Imaging System for Holographic Registration Marks

Abstract

A machine vision system 100 for imaging a target surface 106 is disclosed, which includes a light source 102 for illuminating the target surface and an image sensor 104 for capturing at least one imag

Description

19 06 24 Field This disclosure is related to a machine vision system and corresponding method for imaging 5 a target surface, and in particular where the target surface is a photosensitive material containing a holographic fiducial or holographic registration mark. The disclosure also relates to a system and associated method for aligning a photosensitive target surface having a holographic fiducial or registration mark, with a work surface comprising such a machine vision system. 10 Background Manufacturers and engineers frequently face the challenge of aligning registration marks with a corresponding target on a workpiece. These marks, which are essential for precision during manufacturing processes, can take various forms, including individually or combinations of 15 crosshairs, bullseyes, 2D codes such as QR codes, and / or 1D codes such as barcodes. To read these marks, optical imaging systems with high accuracy are employed, which often require the use of telecentric lenses capable of capturing large fields of view. Consequently, these lenses must have diameters that are greater than the size of the object they are designed to view. The downside to such systems is their high cost and considerable bulk, 20 which presents difficulties when attempting to integrate them into manufacturing environments where space is constrained. Optical imaging systems tasked with reading marks such as fiducials, registration marks, QR codes, bar codes and so on must overcome obstacles related to the accuracy required in correctly reading the marks on a target substrate on which the mark is applied. Traditional 25 high-precision optical systems often rely on such telecentric lenses designed to reduce distortions across the field of view. This typically necessitates the adoption of larger lenses accompanied by broader viewing fields. Such optical systems, whilst highly accurate are known to be costly, bulky, and therefore difficult to integrate into areas where space conservation is crucial. Conversely, more compact lens systems, although beneficial in terms 30 of space efficiency and cost, are prone to introducing distortions that vary with the lens's distance to the mark, thereby requiring the application of software algorithms for post image capture distortion correction. 19 06 24 Moreover, in the field of holographic optical elements such as those used in augmented reality glasses or head-up displays, the marks are typically required for later alignment of the holographic optical element with a work piece such as for example an eyeglass lens or a waveguide of a pair of augmented reality glasses. 5 Presently, the industry's attempts to address the significant issues of size and cost that are inherent in large field of view camera / imaging systems often result in trade-offs that give rise to a distinct set of challenges. Specifically, compact and economically feasible camera systems that forgo the use of telecentric lenses are susceptible to distortions across the field of view. These distortions are not constant but fluctuate depending on the spatial separation 10 between the camera system and the target scene to be imaged, which adds complexity to the imaging process. Although software algorithms have been developed to rectify these distortions, this method is not without its own set of constraints and limitations. In addition, the use of holographic registration marks / fiducials presents its own unique set of constraints which make accurate reconstruction and reading of the marks difficult. Specifically, in order to view 15 the marks, they must be reconstructed using suitable reference light with a particular angle and wavelength combination. This combination of angle and wavelength is dictated by the specific recording geometry and wavelength used to create the marks, and typically necessitates changes in alignment of both camera and light source for each unique holographic registration mark. 20 Figure 1a shows the general replay of a holographic registration mark or fiducial that may be recorded in a photosensitive material. The arrangement of Figure 1a shows a reflection hologram configuration, however the skilled person will appreciate the problems identified are also relevant for transmission type holograms or Fourier transform type holograms. As shown, the reconstructed holographic images I-, I and l+ of the holographic registration mark 25 correspond to respective angular position of an illuminating point source P-, P and P+, where the holographic images are displaced from each other corresponding to the different respective positions of the point sources. A change in angular position of the point source will therefore change the position of the reconstructed image. In other words, positions of the point sources P-, P and P+ give rise to respective images I-, I and l+ of the registration mark. This 30 can lead to positional inaccuracies in the replayed or reconstructed image of the registration mark. To accurately reconstruct the holographic registration mark, the photosensitive material containing the registration mark must be located at the same distance from the point source and at the same angle of recording as when the hologram was recorded in the photosensitive material. 19 06 24 Where the light source is made up of multiple point sources P as shown in Fig 1b, this will result in multiple images of the holographic registration mark being reconstructed at I. Each of the images will overlap and become blurred and extend the size of the replayed holographic registration mark which increases the inaccuracy of the position of the replayed registration 5 mark. Furthermore, a range of wavelengths from the light sources can also blur the replayed holographic registration mark and cause positional inaccuracies. Therefore, unless all the parameters of recording and replaying holographic registrations marks are the same there will be positional inaccuracies when the holographic registration mark is replayed. 10 It is therefore and object of the embodiments disclosed herein to avoid or mitigate one or more of the disadvantages discussed above. Specifically, an aim of embodiments is to provide a solution which can be utilised for a range of holographic recording angles and wavelengths. Summary According to one aspect, a machine vision system for imaging a target surface includes a light 15 source for illuminating the target surface; an image sensor for capturing at least one image of the target surface; wherein the target surface includes an image plane holographic registration mark and the captured image includes a replayed image of the image plane holographic registration mark. According to another aspect, the machine vision system includes a light source that is an 20 extended array light source. According to yet another aspect, the extended array light source comprises a plurality of point sources arranged as a two-dimensional array. According to another aspect, the extended array light source further comprises a diffuser panel. 25 According to another aspect, each of the point sources in the extended array light source is arranged to emit light at substantially the same wavelength and bandwidth. According to another aspect, the light source aligns with respect to the target surface to replay an image of the image plane holographic registration mark. According to another aspect, the target surface comprises holographic photosensitive 30 material. 19 06 24 According to another aspect, the holographic photosensitive material comprises one of a photopolymer based material, a silver halide based material or a dichromated gelatin based material. According to another aspect, the image plane holographic registration mark is a reflection 5 hologram or a transmission hologram. According to another aspect, a system for aligning a target surface with a work surface comprises the machine vision system and a mechanical alignment system; wherein the mechanical alignment system is arranged to hold the work surface and comprises a translation stage, a height adjustment stage and a rotation stage. 10 According to another aspect, the translation stage in the system for aligning a target surface with a work surface is configured and arranged to align in response to an alignment signal from the machine vision system such that the work surface registers with respect to the image plane holographic registration mark. According to another aspect, a method of imaging a target surface includes illuminating the 15 target surface with a light source; capturing images of the target surface using an image sensor; wherein the target surface includes an image plane holographic registration mark and the captured image includes a replayed image of the image plane holographic registration mark. According to another aspect, a method of aligning a target surface with a work surface includes 20 imaging a target surface; the method further includes holding the work surface with a mechanical alignment system and moving the work surface into alignment with the target surface with the mechanical alignment system. According to another aspect, the method of aligning a target surface with the work surface includes aligning the translation stage in response to an alignment signal from the machine 25 vision system such that the work surface is registered with respect to the image plane holographic registration mark. The present disclosure therefore provides a machine vision system which is capable of detecting and reading a wide range of holographic registration marks, recorded under different reference beam geometries. In turn this overcomes any issues with known systems which 30 may require changes to and recalibration of machine vision systems dependent on the recording geometry used to record the holographic registration mark. In other words, replay of the holographic registration mark is invariant to recording setup. 19 06 24 So that the features of the present disclosure can be understood in detail, a more particular description is made with reference to embodiments, some of which are illustrated in the appended figures. It is to be noted, however, that the appended figures illustrate only typical embodiments and are therefore not to be considered limiting of its scope. The figures are for 5 facilitating an understanding of the disclosure and thus are not necessarily drawn to scale. It should be noted that the features as illustrated in the figures have been exaggerated for illustration purposes and no dimensions (unless stated in the text or drawings) should be inferred. Advantages of the embodiments will become apparent to those skilled in the art upon reading this description in conjunction with the accompanying figures, in which like reference 10 numerals have been used to designate like elements, and in which: Figure 1a shows reconstruction, or replay, of a holographic registration mark using three light sources P, P+ and P-; Figure 1b shows reconstruction, or replay, of a holographic registration mark, using a 15 spatially extended light source array; Figure 2 illustrates a side view of a general block diagram of the machine vision system according to embodiments; Figure 3a illustrates a side view of an alignment system incorporating the machine vision system according to embodiments; 20 Figure 3b illustrates an end view of an alignment system incorporating the machine vision system according to embodiments; and Figure 4 is a process flow illustrating a method of imaging and aligning a target surface with a work piece. 25 Description In overview Figure 2 illustrates a machine vision system 100 according to embodiments, configured and arranged to capture images of a target surface 106 and more particularly, 30 replayed images of a holographic registration mark 110 formed in the target surface 106. The machine vision system 100 comprises a spatially extended light source array 102 and an image sensor 104. The spatially extended light source array 102 is configured to illuminate 19 06 24 the target surface 106 such that the image sensor can capture images of the target surface 106. One example of an image sensor 104 suitable for such a machine vision system 100 a digital camera such as the Cognex (RTM) IS9912 Camera. One example of control software for the imaging sensor is In-Sight Explorer 6.50 vision software which in the present example 5 creates a job file and this file then runs on the Cognex (RTM) IS9912 Camera internal microprocessor. The image sensor resolution and lens choice depends on the size and accuracy required in the final product. One example of the extended light source array 102 is OMRON (RTM) QBR3-100030R which is a 100mm x 30mm active area LED bar light. The image sensor may be a CCD, CMOS or other suitable image sensor. 10 The target surface 106 comprises a volume hologram registration or fiducial mark 110 formed within the local thickness of the material and the extended light source array 102 is specifically configured to replay the holographic registration mark or fiducial marks 110 such that it is visible by the image sensor 104. In this regard the target surface 106 may comprise a holographic photosensitive material suitable for recording the holographic registration mark or 15 fiducial marks 110 therein. The photosensitive material capable of recording holograms therein, which may be photopolymer-based material, a silver-halide based holographic emulsion, dichromated gelatin. Methods and holographic recording set-ups for formation of the holographic registration or fiducial marks 110 (herein after holographic registration mark(s)) in the photosensitive material are outside the scope of the present disclosure. 20 However, the holographic registration mark 110 is recorded as an image plane hologram. In the present case the image would be an image of the desired registration mark to be recorded into the photosensitive material. An image plane hologram is any hologram in which the image of an object or the object itself is located near (that is between 0.0 mm and ±3mm, and ideally between ±1mm) the hologram recording plane. In the case where the image location is 25 positive, that is for example +3mm, the image will appear above the hologram recording plane or the target surface 106. Where the image location is negative, that is for example -3mm, the image will appear on the other side of the hologram recording plane (or the target surface). This merely indicates that the hologram may be replayed from either side, that is the hologram can be rotated 180 degrees about the surface in which the registration mark 110 is recorded 30 and still be replayed. One of the key features of image plane holograms is their high resolution and clarity. Since the image is reconstructed in the plane of the hologram, it can be viewed with a level of detail that closely matches the original object therefore making it suitable for use as a holographic registration mark 110. The holographic registration mark 110 may be a reflection type 35 hologram as discussed below, but the skilled person will also appreciate that it may also be a transmission type hologram or a Fourier transform hologram. 19 06 24 The spatially extended light source array 102 comprises multiple point sources 108 arranged to illuminate the target surface 106 and thus replay the holographic registration mark 110. Optionally, a diffuse surface may be placed over the point sources. The addition of a diffuse surface results in a change in light output characteristics for the extended array, most notably 5 the array of point sources change from a finite number to an infinite number of point sources. The multiple point sources are arranged as a grid or a two-dimensional n x m array, where n and m are positive integers. The size of the array, that is the values of n and m are chosen based on the size of the target surface. In the present example a 6 x 20 array is used to illuminate the target surface, and arranging the point sources in this way ensures an even 10 distribution of light over the entire target surface 106. In this arrangement, light from at least one of the point sources in the light source array 102 will be incident on the target surface 106 at the correct angle to suitably reconstruct the holographic registration mark 110. This is true for holographic registration marks anywhere in the illuminated area of the target surface 106 within the field of view of the image sensor 104. 15 Each of the multiple point sources 108 may have substantially the same wavelength of emission and have substantially the same line width (full width half maximum). The wavelength of emission of the point sources 108 may be matched to the replay wavelength of holographic registration mark 110 to avoid the inaccuracies mentioned above. The point sources 108 may be light emitting diodes. Light emitting diodes are preferred over laser diodes as point sources 20 because the light emitting diodes do not suffer from the problem of laser speckle which can cause accuracy problems when trying to image small features such as holographic registration marks 110. Advantageously, the spatially extended light source array 102 results in the target surface 106 being illuminated by light with a wide range of angles. As a result, the machine vision system 25 will be capable of reading different holograms each recorded with different reference beam geometries, that is different holograms may be replayed and imaged using one spatially extended light source array 102. In other words, the range of light angles emitted by the spatially extended light source array 102 could cover the range of angles of the reference beams used to record holographic registration marks 110. This operation over a wide range 30 of angles means that the machine vision system will be invariant to the recording beam reference angle over a range of about 0 Degrees (that is normal to the target surface) to about 89 Degrees to the normal because the extended array provides light covering all of these angles. Therefore, the holographic registration marks 110 will always be reconstructed regardless of the recording geometry and this in conjunction with the image plane holographic 35 registration marks 110 results in a system which has high positional accuracy. 19 06 24 The emission wavelength of the point sources 108 is chosen to match the wavelength of the lasers (within 10nm) used to record the holographic registration marks 110. For example, the laser used to record the holographic registration marks 110 may be a red laser and the point sources 108 may be LEDs chosen to operate in the red portion of the visible spectrum to 5 replay the holographic registration marks 110. Other wavelengths for point sources 108 may be chosen dependent on the recording wavelength of the holographic registration mark 110, for example infra-red. The holographic registration mark 110 may be a reflection hologram and the image sensor 104 and the extended light source array 102 may be arranged on the same side opposing the 10 target surface 106. The extended light source array 102 is aligned with the target surface 106 such that a ray of light from the centre of the array 102 is angled at the reference angle used to record the holographic registration mark 110. The image sensor 104 is aligned with respect to the target surface 106 and captures images of the image plane holographic Registration Mark 110, be that in reflection or in transmission 15 as mentioned above. The image sensor 104 is equipped with a microprocessor including algorithms for distortion correction, displacement correction and so on. The image sensor captures images of the target surface 106, and in an ideal situation the replayed image of the holographic registration mark 110. As mentioned above, the holographic registration mark 110 is formed as an image plane holographic registration mark 110 and the holographic 20 registration mark will be replayed in the plane of the photosensitive material target surface 106. To image the replayed image of the holographic registration mark 110 the image sensor 104 can be focused on the upper surface of the target surface 106 facing the image sensor. The foregoing describes the ideal situation where the holographic registration mark will be replayed “in-plane”. However, in certain circumstances (outside the scope of the present 25 disclosure) the replayed image may be displaced from the plane of the target surface 106 by a known amount in which case a suitable displacement correction algorithm may be implemented the image sensor control software to take account of this image displacement so that the replayed image is accurately read by the image sensor 104. In the case where the holographic registration mark 110 is a transmission hologram the image 30 sensor 104 and the extended light source array 102 may be arranged on opposing sides of the target surface 106. Figure 3a illustrates a side view and respective end view of a workpiece alignment system 300 incorporating the machine vision system 100 discussed above. The machine vision system 100 is used to control a moveable x-y-z-rotation stage 314, which is moveable in response to 35 a control signal from the machine vision system 100. The x-y-z-rotation stage 314 also 19 06 24 includes a chuck, mount or platform (not illustrated) which is suitable for holding or mounting a workpiece 312 thereon. The workpiece 312 can thus be aligned with respect to the target surface. The x-y-z-rotation stage 314 is equipped with linear actuators or motors (not illustrated), which may include stepper motors, servo motors, or piezoelectric actuators, each 5 selected based on the requirements of precision, and speed. The inclusion of such actuators ensures that the x-y-z-rotation stage 314 can achieve fine adjustments in position with a high degree of accuracy in response to the control signal from the machine vision system 100. The x-y-z-rotation stage 314, may be made up of an individual x-stage 320, a y-stage 322 a z-stage 324 and a rotation stage 326 arranged to move the chuck and thus the workpiece 10 relative to the target surface 106. The x-stage 320 is arranged to move the chuck in a first horizontal axis, x, relative to a horizontal plane of the target surface 106. The y-stage 322 is arranged to move the chuck along a second horizontal axis, y, relative to the horizontal plane of the target surface 106. The x- and y- stages are arranged to move orthogonally with respect to each other in parallel with the horizontal plane of the target surface 106. The z-stage is 15 arranged to vary the height of the chuck, that is move the chuck along the vertical axis. In this regard the vertical axis is a direction perpendicular to both the x stage axis and the y stage axis of movement. The chuck may include the rotation r-stage 326 which rotates about the z-stage axis or vertical axis. Each of the x-stage 320, y-stage 322, z-stage 324 and rotation stage may be arranged to move independently of each other in response to control signal from 20 the machine vision system 100. The target surface 106 may be a photosensitive material such as photopolymer. The photopolymer may be provided as a roll of material and transferred from a first transfer roller 316 to a second transfer roller 318 by means of a suitable web transport mechanism. The holographic registration marks 110 have been formed within the photopolymer by any suitable 25 process which is outside of the scope of this disclosure. When a holographic registration mark 110 enters the field of view of the machine vision system 100 the workpiece 312 may be aligned with the holographic registration marks on the photopolymer by the x-y-z-rotation stage 314 as discussed in more detail below. The skilled person will appreciate therefore that the workpiece 312 may serve as a substrate to which the photosensitive material is applied. The 30 workpiece 312 can therefore be positioned with a high degree of precision onto the target surface 106 thus providing accurate and controlled application of the photosensitive material to the workpiece 312 facilitating a repeatable and exact application processes. In this way, the skilled person will appreciate that the x-stage 320, y-stage 322, and the rotation -stage 326 allow precise alignment of a workpiece with respect to the holographic registration mark 110 35 in the photopolymer dependent on control signals from the machine vision system 100. The z-stage 324, by lifting the workpiece up to the photopolymer may bring the workpiece into 19 06 24 contact with the photopolymer, thus allowing the workpiece to be bonded to the photopolymer by any appropriate bonding mechanism. The image sensor 104 illustrated in Figures 3a and 3b, is shown mounted on the y-stage 322, however, the skilled person will appreciate that the image sensor 104 may be located in any 5 suitable location provided that it can accurately read the holographic registration marks 110 as mentioned above. However, in the case where the image sensor is located on the opposite side of the target surface 106 to the light source 108 the skilled person will understand that the holographic registration mark 110 should be a transmission type hologram rather than a reflection type hologram. The rollers of Figure 3a, have been omitted from Figure 3b to aid 10 clarity. Referring now to the process flow of Figure 4, an example process of imaging and aligning a target surface with a workpiece is described. Prior to this process, the workpiece is loaded onto the chuck of x-y-z-rotation stage and the material comprising the target surface 106 is arranged in the field of view of machine vision system 100. Following loading of the workpiece 15 and arrangement of the target surface 106, the target surface 106 is illuminated (Step 402) by the spatially extended light source array 102. As discussed above, the target surface 106, which may be a photosensitive material such as a photopolymer, contains one or more image plane holographic registration marks 110. The target surface 106, and more specifically the one or more image plane holographic registration marks 110 are illuminated and replayed or 20 reconstructed by the spatially extended light source array 102. The image sensor 104 then captures images of the replayed image plane holographic registration mark (Step 404), as described above. The captured images are then processed by the processor of the image sensor 104 (Step 406). The step of processing can include locating the position of the holographic registration 25 marks 110 with respect to a known reference point. This reference point could be a known point on the chuck of the x-y-z-rotation stage 314. Based on the displacement position of the holographic registration mark 110 with respect to the reference point a control signal is generated to control the x-y-z-rotation stage 314 and also rotation of the chuck (Step 410). The x-y-z-rotation stage 314 is then operated, using the control signal, to move the workpiece 30 relative to the holographic registration mark (Step 412). The present system addresses the aforementioned challenges by providing a machine vision system for imaging a target surface 106 that includes an image plane holographic registration mark 110. The system comprises a light source for illuminating the target surface and the image sensor for capturing images of the target surface 106. The light source can be an 35 extended array light source, which includes a plurality of point sources arranged in a two 17 07 24 dimensional array, potentially with a diffuser panel. The system is designed to replay an image of the image plane holographic registration mark, thereby facilitating accurate alignment with a work surface. This approach offers a solution to the problems of distortion and parallax error associated with conventional imaging systems, while also addressing the limitations of size 5 and integration in manufacturing environments. Whilst the foregoing description relates to machine vision systems for reading holographic registration marks and fiducials, the skilled person will also appreciate that the foregoing disclosure also relates to holographic product codes such as QR codes or barcodes, or holographic branding marks such as logos or trademarks. 10 Particular and preferred aspects of the disclosure are set out in the accompanying independent claims. Combinations of features from the dependent and / or independent claims may be combined as appropriate and not merely as set out in the claims. The scope of the present disclosure includes any novel feature or combination of features disclosed therein either explicitly or implicitly or any generalisation thereof irrespective of 15 whether or not it relates to the claimed disclosure or mitigate against any or all of the problems addressed by the present disclosure. The applicant hereby gives notice that new claims may be formulated to such features during prosecution of this application or of any such further application derived therefrom. In particular, with reference to the appended claims, features from dependent claims may be combined with those of the independent claims and features 20 from respective independent claims may be combined in any appropriate manner and not merely in specific combinations enumerated in the claims. Features which are described in the context of separate embodiments may also be provided in combination in a single embodiment. Conversely, various features which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any 25 suitable sub combination within the scope of the appended claims. The term “comprising” does not exclude other elements or steps, the term “a” or “an” does not exclude a plurality. Reference signs in the claims shall not be construed as limiting the scope of the claims. 30

Claims

19 06 245 1. A machine vision system for imaging a target surface, the machine vision systemcomprising:a light source for illuminating the target surface wherein the light source is an extended array light source comprising a plurality of point sources arranged as a two-dimensional array, and10 an image sensor for capturing at least one image of the target surface;wherein the target surface comprises an image plane holographic registration mark and the captured image comprises a replayed image of the image plane holographic registration mark.

2. The machine vision system of claim 1 wherein the extended array light source further 15 comprises a diffuser panel.

3. The machine vision system of claim 1, wherein the each of the point sources are arranged emit light at substantially the same wavelength and bandwidth.

4. The machine vision system of claim 1, wherein the light source is aligned with respect to the target surface to replay an image of the image plane holographic registration mark.20 5. The machine vision system of any preceding claim wherein the target surface isholographic photosensitive material.

6. The machine vision system of claim 5, wherein the holographic photosensitive material comprises one of a photopolymer based material, a silver halide based material, or a dichromated gelatin based material.25 7. The machine vision system of claim 6, wherein the image plane holographicregistration mark is a reflection hologram or a transmission hologram.

8. A system for aligning a target surface with a work surface comprising the machinevision system of any preceding claim and mechanical alignment system wherein the mechanical alignment system is arranged to hold the work surface and comprises a translation 30 stage and a height adjustment stage and a rotation stage.19 06 249. The system of claim 8, wherein the translation stage is configured and arranged to be aligned in response to an alignment signal from the machine vision system dependent on captured images of the replayed holographic registration mark such that the work surface is registered with respect to the image plane holographic registration mark.5 10. A method of imaging a target surface, the method comprising:illuminating the target surface with light source, wherein the light source is an extended array light source comprising a plurality of point sources arranged as a two-dimensional array;capturing images of the target surface using an image sensor;processing images of the target surface10 wherein the target surface comprises an image plane holographic registration mark and the captured image comprises a replayed image of the image plane holographic registration mark.

11. A method of aligning a target surface with a work surface, the method comprising the method of imaging a target surface as claimed in claim 10, the method further comprising holding the work surface with a mechanical alignment system and moving the work surface 15 into alignment with the target surface with the mechanical alignment system dependent on captured images of the replayed holographic registration mark.

12. The method of aligning a target surface with the work surface of claim 11, wherein the translation stage is aligned in response to an alignment signal from the machine vision system such that the work surface is registered with respect to the image plane holographic 20 registration mark.25

Images