The present invention relates to a device and method for optical measuring of thickness of thin planar objects. Measured planar objects are such as paper, tissue, board, film and plastic.
Manufacturing of planar objects such as paper, tissue, board, film and plastic is typically done in continuous process. There are multiple places where the measurement of the product is needed. One of the measurements is the thickness of the product. Thickness can be measured in different ways. A typical way to measure the thickness is to mechanically measure the distance from measurement devices to the surface of the planar object from both sides of the planar object and subtracting the result from known distance between measurement devices.
Another approach is to use radiation sources and measure the radiation that passes through the object and calculate the thickness by knowing the density. The approach has problems. The radioactive sources may be prohibited on some sites and the accuracy may not be good enough due to density variations. The density of paper products is rarely so constant that the measurement using radiation is too inaccurate.
When the thickness is measured using distance from both sides of the object there are several ways to measure the distance. A typical way is to have a glide or shoe that touches the measured surface and the movements of the glide or shoe are measured. This approach has many problems with tear and wear of the measured surface and buildup of contaminants on the contacting elements.
The distance may also be measured contactless from both sides of the object. Here different laser based solutions are invented. E.g. the EP486713 describes a gauge for measuring unsupported web with triangulation sensors attached on a frame both sides of the web. This contactless measurement provides a point measurement without information on inclination angles of the web. This causes big uncertainties with the accuracy.
Other similar approaches with optics based solutions carry following problems: the slight movements of optics used to guide the light and detectors caused by vibrations or temperature changes; the un-linearities of the optics; misalignment of light sources; synchronization of the measurement in both sides of the object. All these affect notably when the measurement accuracy is needed in micrometer level. There is a need to have an accurate contactless thickness measurement that can handle above mentioned problems.
The present invention seeks to describe device and method for improved contactless thickness measurement of thin planar object. This is achieved with device as described in claim 1 and method as described in claim 8.
There are many advantages with the invention. Minimizing the effect of heat expansion and vibration within measurement device, managing the effect of unlinear optics or misaligned light sources and having synchronized distance measurement in both sides of the object.
Example embodiments of the present invention are described below, by way of example only, with reference to the accompanying drawings, in which
The invention uses two optical sensor modules to measure distance from the measured object. The invention is based on use of reference shade and lighting to produce shadow on planar object. The reference shade and shadow are imaged and distance from the measured plane is calculated using the distance of the reference shade and the shadow. Using additional distance measurement on the other side of planar object, it is possible to determine the thickness of the object when knowing the distance between the optical sensor modules.
The optical sensor module 10 for distance measurement of planar object comprises of light source 12, reference shade 13, imaging sensor 14, possible optical elements 15 and computing equipment 16. It may also comprise means to keep the planar object in constant distance from the module. All these are described in detail in following paragraphs.
The light source 12, 22, 22″ can be an LED light, laser light or any other method to produce bright light. The light can be further focused or shaped to illuminate the reference shade with directional light. The light source may produce a wide bandwidth light or light with dedicated bandwidth. The light can be visible, infrared or ultraviolet depending on the imaging sensor used. The light with shorter wavelengths are preferable due to more accurate imaging available. The light may be produced in pulses with interval of 5-100 ms. and duration of 0.1-100 μs. The duration and interval of the pulses are controlled with computing equipment 16 or separate electronics.
The reference shade 13, 23, 23″ can be wires, bars, printed or otherwise processed film, glass with paint or oxidization or some other form of obstacle for light. The shade forms a shading pattern that can also be found as a shadow on the measured surface. This pattern can be three or more dots, lines, or any other 2-dimensional pattern. Examples of shades are shown in
In an embodiment, two or more light sources 12 are installed into the module to illuminate the same target. These light sources can be installed so that there are different angles φ1 or orientation in x,y plane for the light. Two or more light sources lighting the reference shade 13 provide two or more shadows. Different light sources provide illumination with different wavelengths and thus the shadows can be distinguished by selecting different colors from the images captured by the imaging sensor. E.g. blue and green colors can be used simultaneously. Additional light sources provide more combinations of shades and shadows and thus more accurate measurement. Different light sources may also be lit with different timing. This way the lights do not disturb each other.
The imaging sensor 14, 14′, 14″ can be a complementary metal-oxide-semiconductor (CMOS), N-type metal-oxide-semiconductor (NMOS), semiconductor charge-coupled devices (CCD) or Quanta Image Sensor (QIS). The imaging sensor may detect visible wavelengths, ultraviolet or additionally near infrared wavelengths. The imaging sensor converts the image into electrical signals corresponding the image. The imaging sensor has resolution of 1-15 megapixels, with a size of 17 mm2-1000 mm2.
There can be additional optical elements 15, 15′, 15″ between the imaging sensor and the imaging target. The optical elements such as lenses are used to focus the imaging sensor or provide different focal length for the imaging sensor. The optical elements such as optical filters may also be used for polarization or filtering of the light. The elements such as mirrors and prisms may also be used to divert the light so that imaging sensor can be facing to other direction than directly to imaging target. The optical elements may be attached to imaging sensor or to other parts of the module.
The imaging sensor and possible optical elements provide a focal length 17, 17′, 17″ where sharp images can be captured. The focus area 61, 61′ is typically 0.5-2.5 mm depending on used imaging sensor and optical elements and that is a typically restricting factor for the distance between the shade 13 and the measured plane 11.
The computing equipment 16, 26 is one or more microprocessors, Field Programmable Gate Arrays (FPGA) or similar processors and memory, with software capable of distinguishing the shade and the shadow and measure the distance between shade and the shadow from the captured image. The computing equipment is connected to the imaging sensor. Since the shade is static it can be done well detectable by the software. Its location in the image is also very stable and thus minor movements can be pre-programmed to the detection algorithm. The detection of the shadow from the image can be accomplished with several ways. One way is pure edge detection of the shadow. Additionally, looking for the center point of the shadow provides good point for distance calculation. This can be found by detecting the edges of the shadow and looking for the midpoint. If the edges are not clearly visible the midpoint can also be found by applying curve describing the illumination of the image and looking the zero derivate on the area of the shadow. This way the location of the shadow can also be determined even if the resolution of the imaging sensor is not very high. There are multiple known feature detection algorithms for image extracting e.g. Canny edge detector, Sobel filter or Förstner detector. Any other appropriate feature detection algorithm can be used. Additionally, if shade with linear patterns are used, then curve fitting algorithms may be used to detect the possibly smooth shadow. Once the shadow is detected the distance between appropriate shade and its shadow is measured. The shade can be designed so that there is no doubt about correlation between shade and shadow.
The distance between shade and shadow is measured in all visible area A. This approach provides a full shadow that is distorted according to the distance differences. The software finds these distortions and calculates whether there is inclination with the measured plane. If inclination is found, its effect to the thickness measurement can be deducted. The distance calculation can be done purely mathematically by knowing the angle φ1, φ2, φ3 or more preferably by using calibration data as basis.
The software may also be programmed to measure the length of the shadows. This length may also be used for the calculation of the distance. The software may also be programmed to detect winding lines with the shadows. These may reveal uneven surface. The calculated distance data can be stored in local volatile or non-volatile memory.
The location of needed computing equipment is not restricted. They may be integrated within the optical sensor module 10, 20 or they may locate anywhere else within network connection so that the electrical signals from the imaging sensor may be processed. The computing equipment may be distributed 16′, 16″ within the sensor carriage 32, 33 or the frame 31. The reason to move the computing equipment outside of the module chassis may be due to temperature issues or space required. There can be additional user interface 35 for controlling, monitoring and calibrating the sensor arrangement. The computing equipment comprise also network communication interface such as Ethernet interface or industrial fieldbus interface. This interface can be used to connect the measurement arrangement to automation system.
The optical sensor module 10, 20 may also comprise of means to keep the planar object in usable constant distance 18, 19 from the optical sensor modules. These may include air jets from both sides of the planar object. Also, other airflow-based solutions can be used to keep the measured object substantially stable in z direction. The target of these means is to keep the distance of the measured surface 11, 21 within the optical focus area of the focal length 17.
Measurement device for thin planar object requires distance measurement modules mutually arranged to both sides of the object. The distance between the modules must be known.
The measurement device can be a combination of two optical sensor modules 10 with imaging sensor focus area set to fully cover the measured object. It may be done with two modules 20 with tilted imaging sensor focus area set to cover only part of the measured object. The measurement device can also be a combination of these. Each of the approaches has its benefits. A device with two full focus covering modules 10 provides a good accuracy for planar distance measurement on both sides. A device with modules 20 with tilted focus areas that cover only part of the measured object provide larger measurement range in z direction. This way the distance between the modules can be large and the measured object may move on y direction. When the measurement device has different modules on different sides of the measured object, the accurate planar measurement on one side is combined with large range in other side. This way the modules can have a larger distance with good accuracy.
It is not mandatory to pre-define the point where the distance measurement begins. Any point within optical sensor module 10 can be considered the starting point e.g. the level of the shade 13, 23. It is however important to calibrate the measurement device so that the distance between the modules is determined and accurate measurements are possible to do. The calibration can be done using thin objects with known thickness. The calibration may also need moving the measured object in z axis to find the changes shadow positions. Once the extremes are calibrated, the mid-values can be interpolated since the measurement principle is linear. On contrary the calibration may also be done with one point extrapolating the edge values.
The thickness measurement needs a proper positioning for optical sensor modules. The distance measurement made in both sides of the web must be done substantially in same x,y position. This positioning can be done with different methods. The main method is related to the frame control and the way how the frame keeps the sensor carriages 32, 33 on synchronized movement. This is typically sufficient. A more detailed control can be done with e.g. magnetoresistive measurement, where magnetoresistor bridge provides accurate information of the deviation in x,y plane.
The distance measurement 29 between optical sensor modules can be done with different methods. An exemplary method is an electromagnetic measurement described in U.S. Pat. No. 4,160,204. Other exemplary methods could be acoustic with ultrasonic measurement or inductive with eddy current based measurement. Example of eddy current measurement is described in FI111991. If the distance 29 stays fixed it can be measured once and stored into memory of computing equipment 16.
The thickness measurement of fast moving web requires very accurate time synchronization for the measurement of the distance on both sides 11, 12 of the web and the distance 29 between optical sensor modules 10, 20. The synchronization of the distance measurements on both sides 11, 12 must be within 0.1 ms to enable the measurement to catch the thickness of same square millimeter, when the web traverses 30 m/s. This synchronization is easy to accomplish since the measurement is based on lighting the object simultaneously from both sides for measurement. Accurate lighting synchronization between multiple light sources is known technology.
In an embodiment, there are one or more light sources 27 such as lasers attached to the second module 20 so that they illuminate at least one point 43 in the area detected by the imaging sensor 14, 14′ on both sides. This spot or a multitude of similar spots or a pattern created by laser can be used to show a common place for distance measurement on both sides of the object 27. The used light sources can be with visual band e.g. red laser, but also in other bands, preferably infrared. If more than one spot is used or the pattern is large enough, the spots or the pattern can be used to reveal tilt of the imaging sensor or optical elements.
The invention provides a new way to contactless thickness measurement for thin planar objects. The invented approach minimizes the effect of heat expansion and vibration, since the relative distance between shade and shadow stays comparatively stable even if there are minor movements with light source or imaging sensor. Also, optical unlinearities or misalignment of light sources do not affect relative distance between shade and shadow in current measurement arrangement.
Number | Date | Country | Kind |
---|---|---|---|
20187188 | Dec 2018 | FI | national |
Number | Date | Country | |
---|---|---|---|
Parent | PCT/FI2019/050734 | Oct 2019 | US |
Child | 17323463 | US |