The present invention relates to combination of a response adapting filter and a detector, the detector having a predetermined spectral response function to electromagnetic radiation; a method of is preparation, a camera comprising such a response filter and detector combination, and use thereof in e.g. colour measurements in combination with an integrating cavity and a vision inspection system of natural and/or a synthetic material surfaces; also a display and detector combination, a method of displaying optical information, a colour display and monitor system, and a method of controlling colour display, said combination, systems and methods comprising such combination of a response adapting filter and a detector.
The Technical Field
Basics for Colour Measurements
The standard 2° calorimetric observer was defined by CIE in 1931 through the colour matching functions {overscore (x)}(λ),{overscore (y)}(λ) and {overscore (z)}(λ). See CIE publication No 15, COLORIMETRY Official recommendation of the international commission on illumination, 1971. The tristimulus values, X, Y and Z, of a given colour stimulus of a light source S(λ) are defined and calculated or measured as:
wherein k is a factor for normalizing the light source, S(λ) defined as:
and wherein φ(λ) is the colour stimulus in question defined as either following formulas 5a, 5b and 5c:
φ(α)=S(λ)×ρ(λ) (5a)
φ(λ)=S(λ)×β(λ) (5b)
φ(α)=S(λ)×τ(λ) (5c)
wherein
For the above-given definitions of k in formula (4), Y defines the reflectance, the luminance factor, or the transmittance, expressed in percentage.
If the colour stimuli is direct light from the light source then φ(λ)=S(λ) (and if given in Watts) and k=683 lumen×W−1, then the Y-stimulus value is the luminous flux from the light source.
The chromatic coordinates x,y are calculated from the tristimulus values as:
Hence, for a given light source S(λ), the colour is unambiguously given by the chromatic co-ordinates (x,y) and Y. Other chromatic coordinates and colour differences defined by CIE as well as defined by others can also be derived from the tristimulus values X, Y and Z, of the CIE recommendations, ibid.
Colour Measurements
The function φ(λ) can be measured with a colour measuring system comprising a simple sensor, a scanning monochromator and a suitable light source and the tristimulus values X, Y and Z can be derived according to formulas (1), (2) and (3) and tabulated values of {overscore (x)}(λ), {overscore (y)}(λ) and {overscore (z)}(λ).
Most spectroradiometers utilizes this principle, although with substantially modified equipment. Preferably an imaging sensor like a CCD or a CMOS array photo detector is used as the sensor.
Alternatively, a grating, a linear CCD, or an array of photodiodes can be used to simultaneously measure φ(λ).
Measurements with spectroradiometers and uniform illumination can by calibration to a known sample be made independent of the light source used. Consequently, such measurements can be converted to a result for any given light source provided that the object measured is uniformly illuminated during the measurement.
Some spectroradiometers uses a number of LED's with dominant wavelengths throughout the spectrum instead of a ‘white’ light source and a monochromator. Such spectroradiometers work fine on non-fluorescent objects.
A simple calorimeter comprises a X-filter, a Y-filter and a Z-filter in combination with an imaging device and a sensor, each of said X, Y, Z-filters realizing one of the colour-matching functions {overscore (x)}(λ), {overscore (y)}(λ) and {overscore (z)}(λ). Known filters comprise a stack of colour filters, a mosaic of filter segments, and a template with a grating.
Only stacked colour filters in combination with an imaging sensor exhibit imaging properties.
The filters can be positioned in the colour measuring system to either shape the light source or shape the incoming light. Both methods are used in various applications. In many cases, because sensors are small and light sources in many cases are relatively large, the filters are positioned to shape the incoming light in front of the sensor.
Measurements with prior art filter calorimeters must be performed with a given light source the result of which, however, cannot unambiguously be converted from this one light source to another, even if proper uniform illumination has been used.
Prior art filter colorimeters suffers from either poor match to the colour-matching functions or low transmittance in case of the stacked type filter.
Stacked type filters can be made imaging, however they suffer a limited accuracy, and they require very expensive detectors such as cooled CCD's in order to operate under the inherently low transmittance of these filters. Mosaic- and template-type filters cannot be made imaging. Consequently, there is a need for colour measuring systems comprising colour-matching filters and detector combinations which allow imaging and which does not require expensive sensitive and cooled detectors.
Many attempts have been made to establish colour-measuring systems and devices, including calibration procedures therefore.
Colour cameras comprising 3 CCD detectors where the incoming light is split into 3 components of red, green and blue light have been suggested. However, shades of these three colours cannot be distinguished from a change in luminance because only one channel responds to the shades. Colour-measuring systems and devices based on such light splitting function cannot perform repeatable and traceable colour measurements according to the standards set by CIE.
Illumination and Geometry in Colour Measurements
Standard geometry's are defined for measuring reflectance and/or transmittance of an object, cf. CIE publication No 38 ‘Radiometric and photometric characteristics of materials and their measurement’, 1977. The standard geometries are 0°/45°, 0°/diffuse and 0°/total. The 0° measurements are performed in integrating spheres thereby obtaining uniform and diffuse illumination, and simultaneously excluding exterior light. The diffused light is obtained by including a light trap for the specular component.
According to Helmholtz reversal principle, the direction of illumination and observation can be reversed. Prior art measurements set-ups are well described in G. Wyszecki, “COLOR SCIENCE Concepts and Methods, Quantitative Data and Formulae”, 1982, in 3.3.7 “Standard Illuminating and viewing conditions” and in 3.12.3 “Spectrometers”.
Calibration of Calorimeters
For filter colorimeters comprising filters exhibiting given colour matching functions that are not identically with the CIE colour matching functions, the tristimulus values X′, Y′ and Z′ are found for a given colour stimulus. The colorimeter response might be improved in a more or less limited region of the colour space by introducing a 3×3-correction matrix Mcorrection as defined in formula (7). This matrix is found by measuring, at least 3, known samples, and then solving a set of equations to find the correction matrix.
In the extreme case, the response of special samples, monochromatic radiation can be measured for many given wavelengths, and a correction matrix can be found by minimizing with respect to some suitable error metrics, e.g. the f′1 defined in the following section.
Calculation of f′1 Errors for Detectors with Specified Spectral Responsivity
In CIE publication No 53, “Methods of characterising the performance of radiometers and photometers”, 1982, metrics are given for integrated photometer errors.
This error metrics concerns only the relative response. The output from radiometers and photometers must be corrected for linearity and dark current according to well-known procedures.
and
In the case of a calorimeter, 3 separate detectors are used; one for each colour-matching function, and hence 3 f′1 errors are obtained.
Prior Art Disclosures
G. Wyszecki, “COLOR SCIENCE Concepts and Methods, Quantitative Data and Formulae”, 1982, 3.12.5 “Tri-stimulus-Filter Colorimeters”, describes the different types of calorimeters, including the template type, the stacked filter type, and the mosaic filter type. The template type and mosaic filter type can be very accurate but cannot be imaging. The stacked filter type can be accurate but with the expense of very small transmittance and therefore only useful with very sensitive sensors e.g. cooled CCD or photo multipliers.
U.S. Pat. No. 5,850,472 “Colorimetric imaging system for measuring color and appearance” discloses an imaging calorimeter based on a colour video camera with RGB response. In the “real world” the transform from RGB to XYZ colours in CIE space is not possible.
Object of the Invention
It is an object of the present invention to seek to provide an improved colour measuring system, in particular a tristimulus camera.
It is another object of the present invention to seek to provide a colour measuring system with improved transmittance.
It is another object of the present invention to seek to provide a colour measuring system comprising colour-matching filters and detector combinations which allow imaging and which does not require expensive sensitive and cooled detectors.
Further objects appear from the description elsewhere.
Solution According to the Invention
“Combination of Response Adapting Filter and Detector”
In an aspect according to the present invention, there is provided a combination of a response adapting filter and a detector, the detector having a predetermined spectral response function D(λ) to electromagnetic radiation; the response adapting filter comprising:
It has surprisingly turned out that a very high transmittance is achieved whereby it is obtained that a colour measuring system comprising colour matching filters and detector combinations that allow imaging and which does not require expensive sensitive and cooled detectors can be provided.
Preferred embodiments are defined in the dependent claims 2-6.
“Method of Preparing a Response Adapting Filter and Detector Combination”
In another aspect according to the present invention there is provided a method of preparing a response adapting filter and detector combination, the method comprising:
Preferred embodiments are defined in the dependent claims 8-11.
“Camera”
In still another aspect according to the present invention, there is provided a camera, the camera comprising:
Preferred embodiments are defined in the dependent claims 13-19.
“Camera Applications”
In still a further aspect according to the present invention, there is provided use of a camera according to the present invention. Preferred uses are defined in claims 20-24.
In a preferred embodiment a camera according to the invention is used in combination with an integrating cavity.
In another preferred embodiment a camera according to the invention is used in colour measurement in a vision inspection system.
In another preferred embodiment a camera according to the invention is used in colour measurement of a surface of natural and/or a synthetic material, wherein said natural surface is selected from the group consisting of a surface of a biological material including human and animal tissue and skin; and plants tissue including wood, and wherein said synthetic natural surface is selected from the group consisting of a surface of a material of textile, concrete, and paint.
“Display and Detector Combination”
In another aspect according to the present invention, there is provided a display and detector combination, said combination comprising:
Preferred embodiments are defined in claims 26-29.
“Method of Displaying Optical Information”
In another aspect according to the present invention, there is provided a method of displaying optical information, said method comprising:
Preferred embodiments are defined in claims 31-32.
“Colour Display and Monitor System”
In another aspect according to the present invention, there is provided a colour display and monitor system, said display and monitor system comprising:
It is intended that the term “light-emitting means to emit coloured light” include a colour light source, e.g. a phosphorous material emitting coloured light, or e.g. a diffusor emitting transmitted or reflected light, or fluorescence light.
Also, it is intended that the term “a display control signal” includes control signal for any suitable display means, e.g. control signals for an electronic monitor screen device, or e.g. control signals for a colour printer, said control signals optionally triggering further control signals of said means and devices.
Preferred embodiments are defined in claims 34-41.
In a preferred embodiment, said system further comprising signal storage means, said signal storage means storing at least one reference display control signal whereby it is obtained that a reference point for the display can be established.
In a preferred embodiment, said at least one reference display control signal is derived from a detector signal generated by a display and detector combination as defined in an aspect of the invention whereby e.g. electronic information of a colour display provided by a detector having a predetermined spectral detector response can be obtained.
In another preferred embodiment, said at least one reference display control signal is derived from said monitor signal whereby e.g. a reference point and a possible drift therefrom by the displayed colour display can be monitored.
In a preferred embodiment, said system further comprising a signal comparator means for comparing said monitor signal and said at least one reference display control signal, said signal comparator means producing a comparator control signal in response thereto whereby e.g. a possible drift from a reference point can be established.
Generally, a comparator control signal can be used for various applications, e.g. providing a feedback to illumination means for a corrected illumination of an object being measured.
In a preferred embodiment said system further comprising a control means for adjusting said display control signal, said control means adjusting said display control signal in response to said comparator control signal whereby the display can adjusted to a predetermined spectral detector response of the monitor and matching a predetermined spectral-matching function, e.g. that of a CIE standard calorimetric observer.
In a preferred embodiment, said display control signal, said monitor signal, said at least one reference display control signal, or a combination thereof, comprises an electronic tristimulus signal, in particular that of a CIE standard calorimetric observer.
The display means can be any suitable display means for displaying optical colour information.
In a preferred embodiment, said display means comprises a display means such as an electronic display screen, preferably a video display unit; a projector screen system, or an electronic printer, preferably a colour printer.
Generally, connection between said colour display means and monitor means include any suitable signal connecting means known to a skilled person.
In a preferred embodiment, colour display and monitor system further comprises a connection means for connecting said monitor signal to a display and detector combination as defined in an aspect of the invention, in particular a display means such as an electronic display screen, preferably a video display unit; a projector screen system, or an electronic printer, preferably a colour printer.
“Method of Controlling a Colour Display”
In another aspect according to the present invention, there is provided a method of controlling a colour display, said method comprising:
In a preferred embodiment, said display control signal, said monitor signal, said at least one reference display control signal, or a combination thereof, comprises an electronic tristimulus signal whereby in particular an optimized reproduction of a scene on said display means can be obtained.
It should be noted however that the term “light-emitting means to emit coloured light” is intended to have a broad meaning, including a colour light source, e.g. a phosphorous material emitting coloured light, or e.g. a diffusor emitting transmitted or reflected light, or fluorescence light. However, the term is also intended to include e.g. a colour print the colour of which may be controlled by adjusting the printer producing such a colour print by a signal derived from said monitor signal.
In the following, by way of examples only, the invention is further disclosed with detailed description of preferred embodiments. Reference is made to the drawings in which
A tristimulus image is recorded, as three separate images, by an image collecting and detecting means 14, here a photo detector array, through an imaging means 15, here a lens or lens system, and through three filters 11, 12 and 13, one for each separate image, where the three filters are mounted in a filter-wheel or filter sledge.
In
The total spectral system response is given by the CIE colour matching functions {overscore (x)} (λ), {overscore (y)} (λ) and {overscore (z)} (λ) 61. The response of the filters is hence given by the residual spectral response as found by a folding procedure illustrated in
A computer and suitable software can control the whole process and present the result as images on a video display unit (VDU) or as digital files.
The detector signals representing an image recorded by a camera according to an aspect of the invention can be obtained by means known to a skilled person. In an embodiment CCD array signals are stored in a solid-state memory, or other storage device, e.g. a DVD, CD, etc.
Means for displaying said recorded and stored image representation, including signal processing, are known in the art, e.g. comprising display means such as an electronic display screen, preferably a video display unit; a projector screen system, or an electronic printer, preferably a colour printer.
Suitable comparator and signal correction means are known in the art, including analog and digital signal comparators, e.g. realized in a microprocessor or dedicated analog or digital electronic circuit.
The monitor 182 on-line monitors a calibration target 192 on a screen of the display 181 unit, said screen further showing an image 193. A comperator and correction unit 183 providing adjustment of display control signals for the display 181 unit in response to said monitor control signal, whereby an optimized display, optionally corrected for drift, can be obtained.
Preparation of Response Adapting Filters
According to the invention said one or more response matching filters 11, 12, and 13 of the filter camera are adapted to modify the spectral information of the radiant power from the object so that the total response of the camera matches a predetermined colour-matching function (x(λ)).
In a preferred embodiment a response adapting is prepared according to a method comprising:
Preparation of response adapting filters can be carried out in any way suitable for achieving the desired functions for their individual application.
Examples of use of response adapting filters RA generally include configurations: D-L-RA-A-O-S, wherein D is an image collecting and detecting means, L is an imaging means, RA is a response adapting filter, A is an aperture which can be positioned elsewhere in the system, e.g. D-A-L-RA-O-S, D-L-A-RA-A-O-S, O is an object and S is a light source.
The response-adapting filter RA can generally include structures of thin films of different order, e.g. substrate-AR-T-BG-ND, subtrate-T-BG-BG, substrate-T, wherein AR is an anti-reflex coating, BG is a blocking filter, and ND is a neutral density filter.
Preparation of Optical Multilayer Structures of Thin Films
An optical multilayer structure to be applied in a response-adapting filter and detector combination of the present invention can be prepared by any suitable method that allows preparation of a controlled optical thin film structure.
Techniques include multilayer deposition techniques such as sputtering, evaporation, reactive ion-plating evaporation, and chemical vapor deposition.
Suitable thin film preparation techniques are disclosed by Sullivan et al., see e.g. “Deposition of Optical Multilayer Coatings with Automatic Error Compensation. I. Theoretical Description”, Applied Optics, Vol. 31, 3821-3835, 1992, and “Deposition Error Compensation for Optical Multilayer Coatings. II. Experimental Results—Sputtering System”, Applied Optics, Vol. 32, No. 13, 2351-2360, the content of which is incorporated herein by reference, the latter specifically including an automated magnetron-sputtering system.
U.S. Pat. No. 6,217,720, published Apr. 17, 2001 discloses a multi-layer reactive sputtering method with reduced stabilization time for depositing a complex multilayer coating on a substrate, said coating consisting of at least two materials. Optical measurements are taken of deposited layers and compared with model values to continually control and insurances of homogeneity of the deposited layers and allowance of valid thickness determination from said model. It is shown that complex filters have been fabricated.
The system comprises:
The system is operated in a deposition sequence comprising:
For each layer the following steps are performed:
Preferred embodiments of the invention are illustrated by examples of preparation of a response-adapting filter.
Preparation of Response-Adapting Filter
Preparation of a response-adapting filter 61, here exemplified by preparation of X, Y, and Z filters for a CIE tristimulus camera, is illustrated in
The step of realising the transmittance functions T(λ,X), T(λ,Y), T(λ,Z) of the optical multilayered structure of thin films comprises:
Alternatively, the transmittance functions T(λ,X), T(λ,Y), T(λ,Z) may be derived from any suitable combination of the combined elements of the filter, excluding the response of the response adapting filters, and then add the residual response in the transmittance functions of the optical multilayered structure of thin films to match the desired response of the response adapting filter.
The preferred embodiment of the filters is shown in
The neutral density filter types are chosen so that the total system response in the X, Y and Z channels is close to even for the three channels when exposed either to direct tungsten light or direct daylight. This gives the advantages that the three exposures can be performed without changing anything in the camera settings, and thereby reducing time between exposures, give the same conditions for the three channels and a good signal to noise ratio.
Even for a camera with three channels, each channel comprising a neutral density filter, it can be an advantages to exclude the ND filters, e.g. in order to provide desired sensitivities.
For type-I-filters with substrates 52 there is a minimum of waste of expensive BG glass types. The substrates are first AR coated, which is a simple process compared to the next coating process, the response adapting coating. Then the substrates are cemented to the BG glass and optionally the ND glass. If the coating processes did not succeed then only the simple substrate was wasted. This type I has though the advantages that the thin film always is cemented against another glass and thereby protected from the moisture in the environment.
It can be difficult to cement three filters together without introducing errors in alignment. This problem is gone when the type II filters are used, where the coatings are performed directly on the BG type glass.
This type II filters do expose the thin film to the environment, and therefore this type should only be used for very dense coatings, resistible against moisture or in closed systems protected against variation in humidity.
Thin film coatings show a spectral dependency on the angle of incidence. Therefore, in this embodiment of the inventions, the response adapting filters are placed in front of the lens, and further the lens is selected so that the angle of view is restricted to ±10°. Further the response adapting filters are optimised to an angle of incidence of ±3°, hereby minimizing the overall error.
Due to tolerances in all the components, a deviation from the perfect system response must be expected. Therefore the image collecting system, typically comprising a PC with suitable software, can introduce a correction procedure in form of correction matrices as mentioned for formula (7). If necessary different corrections matrices can be implemented for the different part of the image, depending on the angle in the viewing field.
The chromatic coordinates, calculated according the formula (6) is independent of the absolute level of light. To introduce absolute measurements, of luminance or Y from formula (2), the iris aperture 132, see
Preferred embodiments of the invention are further illustrated by examples of production of a camera including a B/W video camera from SONY (XCD-X700) as image collectors, and the front and rear lens group of an objective supplied from Schneider Kreuznach (Xenoplan 1,4/23) as imaging system. A hole aperture wheel or sledge for controlling the light level on the above-mentioned video camera was provided by means known to the skilled person.
Further, three response-adapting filters according to the invention were provided at outlined above, and mounted in a filter wheel or sledge placed in front of the objective for holding and shifting the filters.
Colour Measurements
As preparation for colour measurements, the tristimulus camera is exposed to complete darkness and an image (an average of say 100 image) is recorded, as the ‘dark image’ so the dark noise pattern is known. Then the camera is exposed to a known scene, preferable a uniform illuminated surface. An image (an average of say 100 image) through one of the filters is recorded as the ‘white image’.
A colour measurement is then performed by taking one dark image, calculating the current dark level, scale the previous dark image to the current dark level and hereby produce the current absolute dark noise pattern.
Then the X filter is shifted in front of the lens and first one image is taken to remove so called lag. Then a number, one ore more by choice, of images are averaged and the current absolute dark noise pattern is subtracted (pixel by pixel). The result is multiplied by the reciprocal white image (pixel by pixel). Same procedure is followed to obtain the Y and Z images.
A matrix, and a factor for scaling the images to absolute values (luminance) then correct the images.
Images can be recorded of scenes with controlled lighting conditions. A preferred embodiment is shown in
Most calibration laboratories for measuring the reflection properties of materials use this set up. The integrating sphere 151 provides both indirect and diffuse light and a shield from unwanted light. The light is provided by light sources 152 inside the sphere or light transported into the sphere by example light guides. The latter has the advantages of reducing heat problems.
A shield 153 prevents the target and the camera from receiving direct light. The target 154 is placed against an opening 157 in the sphere. The camera is measuring through another opening 155. Preferably the measurement is done at an angle 158 towards the target, different from 0°, to avoid reflections between the camera and the target. If the port 156 is closed the measurement is with the specular component included, and if the port is open an equipped with a light trap, the measurement is without the specular component.
These integrating spheres are purchased from Porschke or LMT, both Germany.
This set up is normally used to measure homogeneity of targets with non-imaging spectroradiometers. Samples with colour textures can be measured with the camera and the characteristics of the texture can be measured and calculated. Examples are textiles and all kind of granular materials.
As the camera is measuring colour as the human eye, the camera is very suitable for sorting material like marble and wood, production control etc.
Further this camera has potential in automatic screening of medical samples and growth in titre plates, telemedicine etc.
Number | Date | Country | Kind |
---|---|---|---|
PA 2001 00960 | Jun 2001 | DK | national |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/DK02/00414 | 6/19/2002 | WO |