1. Introduction 2. Objectivation of Inspections via Measurements 3. Practical Application of Micro Sensor Systems 4. Summary & Conclusions

Transcription

1 , San Francisco, United States Mobile Devices and Multimedia: Enabling Technologies, Algorithms and Applications 2015 Session 2: Emerging Mobile Applications and Enabling Technologies Wednesday 11 February 2015, 10:50 AM - 12:30 PM Dietrich Hofmann Randolf Margull Paul-Gerald Dittrich SpectroNet c/o Technologie- und Innovationspark Jena GmbH Hans-Knöll-Straße 6, D Jena Telefon: +49 (0) pg.dittrich@spectronet.de URL: Daniel Kraus Mobile micro-colorimeter and micro-spectrometer sensor modules as enablers for the replacement of subjective inspections by objective measurements for optically clear colored liquids in-field > Paper < FUNDED BY: MEMBER OF THE INITIATIVE: CERTIFIED BY:

2 Content: 1. Introduction 2. Objectivation of Inspections via Measurements 3. Practical Application of Micro Sensor Systems 4. Summary & Conclusions 2

3 1. Introduction to the Photonic Characterization of Optically Clear Colored Liquids FUNDED BY: MEMBER OF THE INITIATIVE: CERTIFIED BY:

4 1.1 Inspection or Measurement Task Quality Analysis of Petroleum Oils & Fuels waste still usable fresh Saybolt Scale for Petroleum Oils & Fuels 4

5 1.2 Photonic Inspection & Measurement Tasks Photonic Measurements Emission Remission Transmission e.g. LED and OLED Lights e.g. Papers and Fabrics e.g. Mineral Oils and Fuels Measurement of Illuminations Measurement of Surfaces Measurement of Liquids Remission Emission Transmission 5

6 1.3 From Subjective Inspections to Objective Measurements Subjective Inspections Objective Measurements Human Eye Visual Color Inspection Method Tristimulus Sensors Quasi-Spectral Sensors Spectral Sensors Tristimulus, Quasi-Spectral and Spectral Measurement Methods + 6

7 1.4 Liquid Specific Color Characterizations Colored Liquids Chemicals & industrial oils Petroleum oils & fuels Dark oils & fats Beers, malts and caramel Pharmaceutical solutions Industrial oils and surfactants Sugars, syrups and honeys Water & wastewater Transparent liquids Liquid and Application Specific Color Characterization Pt-Co/Hazen/APHA, Garner, Iodine, CIE values, spectral data Saybolt, ASTM Color, Pt-Co/Hazen/APHA, CIE values, spectral data FAC, Gardner, CIE values, spectral data EBC (CIE & 430 nm), ASBC (CIE & 430 nm), CIE values, spectral data EUR, US & Chinese Pharmacopoeia Color, Pt-Co/Hazen/APHA, CIE values, spectral data Klett Color (blue filter KS-42), Pt-Co/Hazen/APHA, CIE values, spectral data ICUMSA Color (420, 560, 710 nm), Honey Color, CIE values, spectral data ADMI (spectral & tristimulus filter methods), Pt-Co/Hazen/APHA, CIE values, spectral data CIE values, L*a*b* or L*C*h* color space, Hunter Lab color space, spectral data 7

8 1.5 Standards for Liquid Specific Color Characterizations DIN 5033: Colorimetry; spectrophotometric method colored liquids as XYZ tristimulus values in transmission ASTM E : Standard Practice for Computing the Colors of Objects by Using CIE System DIN EN 1557: Colorimetric characterization of optically clear colored liquids as XYZ tristimulus values in transmission ASTM D : Standard Test Method for Color of Petroleum Products by the Automatic Tristimulus Method 8

9 1.6 Solid & Liquid Color Standards for Saybolt Scale Solid Color Standards Liquid Color Standards Saybolt Scale for Petroleum Oils & Fuels 9

10 2. Objectivation of Inspections via Measurements for Photonic Characterization of Optically Clear Colored Liquids FUNDED BY: MEMBER OF THE INITIATIVE: CERTIFIED BY:

11 2.1 Micro Sensors for Photonic Measurements Dimensional RGB Spectral XYZ Spectral Quasi-Spectral Spectral Hyperspectral 11

12 2.2 Mobile Photonic Measurement Systems Modularization of Micro Sensors Miniaturization of Smart Computers Consumerization of Applications Convenient, Reliable and Affordable Micro Sensor Modules and Smart Computers for Photonic Characterizations of Optically Clear Colored Liquids 12

13 3. Practical Application of Micro Sensor Systems for Photonic Characterization of Optically Clear Colored Liquids FUNDED BY: MEMBER OF THE INITIATIVE: CERTIFIED BY:

14 Relative intensity Relative intensity 3.1 Characteristics of Investigated Photonic Micro Sensors Spectral responsitivity Tristimulus Sensor = 1 0,8 0,6 0,4 0, Wavelength in nm XYZ Tristimulus Values Spectral responsitivity Quasi-Spectral Sensor = 1 0,8 0,6 0,4 0, Wavelength in nm Six Sampling Points + (center and dark) = Spectral curve Spectral Sensor 14

15 3.2 Challenges & Chances of Photonic Micro Sensors Tristimulus Sensors Quasi-Spectral Sensors Spectral Sensors Items Resolution Low Medium High Costs Low Medium High Speed High High Low Accuracy Low Medium High 15

16 3.3 Calibration of Tristimulus & Quasi-Spectral Sensors The measurement of the XYZ tristimulus color values are realized in parallel with a spectral sensor and a tristimulus sensor. The relationship between the spectral sensor values (T) and the tristimulus sensor values (S) are used for the target based correction matrix (equation 1). T = K S (1) After the transposition of equation 1 the correction matrix (K) can be calculated (equation 2). K = T S T S S T 1 (2) For a compensated, balanced and calibrated tristimulus and quasi-spectral sensor measurement the following steps have to be accomplished: Calibration Step 1: Offset compensation Calibration Step 2: White balancing Calibration Step 3: Measurement with calibration references Calibration Step 4: Calculation of the correction matrix 16

17 3.4 XYZ-Tristimulus Values with Tristimulus Sensors Ilumination Liquid Probe Sensor XYZ Sensor Values Ilumination Spectrum Liquid Spectrum Sensor Sensitivities Standard Illumination Standard Observer Correction Matrix K Calculation Method XYZ-Values 17

18 3.5 XYZ-Tristimulus Values with Quasi-Spectral Sensors Ilumination Liquid Probe Sensor Quasi-Spectral Sensor Values Ilumination Ilumination Spectrum Liquid Spectrum Sensor Sensitivities Standard Illuminant Standard Observer Correction Matrix K Calculation Method XYZ-Values 18

19 3.6 XYZ-Tristimulus Values with Spectral Sensors Ilumination Liquid Probe Sensor Spectral Sensor Values Ilumination Ilumination Spectrum Liquid Spectrum Sensor Sensitivities Standard Illuminant Calculation Method XYZ-Values Standard Observer 19

20 3.7 Transformation from XYZ to L*a*b* Values L = 116 Y Y n or L = Y 16 Y n a = 500 X X n 1 3 Y 1 3 Y n or a = X X n 3 Y Y n b = 200 Y Y n 1 3 Z 1 3 Z n or b = Y Y n 3 Z Z n with: X, Y, Z tristimulus values X n, Y n, Z n ideal values for white from standard illuminant and observer combinations 20

21 3.8 Calculation of Color Differences ΔE in L*a*b* according to ASTM E ΔE = L 1 L a 1 a b 1 b 0 2 with: ΔE color difference between investigated probe, calibration standard and distilled water probe L 1, a 1, b 1 lightness and chroma coordinates target-color L 0, a 0, b 0 lightness and chroma coordinates actual-color 21

22 ΔE Color Difference 3.9 Transformation from ΔE in L*a*b* to Saybolt Scale according to ASTM D calibration curve is derived by plotting the color differences versus the Saybolt color numbers S = α + with: β (log ΔE τ) S Saybolt color number ΔE color difference between investigated probe, calibration standard and distilled water probe α intercept correction constant (typical = 51.1) β slope correction constant (typical = 44.5) Saybolt Color Number τ correction constant (typical = 2.55) 22

23 3.10 Measurement Results for Tristimulus & Spectral Sensors Probe5 Probe3 Probe2 Probe6 Probe7 Probe8 Probe1 Probe4 Color Values: Normal Probe Spectral Sensor Tristimulus Sensor Probe1 Probe4 Probe8 Probe7 Probe6 Probe2 Probe3 Probe5 ΔE 0,093 0,067 0,071 0,052 0,258 0,082 0,133 0,524 23

24 3.11 Measurement Results for Quasi- & Spectral Sensors Probe5 Probe3 Probe2 Probe6 Probe7 Probe8 Probe1 Probe4 Color Values: Normal Probe Spectral Sensor Quasi-Spectral Sensor Probe1 Probe4 Probe8 Probe7 Probe6 Probe2 Probe3 Probe5 ΔE 0,185 0,066 0,098 0,068 0,151 0,072 0,036 0,047 24

25 3.12 Measurement Results for Values in Saybolt Scale Probe5 Probe3 Probe2 Probe6 Probe7 Probe8 Probe4 Probe1 Tristimulus Sensor -32,759-14,204-8,835-6,347 1,602 6,490 29,996 42,862 Quasi-Spectral Sensor -32,279-14,020-8,786-6,056 1,573 6,408 29,835 42,240 Spectral Sensor -32,314-13,969-8,659-5,759 1,748 6,718 30,052 42,982 Saybolt Scale for Petroleum Oils & Fuels

26 4. Summary & Conclusions and Further Information FUNDED BY: MEMBER OF THE INITIATIVE: CERTIFIED BY:

27 4.1 Summary & Conclusions Innovative convenient, reliable and affordable measuring instruments with micro sensor modules, smartpads and software apps enable objective colorimetric characterizations of optically clear colored liquids at the point of interest. Micro-colorimeter modules with tristimulus sensors have the same color accuracy as quasi-spectral and spectral sensor modules in terms of colorimetric characterizations of optically clear colored liquids. Micro colorimeter modules with tristimulus and quasi-spectral sensors can be calibrated by any kind of linear independent liquid color solutions. Highly accurate colorimetric characterization of liquids according to DIN EN 1557 and ASTM D can be seamless applied to micro-colorimeter modules with tristimulus and quasi-spectral sensors. Smartpads are convenient, reliable and affordable mobile quality assurance systems in combination with micro-colorimeter and micro-spectrometer modules and specialized image processing software. Imaging software apps can be unified for the numerical and graphical representation of color spectra and color coordinates in the color spaces. 27

28 4.2 Further Information 10 th Anniversary JENCOLOR ZEISS SpectroNet Collaboration Forum , Jena Germany Innovative Photonic Micro Sensor Applications > RGB, XYZ, Quasi-Spectral, Spectral, Multi-Spectral, Hyper-Spectral < > Pixel, Matrix, Cube < You are cordially invited as speakers and/or participants Further Information see in time at 28

29 , San Francisco, United States Mobile Devices and Multimedia: Enabling Technologies, Algorithms and Applications 2015 Session 2: Emerging Mobile Applications and Enabling Technologies Wednesday 11 February 2015, 10:50 AM - 12:30 PM Thank you for your attention Dietrich Hofmann Randolf Margull Paul-Gerald Dittrich Daniel Kraus SpectroNet c/o Technologie- und Innovationspark Jena GmbH Hans-Knöll-Straße 6, D Jena Telefon: +49 (0) pg.dittrich@spectronet.de URL: FUNDED BY: MEMBER OF THE INITIATIVE: CERTIFIED BY: