occur in leather-like structures.
To overcome this limitation, the spectro2profiler offers a new camera-based technology to capture the spatial distribution of reflectivity. An in-line illumination setup eliminates cast shadows, invisible areas, and perspective distortions so that the measurement is independent of orientation. The camera acquires 2D reflectivity images. Figures 3 and 4 show the measurement principle of the spectro2profiler and an example of a grey scaled reflectivity map in which every pixel represents a reflectivity value allowing more detailed analysis of reflectivity distributions of a surface.
 

Spatial_Resolved_Reflectivity_Measurement_585x585_RGB.jpg

Figure 3 Setup of spatial resolved reflectivity measurement

Reflectivity_Map_585x585_RGB.jpg

Figure 4 Reflectivity map of a powder coated surface acquired with spectro2profiler

4. 3D topography analysis using photometric stereo technique

So far visual assessment was the only way to deliver a complete judgement of a textured surface. Therefore, 3D microscopes are used to provide very detailed information of the surface structure in the laboratory for research purposes, but not suitable for fast and easy analysis of production quality.
The spectro2profiler uses photometric stereo technique for estimating surface normals in order to calculate a 3D topography of that surface. The technique was originally introduced by Woodham in 1980. (1) The surface normals are calculated by observing an object from different illumination directions. With each direction, the object casts different shadows on the surface and a camera acquires images for each illumination. Using shape from shading, the surface curvature is estimated, and the height map of the object can be calculated. The result is a real 3D topography of the measured object surface. The unit P-µm is perceived height.

Image_Acquisition_585_RGB.jpg

Figure 5 Image acquisition of different illuminations to calculate surface topography (2)

Height_Map_585_RGB.jpg

Figure 6 Height map of a powder coated surface. The unit P-µm is perceived height.

5. Watershed analysis to define structure cell sizes

Topographies such as leather grains or coarse powder coated structures can be characterized by their structure cells. To divide the topography into cells, the watershed algorithm is used, a region-based segmentation approach. One can imagine that the algorithm gradually floods the valleys of the topography, building rivers until hill areas are surrounded. (3) These areas will be defined as cells, marked as green lines in Figure 7.
Characteristic features of the surface can be calculated based on the watershed segmentation results to compare different structures or grains. Spatial length scales result from camera calibration and are traceable to SI-units. The calculated average cell size correlates to our visual impression of coarseness. The distribution of individual cell sizes is an indication for the uniformity of the surface structure. For example, a natural leather structure varies in uniformity depending on the part of the cow skin. A textured paint can form agglomerations during the wet paint application if the application parameters vary resulting in an inhomogeneous appearance. The normalized cell size deviation is calculated by dividing the cell size distribution with the mean cell size. It is an objective measure to compare the uniformity of different structures independent of its absolute cell size.

3D_Topography_Data_Set_585x585_RGB.jpg

Figure 7 3D topography data of a leather grain

Watershed_Segmentation_585x585_RGB.jpg

Figure 8 Watershed segmentation of topography data

6. Combination of 3D topography data with 2D reflectivity data

To assess the overall appearance of an object, it is necessary to measure surface structure and reflectivity in parallel, as they are mutually interdependent, but are combined for an overall visual assessment. (4) Because our eyes are only capable to acquire 2D information, the human visual system reconstructs 3D information of objects in our brain using shading and reflections. (5) That means, the perceived depth of a structure is dependent on the reflection behaviour on the hills and valleys. Since the spectro2profiler uses the same camera and lens system for the acquisition of 3D topography and 2D reflectivity data, it is possible to combine the data of both measurement principles (Figure 8 and Figure 9). Thus, the reflection of hills and valleys can be separated. The difference between reflection of hills and valleys, describes the contrast and perceived depth of a structured surface.

Topography_Data_585x585_RGB.jpg

Figure 9 3D topography data set: Height is gray scale coded

Reflectivity_Data_585x585_RGB.jpg

Figure 10 2D reflectivity data set: Intensity of reflectivity is gray scale coded

7. Practical example from the automotive industry

Many automotive interior components have a leather-like look and are manufactured by different suppliers with different processes and made of various materials. The appearance of the products surface is analyzed in the different development phases, e.g., at the very beginning by the design department in the grain development to approve suppliers and at the very end by quality control in production. Leather grain structures can appear different in contrast although color and 60° gloss are the same (Figure 10).  This can be caused due to different reflectivity levels of the surface on hills and valleys. Up till now this had to be evaluated visually which is subjective and non-repeatable. The measurement results in the table display how the reflectivity contrast Rc can distinguish the samples despite having the same color and 60° gloss. Moreover, the results of the reflectivity for hills and valleys provide details about what causes the different reflectivity contrasts.

 

CheckzoneSample 1Sample 2Sample 3Sample 4
Mean Reflectivity R (a.u.)162156156155
Reflectivity Hills Rh (a.u.)209188195190
Reflectivity Valleys Rv (a.u.)115122115117
Reflectivity Contrast Rc0.290.210.260.24
60° Gloss (GU)1.31.31.21.3

 

The new measurement parameter reflectivity contrast is an ideal measure for production QC of injection or slush moulded parts.

 

4_grey_Skins_585_RGB.jpg

Figure 11 Four dashboard slush skins of same material with different levels of contrast

8. Coarse paint - powder coated samples

In this example power coated panels of the same color with a fine to coarse structure are evaluated. Visually the samples differ due to different cell sizes. This specific difference is caused by variations in film thickness, but also additives or temperature changes can have an impact on surface texture.
In the smart-chart data table (Figure 12) one can clearly see that the four samples have the same color and 60° gloss values. A differentiation can be clearly done by the mean cell size.

4_blue_Panels_585_RGB.jpg

Figure 12 Four powder coated panels with different structure

Measurement_Results_smart-chart_01_1180_RGB.jpg

Figure 13 Measurement results displayed in smart-chart

9. Eroded plastic parts or fine structured paint

Eroded plastic parts or fine structured paint as shown in Figure 13 have structures too small to segment into visible cells. Therefore, another approach is necessary to evaluate the topography data.
Local maxima and minima are detected and the Micro Peak Distance µPd (µm) is calculated as the peak distance between adjacent peaks on the topography (Figure 14). It correlates with the visually perceived roughness of these fine structures. The higher the value, the rougher the structure appears. The effect of roughness is often enforced, indicated or illustrated by the amplitude of the structure peaks which is measured by the Micro Local Amplitude µA (P-µm).
The results in the smart-chart data table (Figure 15) show the rougher the sample appears, the higher is the micro peak distance and the micro mean amplitude, respectively.
In addition to roughness the visual perception is also influenced by the reflectivity of the surface. This “glossy appearance” is mainly dominated by the contrast between sparkling spots and non-sparkling spots. The spectro2profiler captures the effect with the measure Micro Reflectivity Contrast µRc using the 2D spatial reflectivity information from the camera image.

Micro_Peak_Distance_585_RGB.jpg

Figure 14 Calculation of Micro Peak Distance µPd (µm)

3_grey_finePaint_Panels_585_RGB.jpg

Figure 15 Three panels with fine structured paint

Measurement_Results_smart-chart_02_1180_RGB.jpg

Figure 16 Measurement results displayed in smart-chart

10. Summary

spectro2profiler is a game changer and marks a turning point in the analysis of structured surfaces. The combination of 45°c: 0° color measurement, 60° specular gloss, 3D topography and 2D reflectivity in one easy to use instrument is a milestone in the objective measurement control of textured surfaces. At this moment, the spectro2profiler incorporates four algorithms for surface structure analysis - leather-like structures, inverted leather-like structures, coarse paint textures and fine paint or plastic textures.  Due to its excellent technical performance regarding repeatability and inter-instrument agreement, digital standards can be used as a reference, allowing a flawless communication within a global supply chain.
From now on, our visual perception of colour, gloss and structure can be evaluated in a holistic and objective approach, color and appearance harmony when combining different components can be optimized and all this is possible in the laboratory as well as on the production line with the portable spectro2profiler.

Standards and Literature

(1) Woodham, R.J. 1980. Photometric method for determining surface orientation from multiple images. Optical Engineerings 19, I, 139-144
(2) by Meekohi - Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=44925507
(3) Serge Beucher and Christian Lantuéj workshop on image processing, real-time edge and motion detection (1979).  http://cmm.ensmp.fr/~beucher/publi/watershed.pdf
(4) Qi, L., Chantler, M. J., Siebert, J. P., & Dong, J. (2012). How mesoscale and microscale roughness affect perceived gloss. Edinburgh, Scotland: Lulu Press, Inc.
(5) A. Nischwitz et al., Computergrafik und Bildverarbeitung, Vieweg+Teubner Verlag | Springer Fachmedien Wiesbaden GmbH 2011

CANCEL