Method and apparatus for high resolution measurement of workpieces

Abstract

Methods and systems for inspecting a workpiece (e.g., a honeycomb) are described herein. The methods and systems include collecting a plurality of images of the honeycomb, extracting measurement data from each of the plurality of images, converting the measurement data extracted from each image into a common frame of reference, and combining the measurement data together.

Description

[0001] Cross-reference to related applications

[0002] This application claims priority to U.S. Provisional Application No. 62 / 928,847, filed October 31, 2019, pursuant to 35 U.S. SC §120, which is based on and incorporated herein by reference in its entirety. Background Technology

[0003] Cellular structures are used in a variety of applications, such as particulate filters for treating unwanted components in working fluids (e.g., pollutants in combustion exhaust gases) and in the construction of catalytic converters. The fabrication of cellular structures may involve examining the characteristics of the cellular structure. Summary of the Invention

[0004] The various methods described herein provide improvements for inspecting honeycomb structures or honeycomb extrusion dies. In these methods, multiple images derived from portions of the honeycomb structure are first converted to a common reference frame and then combined, rather than generating a composite image of the entire honeycomb structure and measuring from the composite image. Furthermore, improved devices for imaging honeycomb structures are described and can be configured to reduce pattern deviations in the honeycomb structure images.

[0005] In one aspect, a method for measuring features of a workpiece is provided. The method includes: capturing a first image of a first portion of the workpiece using a camera of an imaging system; determining a first set of multiple measurements of the first portion of the workpiece from the first image in a first reference frame of the first image; moving at least one of the workpiece or the camera relative to each other; capturing a second image of a second portion of the workpiece using the camera, wherein the first portion and the second portion at least partially overlap and share common features; determining a second set of multiple measurements of the second portion of the workpiece from the second image in a second reference frame of the second image; identifying common features in the first image and the second image, wherein the common features have a first position in the first reference frame and a second position in the second reference frame; determining a spatial relationship between the first reference frame and the second reference frame based on a comparison of the first position and the second position; converting the first set of multiple measurements, the second set of multiple measurements, or both, to a common reference frame, at least partially based on the spatial relationship; and generating a set of dimensional features of the workpiece by combining the first and second sets of multiple measurements when converted to the common reference frame.

[0006] In some embodiments, the movement includes moving at least one of the workpiece or camera a predetermined distance, and the spatial relationship between the first reference frame and the second reference frame is at least partially based on the predetermined distance.

[0007] In some embodiments, the method further includes determining the error in the relative movement of at least one of the workpieces or cameras compared to each other.

[0008] In some embodiments, the error is determined by comparing the predetermined distance with the difference between a first and a second position in a shared reference frame. In some embodiments, the spatial relationship between the first and second reference frames is based on the predetermined distance and the error.

[0009] In some embodiments, the common reference frame is the first reference frame. In some embodiments, the common reference frame is the second reference frame.

[0010] In some implementations, the shared reference frame is a third reference frame that is different from the first and second reference frames.

[0011] In some implementations, both the first and second reference frames are oriented relative to the Cartesian coordinate system.

[0012] In some implementations, the method further includes removing duplicate measurements after combining the first set of multiple measurements and the second set of multiple measurements in a common reference frame.

[0013] In some implementations, the shared reference frame includes a scale defined between two points.

[0014] In some embodiments, the workpiece is a honeycomb structure defining multiple longitudinal channels. In some embodiments, a common feature is the centroid of the longitudinal channels of the honeycomb structure.

[0015] In some embodiments, the workpiece includes a honeycomb structure and the length of the wall of the longitudinal channels of the honeycomb structure as a common feature.

[0016] In some embodiments, the dimension includes the distance between the first centroid of the first channel of the honeycomb and the second centroid of the second channel of the honeycomb.

[0017] In some embodiments, the imaging system includes a camera, a lens, a first light source, and a second light source.

[0018] In some embodiments, the first light source is configured to provide bright field illumination, and the second light source is configured to provide dark field illumination.

[0019] In some embodiments, the lens defines an optical axis, the first light source defines a first illumination axis, and the second light source defines a second illumination axis, wherein the first illumination axis forms an angle α with respect to the optical axis of 0° to 20°, and the second illumination axis forms an angle θ with respect to the optical axis of 70° to 90°.

[0020] In some implementations, the first light source is a ring light, and the second light source is a ring light.

[0021] In some implementations, the workpiece is simultaneously illuminated by a first light source and a second light source when capturing the first and second images.

[0022] In some implementations, the workpiece is static relative to the camera of the imaging system when capturing the first and second images.

[0023] In some embodiments, the method further includes: capturing a third image of a third portion of a workpiece, wherein the common feature is a first common feature and a second common feature that overlaps with and includes the second image; determining a third set of multiple measurements of the third portion of the workpiece from the third image in a third reference frame of the third image; identifying the second common feature in the second and third images, wherein the second common feature has a third position in the second reference frame and a fourth position in the third reference frame; comparing the third position and the fourth position, and determining a second spatial relationship between the second reference frame and the third reference frame based on the comparison;

[0024] Based at least in part on the second spatial relationship, the third set of multiple measurements is converted to a common reference frame; and by converting to the common reference frame, the third set of multiple measurements is incorporated into a scale group.

[0025] In one aspect, a method for measuring features of a workpiece is provided. The method includes: providing an imaging system; capturing an image of a first portion of the workpiece, wherein the image of the first portion of the workpiece defines a reference frame; calculating a first set of multiple measurements based at least partially on the image of the first portion of the workpiece; capturing an image of a second portion of the workpiece, wherein the first portion and the second portion of the workpiece include multiple common features, wherein the second portion includes at least one feature not included in the first portion of the workpiece; calculating a second set of multiple measurements based at least partially on the image of the second portion of the workpiece, wherein at least one of the first set of multiple measurements is a first reference scale defined by the common features, wherein at least one of the second set of multiple measurements is a second reference scale defined by the common features, wherein the first reference scale is defined at least partially based on the image of the first portion of the workpiece by a scale between the common features and the second reference scale is defined at least partially based on the image of the second portion of the workpiece by the same scale between the common features; and calculating a transformation mode by comparing the first reference scale and the second reference scale.

[0026] The transformation method is applied to the second group of multiple measurements to transform the second group of multiple measurements into a reference system; and the first group of multiple measurements is combined with the transformed second group of multiple measurements.

[0027] In another aspect, an imaging system for measuring the dimensions of a workpiece is provided. The system includes: a camera configured to capture an image of a workpiece; an actuator configured to move the camera relative to the workpiece or to move the workpiece relative to the camera; a controller data-communicated with the camera and the actuator and configured to cause an imaging system to capture a first image of a first portion of the workpiece via the camera; a first set of multiple measurements determining features of the workpiece from the first image, wherein the first set of multiple measurements is defined by a first reference frame relative to the first image; positioning the field of view of the camera relative to a second portion of the workpiece via the actuator, wherein the second portion overlaps with the first portion and includes common features with the first portion; capturing a second image of the second portion of the workpiece; determining a second set of multiple measurements determining features of the workpiece from the second image, wherein the second set of multiple measurements is defined by a second reference frame relative to the second image; identifying a first position of a common feature in the first image and a second position of a common feature in the second image; determining a spatial relationship between the first reference frame and the second reference frame based on a comparison of the first position and the second position; converting the first set of multiple measurements, the second set of multiple measurements, or both, to a common reference frame based on the spatial relationship; and generating a set of dimensional features of the workpiece by combining the first and second sets of multiple measurements when converted to the common reference frame.

[0028] In some embodiments, the actuator is configured to position the camera such that at least one of the workpiece or the camera moves a predetermined distance, and the spatial relationship between the first reference frame and the second reference frame is at least partially based on the predetermined distance.

[0029] In some embodiments, the controller is also configured to determine an error in the movement of at least one of the workpiece or the camera relative to each other when the field of view of the camera is positioned relative to the second part of the workpiece.

[0030] In some implementations, the error is determined by comparing the predetermined distance with the difference between the first and second positions in a common reference frame.

[0031] In some implementations, the spatial relationship between the first reference frame and the second reference frame is based on the predetermined distance and the error.

[0032] In some implementations, the common reference frame is either the first reference frame or the second reference frame.

[0033] In some implementations, the shared reference frame is a third reference frame that is different from the first and second reference frames.

[0034] In some implementations, the workpiece is a honeycomb structure defining multiple longitudinal channels.

[0035] In some implementations, the common feature is the centroid of the longitudinal channels of the honeycomb.

[0036] Other implementation methods are described in the detailed embodiments below. Attached Figure Description

[0037] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate embodiments of the invention and, together with the description, further serve to explain the principles involved and enable those skilled in the art to make and use the disclosed techniques.

[0038] Figure 1 This is a perspective view of an exemplary honeycomb structure;

[0039] Figure 2 yes Figure 1 An exemplary end view of a cellular structure.

[0040] Figure 3 This is a side view of an exemplary imaging system that can be used in accordance with embodiments disclosed herein.

[0041] Figure 4 This is an end view of an exemplary cell.

[0042] Figures 5A-5C This is a partial end view of an exemplary cellular structure.

[0043] Figures 6A-6C This is a partial end view of an exemplary cellular structure.

[0044] Figure 7 A flowchart illustrating an exemplary measurement method for features of a workpiece (e.g., a honeycomb structure) according to embodiments disclosed herein.

[0045] The features and advantages of the disclosed technology will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which the same reference numerals consistently identify corresponding elements. In the drawings, the same reference numerals generally denote identical, functionally similar, and / or structurally similar elements. The first appearance of an element in a graphic is indicated by the leftmost numeral in the corresponding reference numeral. Detailed Implementation

[0046] The following detailed description relates to the accompanying drawings, which illustrate exemplary embodiments of the invention. However, the scope of the invention is not limited to these embodiments, but rather defined by the appended claims. Therefore, the invention may also include more than those embodiments shown in the drawings (e.g., modifications of the illustrated embodiments).

[0047] In this specification, references to "one embodiment," "an embodiment," or "an exemplary embodiment" indicate that the described embodiment may include specific features, structures, or characteristics, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, such expressions do not necessarily refer to the same embodiment. In addition, when a specific feature, structure, or characteristic is described in connection with an embodiment, it means that those skilled in the art can influence such feature, structure, or characteristic in combination with other embodiments, whether or not it is explicitly described.

[0048] The exemplary embodiments described herein provide improvements over known methods and systems for measuring cellular features. An exemplary embodiment of the method for measuring workpiece features includes: capturing an image of a first portion of the workpiece; determining a first set of multiple measurements based on the image of the first portion; capturing an image of a second portion of the workpiece; determining a second set of multiple measurements based on the image of the second portion; locating a common feature in an overlapping region between each image; converting the first set of multiple measurements, the second set of multiple measurements, or both, to a common reference frame using the location of the common feature in the images; and combining the measurements in the common reference frame.

[0049] Compared to known systems (e.g., those employing flatbed scanners and microscope-based imaging systems), the improvements offer several advantages. These improvements include combining measurement data, rather than stitching images together to create a composite image. This reduces errors in the measurement data that would arise when determining measurements on a composite image formed by stitching images together. In exemplary embodiments, this reduction in errors in the measurement data can be used to more accurately predict workpiece performance characteristics (e.g., isostatic strength). The improvements also reduce pattern deviations in the images while decreasing image capture time on the end faces of the workpiece (e.g., a honeycomb structure or a honeycomb extrusion die). In some cases, the improvements also provide higher resolution images of specific parts of the workpiece. Furthermore, the improvements are effective for workpieces constructed from both dark and light materials and enable the inspection of the entire end face of the workpiece within a shorter timeframe.

[0050] Figure 1 and 2An exemplary honeycomb structure 100 is shown. The honeycomb structure 100 includes a plurality of spaced-apart and intersecting inner walls 102 (or networks) extending longitudinally from a first end face 104 through the honeycomb structure 100 to a second end face 106. The inner walls 102 combine to define a plurality of channels 108 (or pores) that form a pore-like honeycomb structure of the honeycomb structure 100. The honeycomb structure 100 may also include perimeter channels 110, typically partial channels, intersecting with the outer skin 112 of the honeycomb structure 100. As shown, the honeycomb structure 100 includes channels 108 having a square cross-sectional shape, but the channels 108 may also have other cross-sectional shapes, such as triangular, hexagonal, octagonal, wedge-shaped, or combinations of these or other shapes. Similarly, as shown, the honeycomb structure 100 has a circular cross-sectional shape, but other shapes may also be used, such as rectangular, square, triangular or trefoil-shaped, or elliptical, etc. The honeycomb 100 defines a longitudinal axis L that extends between the first end face 104 and the second end face 106 and is substantially parallel to the longitudinal axis of the channel 108.

[0051] The honeycomb 100 can be formed in any desired manner, for example, by extruding the ceramic mixture through an extrusion die to form a green body, drying the green body, cutting the green body to a certain length, and firing the green body to form a ceramic material. The ceramic material of the honeycomb 100 can be a porous ceramic material. The honeycomb is inspected when it is in the green body stage (before firing) or when it is in the ceramic stage (after firing). The honeycomb 100 can be used in catalytic converter assemblies by loading catalytic material onto the walls 102, and / or in particulate filter assemblies by blocking some channels 108 (e.g., alternately blocking channels 108 at the inlet and outlet faces).

[0052] See Figure 3 An exemplary imaging system 320 that can be used to capture high-resolution images of a cell 100 will be described herein. The imaging system 320 can be used to capture images of a portion of the cell 100, which can be analyzed using machine vision software to collect measurement data of the features of the cell 100. In some embodiments, the imaging system 320 is configured to collect images of the cell 100 having a resolution of at least 5000 x 5000 pixels, at least 10000 x 10000 pixels, or even at least 20000 x 20000 pixels. The imaging device 320 includes a camera 322 and a lens 324. As discussed herein, the imaging device 320 may also include a first light source 326, a second light source 328, a component holder 330, and a controller 332.

[0053] A camera 322 is disposed on a first side of the cell 100 and configured to capture high-resolution images of the end faces (e.g., inlet face 104 or outlet face 106) of the cell 100. The camera 322 may be a digital camera configured to record digital image data corresponding to the cell 100, thereby enabling the collection of measurement data on the features of the cell 100. The digital image data is based at least in part on an image of the cell 100 that passes through a lens 324 and is projected onto a digital imaging sensor in the camera 322. The camera 322 may be configured to collect monochrome or multicolor image data. Exemplary digital cameras that may be used are: the Dalsa Falcon4 (86MP digital camera); the Prosilica GT 6600 (28.8MP digital camera); and the Adimec S25A80 (25MP digital camera), but other cameras are also possible. In some embodiments, the camera 322 has a resolution related to the physical scale of the cell 100, corresponding to approximately 4-50 μm per pixel (e.g., μm per pixel).

[0054] Lens 324 has an optical axis OP. Lens 324 can be integrated into camera 322 or optically coupled to camera 322 in any other way, such that an image of cell 100 (e.g., in the form of reflected light from cell 100) passes through lens 324 and is guided to camera 322 (e.g., to an imaging sensor of camera 322). Lens 324 can provide any selected magnification to provide the desired scale resolution. The lens can be constructed as a telecentric or macro lens. In an exemplary embodiment, the lens is a telecentric lens having 1x magnification, 0.03° telecentricity, and 0.02% distortion. An exemplary lens that can be used includes the TC16M036 lens supplied by Opto Engineering of Houston, TX.

[0055] The imaging system may include a movable stage 334. The movable stage 334 may be configured to provide relative movement between the cell 100 and the camera 322, for example, causing the camera 322 and lens 324 to move toward and away from the cell 100, for example, parallel to... Figure 3 The Z-axis direction is shown. A movable stage 334 can be used to provide relative movement, thereby placing a selected portion of the honeycomb 100 within the depth of field of the lens 324. In an exemplary embodiment, the depth of field of the lens 324 is less than 5 mm, and in another exemplary embodiment, the depth of field of the lens 324 is approximately 1 mm.

[0056] The first light source 326 can be arranged on a first side of the honeycomb structure 100, i.e., on the same side of the honeycomb structure 100 as the camera 322 and the lens 324. This position allows the first light source 326 to directly illuminate the end face of the honeycomb structure 100 closest to the lens 324. The first light source 326 can be arranged adjacent to and coupled to the lens 324. The first light source 326 can be a high-intensity monochromatic ring light, which is generally annular and surrounds the optical axis OP (and field of view) of the lens 324. The first light source 326 can be constructed from multiple light sources (e.g., light-emitting diodes (LEDs) distributed around the optical axis). In some embodiments, the light source of the first light source 326 is selected to emit uniform monochromatic light of a selected color (e.g., monochromatic green light).

[0057] like Figure 3 As shown, the first light source 326 can be configured to provide direct bright-field illumination to the honeycomb 100. In some embodiments, the first light source 326 provides bright-field illumination in which the illumination axis forms a low illumination angle α relative to the optical axis OP of the lens 324, for example, ranging from about 0° (parallel) to about 45°, or from about 0° to 20°. The low illumination angle α causes light reflected off the end face of the honeycomb 100 to return directly to the lens 324. In an exemplary embodiment, the first light source 326 is configured as a ring light, and in some embodiments, the first light source 326 is coupled to the lens 324. In an exemplary embodiment, the illumination angle α provided by the light source 326 relative to the optical axis OP of the lens 324 is less than 20°, in another exemplary embodiment, the illumination angle α is less than 15°, and in yet another exemplary embodiment, the illumination angle α is from about 10° to about 0°.

[0058] Imaging system 320 may optionally include a second light source 328 configured to provide dark-field illumination to the cell 100. The inventors have determined that adding a light source providing dark-field illumination is useful for reducing pattern aberrations (i.e., image portion distortion at the end faces of the cell 100) present in some images captured only with bright illumination. In some embodiments, the illumination axis provided by the second light source 328 forms a high illumination angle θ relative to the optical axis OP of the lens 324, for example, ranging from about 45° to about 90°, or about 70° to 90°. In some embodiments, the second light source 328 is constructed as a ring of light similar in configuration to the first light source 326, but provides a different illumination angle. In an exemplary embodiment, the illumination angle θ provided by the second light source 328 relative to the optical axis OP of the lens 324 is greater than 45° or even greater than 70°, greater than 75%, or about 75° to about 90°.

[0059] The second light source 328 can be positioned close to the honeycomb cell 100 to provide dark-field illumination. The second light source 328 can be arranged as close as possible to the honeycomb cell 100 without risk of collision during relative movement between the honeycomb cell 100 and the imaging system 320. In an exemplary embodiment, the distance between the second light source 328 and the honeycomb cell 100 is less than 10 mm, and in another exemplary embodiment, it is approximately 6-7 mm.

[0060] The component holder 330 is configured to hold and / or position the honeycomb 100 in a desired orientation, thereby enabling imaging of selected portions of the honeycomb 100. The component holder 330 includes a movable stage 336 (e.g., an XY stage and / or a tilting stage) so that the honeycomb 100 can move relative to the camera 322 and lens 324. The relative movement of the honeycomb 100 with respect to the camera 322 and lens 324 along the X, Y, and Z axes can be accomplished using actuators directly connected to the component holder 330, directly connected to the camera 322 and lens 324, or both.

[0061] Controller 322 can control the relative motion between cell 100 and imaging system 320, image capture, image processing, and the combination of measurement data, as described herein. Controller 332 may include a processor, data storage, and a display. Along with the hardware components, controller 322 may include software configured to perform operations on components of imaging system 320, such as causing camera 322 to capture images or causing component holder 330 to adjust the relative position of cell 100 and imaging system 320. Furthermore, controller 332 may be configured to perform feature measurements by executing imaging measurement software. Controller 332 may also include image acquisition and processing software that provides a user interface for collecting and processing images.

[0062] In some embodiments, the imaging system 320 includes a distance sensor 340. The distance sensor 340 can be used to determine the presence or absence of the honeycomb 100 and / or to determine the position of the honeycomb 100, for example, by measuring the distance between the honeycomb 100 and the lens 324. The distance between the honeycomb 100 and the lens 324 can be used to control the movable stage 334 of the camera 322 and the lens 324, as well as the movable stage 338 of the honeycomb 100, thereby placing the honeycomb 100 within the depth of field of the lens 324 for imaging. In some embodiments, the distance sensor 340 is a laser line profilometer.

[0063] Imaging system 320 can be used to measure various physical characteristics of the honeycomb 100. For example, the dimensions of walls 102 and / or channels 108 can be measured. In some embodiments, features captured and measured by imaging system 320 include physical defects, i.e., geometries (e.g., dimensions or shape) that differ from the designed geometry of the honeycomb 100. For example, walls with cracks, fissures, tears, or gaps can be identified as defects. Geometric defects may occur during the extrusion or other manufacturing processes of the honeycomb 100, and those defects may alter the properties of the honeycomb 100 (e.g., isostatic compressive strength).

[0064] See Figure 4 Multiple images of the cell 100, covering different portions of the cell 100, are captured. The cell 100 and the imaging system 320 move relative to each other to image different portions of the cell 100, and the movement can be step-by-step or continuous. In an exemplary embodiment, images are captured by the imaging system 320 traversing the cell 100 in a step-by-step manner along a two-dimensional grid pattern oriented above the end face of the cell 100, for example as... Figure 4 The grid shown. During the capture of each image, the cell 100 and the imaging system 320 can remain fixed relative to each other. Capturing images along the grid pattern results in multiple images, arranged in rows 444a-g and columns 446a-g. The grid size can be adjusted based on the specific part size (e.g., the diameter of the selected cell). Alternatively, the grid can be large enough to contain the largest cell, allowing the grid to capture images of any workpiece up to that maximum size.

[0065] Collecting multiple images on the cell 100 can result in image resolution higher than any other possible with a camera used to collect the images. That is, each image can be configured to capture a portion of the cell 100 instead of the entire cell. As a result, the entire resolution of the camera can be dedicated to capturing only a portion of the cell 100 in each image.

[0066] like Figure 4 As shown, the grid pattern can be cropped so that the multiple images do not necessarily contain the same number of images in each row or column. For example, it is not necessary to collect images located in the grid pattern that do not capture any part of the cell 100.

[0067] Each captured image at least partially overlaps with one or more adjacent images. The positions of the adjacent images are such that the portion of the cell 100 captured in the adjacent images contains at least one common feature. For example, the common feature could be the centroid of a specific channel 108, a specific corner of a specific channel 108, or other visually identifiable features. As described herein, the overlap of adjacent images and the common features captured by the images provide information for combining measurement data obtained from the multiple images. As discussed in more detail below, the positions of the identified common features in the adjacent images are used to determine the spatial relationships between the adjacent images, and then used to transform the measurement data determined from the adjacent images to a common reference frame. After the measurement data is transformed to the common reference frame, the data can be combined to produce a set of feature scales for the cell 100 without stitching the images together.

[0068] exist Figure 4 In this implementation, the first row 444a of the images comprises five images. For ease of discussion herein, three such images are identified as first image 448, second image 450, and third image 452. The first image 448 can be captured from a first relative position between the cell 100 and the imaging system 320. The second image 450 can be captured from a second relative position between the cell 100 and the imaging system 320. The third image 452 can be captured from a third relative position between the cell 100 and the imaging system 320, and so on for each sequential image in each row and column. Adjacent images in a row include overlapping regions. For example, an overlapping region 449 is created between the first image 448 and the second image 450 by selecting a distance between the first and second relative positions that is less than the width of the portion of the cell 100 as shown in the first image 448. The overlapping region 449 ensures that common features of the cell are displayed in adjacent images, i.e., consistent features displayed in both the first image 448 and the second image 450. Similarly, the second image 450 also overlaps with the third image 452, as shown in the overlapping area 451. Furthermore, adjacent images arranged in each column may include overlapping areas. For example, the first image 448 in the first row 444a of column 446b overlaps with the second image 456 in the second row 444b in the overlapping area 453. In the exemplary grid pattern, each image displays a square field of view of 5mm to 50mm, for example, a square field of view of approximately 20mm to 30mm, and the step size between images is approximately 50% to 90% of the field of view width. For example, in one embodiment, for image sizes of approximately 20mm to 25mm, a step size of approximately 15mm is used, thereby creating an overlap of approximately 5mm to 10mm between adjacent images.

[0069] The measurement data generated from each image can include image identification measurements and pore property measurements. Image identification measurements can include the image X index (i.e., the x-coordinate of a known location on the image, such as the x-coordinate of the image's centroid) and / or the image Y index (i.e., the y-coordinate of a known location on the image, such as the y-coordinate of the image's centroid). By using the X and Y indices, the relative positions of the images with respect to each other can be established. For example, see... Figure 4 For example, the X index can be associated with the column where each image is located, while the Y index can be associated with the row where each image is located. The proximity of images can also be determined when the relative positions (e.g., rows and columns) are known.

[0070] Orifice property measurements may include orifice wall angles, horizontal and vertical orifice pitches, horizontal and vertical wall thicknesses, horizontal and vertical wall sags, shear angles, network deformation, orifice area and aspect ratio, perimeter, etc. Measurement data can be extracted from images using machine vision software by performing one or more processes (e.g., the processes contained in controller 332). Examples of systems and methods for extracting scale data from images are described in US 8,285,027, published by Zoeller, III, October 9, 2012, and US 9,996,766, published by Richard et al., June 12, 2018, both of which are incorporated herein by reference in their entirety. One or more processes may be integrated within an imaging system or they may be separate processes communicating over a network. In examples, computer vision libraries such as OpenCV may be employed.

[0071] Therefore, as described above, measurement data of various features captured in each of the images (e.g., scale data of the pores 108 and / or walls 102 of the cell 100) can be collected. Advantageously, even when extracting partial measurement data from multiple different images, system 320 is able to combine the measurement data into a common data set. Specifically, system 320 is still able to combine such measurement data even when errors are introduced during the movement of the camera or cell between sequentially captured images. Furthermore, as described herein, system 320 can generate such combined measurement data sets without stitching all images into a single composite image. Advantageously, avoiding the need to stitch images together significantly reduces the computational time and resources required to extract measurement data from captured images, while maintaining the image to be analyzed at a pixel resolution that would otherwise be infeasible.

[0072] See Figures 5A-5C Other operational details of the imaging system 320 will be readily apparent. As described above, by capturing multiple images (e.g., the first image 670 and the second image 672, e.g., similar to those about...) Figure 4Images 448 and 450 image the cell 100. For clarity, the field of view of the first image 670 is represented by a dotted-dash line (–.–.–), while the field of view of the second image 672 is represented by a double-dash line (–..–..–). The first and second images 670 and 672 are adjacent in position and share an overlapping region 678 (e.g., similar to…). Figure 4 The overlapping region 449). Therefore, the features (e.g., physical structure or geometry) of the honeycomb 100 in the portion shown in the overlapping region 678 are shared by both the first image 670 and the second image 672 (and also by the corresponding first and second portions of the honeycomb 100 shown in the first image 670 and the second image 672).

[0073] exist Figures 5A-5C In the example, the centroid C1 of the first channel 674 is located in the overlapping region 678, and therefore can be used as a shared feature of both the first image 670 and the second image 672. That is, both the first image 670 and the second image 672 show the centroid C1 of the first channel 674, thus allowing the centroid C1 to be identified in both images. By controlling and monitoring the states of the movable stages 334 and 336, the position of the cell 100 relative to the camera 322 (and / or lens 324) when the first image 670 and the second image 672 are captured is known, for example, relative to... Figure 5A The global reference frame is shown. A global reference frame can be used to determine the position of images relative to each other, such as the X and Y indices discussed above. For example, referring again to the example in Figure 5, the global coordinate system can be a system in which rows and columns are defined. Therefore, the camera 322 and / or the cell 100 can move relative to each other in the X and / or Y directions of the global coordinate system at predetermined steps (i.e., the step size described above), thereby capturing each image sequentially.

[0074] As described in more detail below, although the position of the camera 322 (and / or lens 324) relative to the cell 100 is referred to as "known" above, there is a degree of uncertainty or error regarding the accuracy of this position. Therefore, each captured image will have a reference frame separate from the overall reference frame, for example, in... Figure 5A In the image, the first image 670 has a first reference frame and the second image 672 has a second reference frame. The reference frames can use the same coordinate system (e.g., X and Y axes) and orientation as all reference frames (e.g., the X and Y axes of the first and second reference frames of the first and second images 670 and 672 can be parallel to the X and Y axes of the global coordinate system, respectively). For consistency, the image reference frames can be set at the same position relative to each image, for example, relative to... Figures 5A-5C The orientation is located at the lower left of each image.

[0075] Therefore, the centroid C1 is located in the first position, as shown in... Figure 5B The first image 670 shown has first coordinates (X11, Y11) within a first reference frame defined by a first image. As used herein, the coordinate Xnm is used to represent the X-axis coordinate of feature “n” in the “m” reference frame. Therefore, X11 represents a first common feature (e.g., centroid C1) in the first reference frame, while X12 represents a first common feature (e.g., centroid C1) in the second reference frame. Thus, after the camera 322 is moved relative to the cell 100 to capture the second image 672, the same centroid C1 is also located in a second position, as shown in the image. Figure 5C The second reference frame defined by the second image 672 shown has second coordinates (X12, Y12). For example, for Figure 5A In this context, the first position, relative to the first image 670, faces right-hand side from a first reference frame, while the second position, relative to the second image 672, faces far-left side from a second reference frame. In other words, the value of coordinate X11 is expected to be greater than the value of coordinate X12 because these coordinates are determined relative to their respective reference frames, rather than relative to all reference frames.

[0076] See Figure 5A If there is no movement error in the relative position of the cell 100 with respect to the camera 322 (and / or lens 324) between the capture of the first and second images, then the field of view of the first image 670 is offset by a predetermined distance S in the global reference frame relative to the field of view of the second image 672. That is, the predetermined distance S represents the step size performed by the controller 332 instructing the movable stages 334 and / or 336. Therefore, if there is no error in the movement and if the field of view size does not change, the second coordinate (X12, Y12) in the second reference frame can be obtained by subtracting the predetermined distance (e.g., distance S) from the first coordinate (X11, Y11) of the centroid C1. In the example shown, the second image 672 is offset only relative to the first image 670 in the X direction (i.e., the first image 670 and the second image 672 are in the same row, and therefore the Y index or coordinate in the global reference frame should be the same). Therefore, at least in part, a measurement determined in one reference frame (e.g., the coordinates of the centroid C1) can be transformed to another reference frame (assuming no error) based on a distance S according to the following coordinate relationship.

[0077] If the adjacency between the first and second images is in the X direction, the step size is S:

[0078] X12 = X11 – S; and (1)

[0079] Y12 = Y11. (2)

[0080] If the adjacency between the first and second images is in the Y direction, the step size is S:

[0081] X12 = X11; and (3)

[0082] Y12 = Y11 - S. (4)

[0083] Based on those relationships, the coordinates of features measured from either the first image 670 or the second image 672 can be transformed into either reference frame. For example, by subtracting the distance S from each X-coordinate while keeping the Y-coordinate unchanged, coordinates measured in the first reference frame defined by the first image 670 can be transformed into coordinates in the second reference frame defined by the second image 672. Therefore, this transformation enables either the first or second reference frame to be used as a common reference frame. Measurements of other features corresponding to the cell 100 can also be transformed into the common reference frame in a similar manner. For example, the centroid Cn of the pore “n” shown in the second image, whose coordinates in the second reference frame are (Xn2, Yn2), can be transformed into coordinates in the first reference frame. When each desired measurement is transformed into the common reference frame, they can be combined into a single common measurement data set without combining image data from multiple images.

[0084] Measurements of other features shown in the second image 672 can also be determined relative to shared features in the second image 672. Specifically, measurements of features shown only in the second image 672 can be combined with features shown only in other images by associating them with shared features. For example, measurements (e.g., coordinates) of the centroid Cn shown only in the second image 672 can be determined. In the second image 672, the shared features (centroid C1) and the centroid Cn are spaced apart by a distance dx in the X direction and a distance dy in the Y direction, resulting in the following relationship for the coordinates:

[0085] Xn2=X12+dx; and (5)

[0086] Yn2 = Y12 + dy. (6)

[0087] When converting multiple measurements derived from multiple images into a common reference frame, the measurements can be combined into a single common measurement data set.

[0088] Although the example described refers to a second image 672 that is displaced by a distance S relative to the first image 670 only in the X direction, images can be offset in either or both of the X and Y directions, thereby changing the coordinate relationships. For example, combining measurement data from multiple images (e.g., as...) Figure 4In the grid pattern shown, different spatial relationships relative to a common reference frame need to be applied to each image.

[0089] Additional images can be captured, and measurements can be collected from the images. Measurement data generated from subsequent images can be converted to a common reference frame using transformations from intermediate adjacent images. For example, measurements generated from a third image (which is spaced apart from the first image 670 so that they do not contain any overlapping areas) can be converted to a common reference frame by combining measurement transformations between the first image 670 and the second image 672, and measurement transformations between the second image 672 and the third image.

[0090] In another example, measurements can be transferred to a common reference frame after the positional error has been determined and / or corrected. For example, the positional error can be determined by comparing the expected position of a common feature in a given reference frame with the actual position of that common feature in that reference frame. For example, as described above, it is expected that X12 = X11 – S and Y12 = Y11 in the example described above. However, when the camera 322 and the cell 100 are moved relative to each other between each successive image capture, an error between the expected and actual positions is introduced. For example, the second position of the centroid C1 will be a position in the second image 672 that differs from the expected second position, such as... Figure 5C The dashed line indicates this. Specifically, the actual second position is displayed as C1' with coordinates (X12', Y12'), while the expected second position has coordinates (X12, Y12). In the example shown, the error in the relative position results in an error in the X coordinate (e). x1 ) and Y coordinate error (e y1 And the coordinates can be transformed using the following relationship:

[0091] If the adjacency between the first and second images is in the X direction, the step size is S:

[0092] X12' = X11 – S + ex1; and (7)

[0093] Y12' = Y11 + ey1. (8)

[0094] If the adjacency between the first and second images is in the Y direction, the step size is S:

[0095] X12' = X11 + ex1; and (9)

[0096] Y12' = Y11 - S + ey1. (10)

[0097] Based on those relationships, by adjusting the distance S and the error in the X-axis for each X-coordinate and simultaneously adjusting the error in the Y-axis for the Y-coordinate, the feature coordinates measured from the second image 672 each time can be accurately converted from coordinates in the second reference system defined by the second image 672 to coordinates in the first reference system defined by the first image 670.

[0098] See Figures 6A-6C The multiple shared features displayed in the overlapping region can also be used to determine any scaling errors (variations in the field of view size of lens 324 when capturing different images) before combining the measurement data. First image 770 shows a first portion of the honeycomb 100 including a first feature (e.g., the centroid C1 of the first channel 774) and a second feature (e.g., the centroid C2 of the second channel 776). Second image 772 shows a second portion of the honeycomb 100 and includes at least a portion of the first portion of the honeycomb 100 to define an overlapping region 778. First image 770 and second image 772 are sized and oriented such that the overlapping region 778 includes the centroid C1 of the first channel 774 and the centroid C2 of the second channel 776. As a result, both first image 770 and second image 772 show the centroid C1 of the first channel 774 and the centroid C2 of the second channel 776.

[0099] The centroids provide shared features in the first image 770 and the second image 772. Measurement data generated from the images can be normalized using a scale that can be measured in the first image 770 and the second image 772 (e.g., the distance between the centroids) to ensure that the images display portions of the cell 100 at the same scaling factor or magnification. For example, generating measurement data from the images may include generating the distance between centroids C1 and C2 based on the first image 770, and generating the distance between centroids C1 and C2 based on the second image 772. The distance between centroids C1 and C2 extracted from the first image 770 defines a first reference scale R1, such as... Figure 6B As shown. The distance between the centroids C1 and C2 extracted from the second image 772 defines the second reference scale R2, as follows. Figure 6C As shown, the scaling error is calculated by comparing a first reference scale R1 with a second reference scale R2. This can be applied to measurement data obtained from multiple images, thus standardizing all measurement data before combining them into a common set of measurement data.

[0100] Next, the data combining process involves combining measurement data into a single common set of measurement data. For example, measurement data extracted from a baseline image is combined with transformed measurement data extracted from subsequent images into a single set of measurement data for the entire cell 100. The measurement data combining step may also include removing duplicate measurement data caused by extracting measurements from overlapping areas of the images. As a result, a single set of measurement data is produced, which reduces errors caused by imaging and is free of duplicate measurement data.

[0101] Figure 7 A flowchart 560 shows the measurement of features of a workpiece (e.g., a honeycomb structure 100). Since the honeycomb extrusion die has features corresponding to those of the honeycomb structure 100 (e.g., the slit-forming wall 102 of the extrusion die and the pin-forming channel 108 of the extrusion die), the imaging system 320 and method 560 can be used to inspect the dimensions of the pins and slits of the honeycomb contact die, because both the honeycomb structure and the honeycomb extrusion die are workpieces having the honeycomb pattern described herein. For example, Figure 2 The image can also represent a honeycomb extrusion die, with reference numeral 108 indicating a pin and reference numeral 102 indicating a slit formed between the pins. Examples of such methods include... Figure 3 As shown and relative to Figure 4 , 5A The imaging system 320 described in -5C and 6A-6C is illustrated in flowchart 560. Based on the discussion of flowchart 560, other structural and operational implementations will be apparent to those skilled in the art.

[0102] like Figure 7 As shown, the method in flowchart 560 begins at step 562. In step 562, the workpiece (e.g., the honeycomb structure 100) is inspected by loading the workpiece into an imaging system (e.g., imaging system 320). For example, the honeycomb structure 100 may be loaded onto a component holder 330.

[0103] In step 564, the position of the workpiece is determined. Specifically, the position of the honeycomb cell 100 relative to the imaging system 320 is determined. In an exemplary embodiment, the position of the honeycomb cell 100 is determined by measuring the distance between the lens 324 and the honeycomb cell 100. The distance can be measured using a distance sensor 340 of the imaging system 320. The distance can be measured at multiple locations on the workpiece so that the angle of the end face of the workpiece relative to the optical axis of the lens 324 can be determined.

[0104] In step 566, the workpiece is positioned relative to the camera (e.g., camera 322 and / or its lens 324) to provide the alignment and position required for imaging a portion of the workpiece. In this example, the honeycomb 100 is positioned relative to the lens 324 by displacing it in the XY plane and rotating it about the X-axis and / or Y-axis, thereby placing the desired portion of the honeycomb 100 within the field of view of the camera 322 and the lens 324. The honeycomb 100 may be tilted such that its end face is normal to the optical axis of the lens 324 to improve imaging. The movement in the XY plane may correspond to... Figure 4 The position of the honeycomb 100 relative to the lens 324 can be determined by any combination of movable stages (e.g., movable stages 334 and / or 336 for the camera 322 and / or lens 324). The movable stages are selected such that the accuracy and repeatability of the relative position between the honeycomb and the imaging system 320 are known to fall within a predetermined tolerance. For example, the accuracy of the movable stages is selected such that the position of a common feature shown in adjacent images is known to fall within a tolerance of half the aperture width. In another exemplary embodiment, the accuracy of the movable stages is selected such that the position of a common feature shown in adjacent images is known to fall within a tolerance of 20 pixels. In this way, even if there is some error during movement that causes the feature to not be exactly in the expected position, the feature can still be identified because it falls within a small tolerance.

[0105] Step 566 may further include setting the relative Z-axis position between the workpiece and the imaging system at a distance such that the workpiece is within the depth of field of the imaging system. In an exemplary embodiment, the camera 322 and lens 324 are displaced relative to the honeycomb 100 in a direction parallel to the Z-axis via the movable stage 334. This movement places the end face of the honeycomb 100 within the depth of field of the lens 324, thereby bringing a portion of the honeycomb 100 into focus. The combination of movements in steps 566 and 568 effectively provides automatic leveling and automatic focusing of the workpiece in the imaging system.

[0106] In step 568, an image is captured. For example, the honeycomb is illuminated and an image of at least a portion of the workpiece is captured. In an exemplary embodiment, a processor (e.g., included in controller 332) determines the desired location of the honeycomb in the grid (e.g., having the desired X and / or Y coordinates or indices), guides the first and second light sources to illuminate the workpiece, and / or guides camera 322 to capture an image of the honeycomb 100.

[0107] In step 570, features of the workpiece are measured by analyzing the captured images. In other words, measurement data related to the feature scales of the honeycomb 100 (e.g., the scale of the walls 102 and / or channels 108) are extracted from the captured images. The images are used to identify the type, location, and scale of the features of the honeycomb 100. For each image, measurement data corresponding to the features of the honeycomb 100 are compiled. In an exemplary embodiment, the honeycomb 100 and the imaging system 320 are arranged in a stepped manner through a grid pattern (e.g., Figure 4 The position in the grid pattern shown is captured, and an image is captured and analyzed at each position.

[0108] In step 572, shared features in the overlapping areas between each pair of adjacent images are used to transform the measurement data into a shared reference system, allowing the measurement data to be combined into a single measurement data set in step 574. In this way, although measurement data is extracted from multiple images with different reference systems, the combined measurement data set corresponds to the entire workpiece. In an exemplary embodiment, the measurement data can be standardized so that all measurement data corresponds to a single coordinate system, and the standardized measurement data can be transformed into a single reference system and combined. In an exemplary embodiment, images from the multiple images can be selected to define a reference system used as a shared reference system for all measurement data. Alternatively, a global reference system different from any reference system from the multiple images can be used as the shared reference system for all measurement data. Measurement data from adjacent images is collected, including measurement data applied to shared features captured in adjacent images. Shared features enable the transformation of measurement data acquired from multiple images into a shared reference system, allowing all measurement data to be combined into a single measurement data set.

[0109] Common features can be one or more features of the cell 100 captured in two images. Common features can be the corner pillars of a single channel, the centroid of a channel, the centroids of multiple channels, etc., displayed in the overlapping region between the two images. Common features can be measurements of geometric defects, such as: curved walls, discontinuous walls, wall thickness anomalies, or other wall defects. Based on common features, spatial relationships between images (e.g., relative offsets in the X and / or Y directions) can be determined and can be used to transform measurement data from a reference frame of any image to a common reference frame, thereby combining all measurements into a single set of measurement data based on the common reference frame.

[0110] Based on a predetermined expected offset distance, the identification of shared features can also be used to determine the positional error of the images relative to each other. For example, based on the predetermined expected offset, the position of a shared feature measured in an image can be compared with the expected position of the shared feature in that image to define the error of the relative movement between the cell 100 and the imaging system 320 between the captured images. The predetermined expected offset distance, incorporating any error, can be used to define the spatial relationship between the images, corresponding to the spatial relationship between a first reference frame and a second reference frame.

[0111] Furthermore, the scale between multiple shared features in adjacent images can be used to determine errors (e.g., scaling errors). For example, the same scale can be measured in adjacent images to provide a first reference scale and a second reference scale. The measured scale values ​​are compared to determine whether an error (e.g., scaling error) exists between adjacent images. If no error exists, the first and second reference scales are the same. However, if imaging errors exist, the reference scales will be different, and the difference can be used to constrain the transformation. The transformation can be applied to the measurement data to obtain transformed measurement data that has been standardized relative to a shared reference frame.

[0112] In an exemplary embodiment, each image is analyzed to extract measurement data before moving to the next position, thereby repeating steps 566, 568, 570, and 572 until the entire end face of the cell 100 has been imaged and the measurement data has been analyzed. However, it is not necessarily necessary to perform steps 570 and / or 572 on each image immediately after step 568. Instead, the images and / or the extracted measurement data can be stored (e.g., stored in the data memory in the controller 332) and analyzed later.

[0113] In step 574, the measurement data are combined (e.g., via controller 332). As discussed herein, the measurement data generated from each of the multiple images are directly combined rather than generating a composite image and extracting measurements from the composite image.

[0114] Although the subject matter has been described using language specific to structural features and / or behaviors, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or behaviors described above. Rather, the specific features and behaviors described above are disclosed as examples for practicing the claims, and other equivalent features and behaviors are intended to fall within the scope of the claims.

Claims

1. A method for measuring features of a workpiece including a honeycomb structure or a honeycomb extrusion die, the method comprising: A first light source located on a first side of the workpiece illuminates the first end face of the workpiece, the first light source providing bright field illumination to the first end face; A camera using an imaging system captures a first image of a first portion of a first end face; From the first image, a first set of multiple measurements of the first portion are determined in a first reference frame of the first image; This causes at least one of the workpieces or the camera to move relative to each other; A second image of a second portion of a first end face is captured using a camera, wherein the first portion and the second portion at least partially overlap and share common features; From the second image, a second set of multiple measurements of the second portion is determined in a second reference frame of the second image, wherein the first and second sets of multiple measurements correspond to the scale of the cross walls of the honeycomb or the scale of the cross slits of the honeycomb extrusion die; The common features in the first image and the second image are identified, wherein the common features have a first position in the first reference frame and a second position in the second reference frame; Based on the comparison between the first position and the second position, the spatial relationship between the first reference frame and the second reference frame is determined; Based at least in part on spatial relationships, the first set of multiple measurements, the second set of multiple measurements, or both, are transformed to a common reference frame; and A set of dimensional features of the workpiece is generated by combining the first and second sets of multiple measurements when transformed to a common reference frame, without forming a synthetic image.

2. The method as described in claim 1, wherein, The movement includes at least one of the workpieces or cameras moving a predetermined distance, and the spatial relationship between the first reference frame and the second reference frame is at least partially based on the predetermined distance.

3. The method of claim 2, further comprising determining an error in the relative movement of at least one of the workpieces or the camera compared to each other.

4. The method of claim 3, wherein, The error is determined by comparing the predetermined distance with the difference between the first and second positions in a shared reference frame.

5. The method of claim 3, wherein, The spatial relationship between the first reference frame and the second reference frame is based on the predetermined distance and the error.

6. The method according to any one of claims 1-5, wherein, The shared frame of reference is the first frame of reference.

7. The method according to any one of claims 1-5, wherein, The shared frame of reference is the second frame of reference.

8. The method according to any one of claims 1-5, wherein, The shared frame of reference is a third frame of reference that is different from the first frame of reference and the second frame of reference.

9. The method according to any one of claims 1-5, wherein, Both the first and second reference frames are oriented relative to the Cartesian coordinate system.

10. The method of any one of claims 1-5, further comprising removing duplicate measurements after combining the first set of multiple measurements and the second set of multiple measurements in a common reference frame.

11. The method according to any one of claims 1-5, wherein, Shared features include the scale defined between two points.

12. The method according to any one of claims 1-5, wherein, If the workpiece is a honeycomb structure, the common feature is the centroid of the longitudinal channels; and if the workpiece is a honeycomb extrusion die, the common feature is the centroid of the pins.

13. The method according to any one of claims 1-5, wherein, The workpiece includes a honeycomb structure, and a common feature includes the length of a wall in the longitudinal channels of the honeycomb structure.

14. The method according to any one of claims 1-5, wherein, The dimension includes the distance between the first centroid of the first channel of the honeycomb and the second centroid of the second channel of the honeycomb, which are defined by the cross walls of the honeycomb.

15. The method according to any one of claims 1-5, wherein, The imaging system includes a second light source disposed between the first light source and the first end face.

16. The method of claim 15, wherein, The second light source illuminates the first end face with dark field illumination.

17. The method of claim 15, wherein, The lens defines the optical axis, the first light source defines the first illumination axis, and the second light source defines the second illumination axis, wherein the first illumination axis forms an angle α with respect to the optical axis of 0° to 20°, and the second illumination axis forms an angle θ with respect to the optical axis of 7° to 90°.

18. The method of claim 17, wherein, The first light source is a ring light, and the second light source is a ring light.

19. The method of claim 15, wherein, When capturing the first and second images, the first end face is simultaneously illuminated by the first light source and the second light source.

20. The method according to any one of claims 1-5, wherein, When the first and second images are captured, the workpiece is stationary relative to the camera of the imaging system.

21. The method of claim 1, further comprising: A third image of the third portion of the first end face is captured, wherein the shared feature is a first shared feature and the third portion overlaps with and includes a second shared feature of the second image; From the third image, a third set of multiple measurements of the third portion are determined in a third reference frame of the third image; In the second image and the third image, the second common feature is identified, wherein the second common feature has a third position in the second reference frame and a fourth position in the third reference frame; Compare the third position and the fourth position, and determine the second spatial relationship between the second reference frame and the third reference frame based on the comparison; Based at least in part on the second spatial relationship, the third set of multiple measurements are transformed to a common reference frame; and The third set of multiple measurements, when converted to a common reference frame, is combined into a scale group instead of forming a composite image.

22. A method for manufacturing a ceramic honeycomb, comprising the method as described in claim 1.

23. A method for measuring features of a workpiece including a honeycomb structure or a honeycomb extrusion die, the method comprising: A first image of a first portion of a first end face of a workpiece is captured by a camera of an imaging system, the first image of the first portion defining a reference frame; The first set of multiple measurements is calculated based at least in part on the first image of the first portion; The field of view of the camera is positioned relative to a second portion of the first end face by an actuator. The first and second portions include a plurality of shared features, and the second portion includes at least one feature not included in the first portion. Capture an image of the second portion of the first end face; A second set of multiple measurements is calculated based at least in part on the image of the second portion, wherein the first and second sets of multiple measurements correspond to the scale of the cross walls of the cell or the scale of the cross slits of the cell extrusion die, wherein at least one of the first set of multiple measurements is a first reference scale defined by common features, wherein at least one of the second set of multiple measurements is a second reference scale defined by common features, wherein the first reference scale is defined by a scale at least in part based on the common features between the images of the first portion, and the second reference scale is defined by the same scale at least in part based on the common features between the images of the second portion; The transformation method is calculated by comparing the first reference scale and the second reference scale. The transformation method is applied to the second set of multiple measurements to convert the second set of multiple measurements to a reference system; and The second set of multiple measurements is obtained by combining the first set of multiple measurements and transformations, without forming a composite image.

24. An imaging system for measuring the dimensions of a workpiece including a honeycomb structure or a honeycomb extrusion die, the system comprising: A first light source is configured to illuminate the first end face of the workpiece with bright field illumination. A camera configured to capture an image of the first end face; An actuator configured to move a camera relative to a workpiece or a workpiece relative to a camera; The processor, which communicates data with the camera and actuators and is configured to cause the imaging system: A first image of a first portion of a first end face is captured by a camera, wherein the first image defines a frame of reference; At least a first set of multiple measurements is determined based on the first image; The field of view of the camera is positioned relative to a second portion of the first end face by means of an actuator, wherein the first portion and the second portion include a plurality of shared features, and the second portion includes at least one feature not included in the first portion; Capture the second image of the second part; A second set of multiple measurements is determined based at least on the second image of the second portion, the first and second sets of multiple measurements corresponding to the scale of the cross walls of the honeycomb or the scale of the cross slits of the honeycomb extrusion die, wherein at least one of the first set of multiple measurements is a first reference scale defined by common features, at least one of the second set of multiple measurements is a second reference scale defined by common features, the first reference scale being defined at least partially based on the scale between common features of the first image, and the second reference scale being defined at least partially based on the same scale between common features of the second image; The transformation method is calculated by comparing the first reference scale and the second reference scale. The transformation method is applied to the second set of multiple measurements to convert the second set of multiple measurements to a reference system; and The second set of multiple measurements is obtained by combining the first set of multiple measurements and transformations, without forming a composite image.

25. The system of claim 24, wherein, The actuator is configured to position the camera such that at least one of the workpiece or the camera moves a predetermined distance, and the spatial relationship between the first reference frame and the second reference frame is at least partially based on the predetermined distance.

26. The system of claim 25, wherein, The controller is also configured to determine the error in the movement of at least one of the workpiece or the camera relative to each other when the field of view of the camera is positioned relative to the second part of the workpiece.

27. The system of claim 26, wherein, The error is determined by comparing the predetermined distance with the difference between the first and second positions in a shared reference frame.

28. The system of claim 27, wherein, The spatial relationship between the first reference frame and the second reference frame is based on the predetermined distance and the error.

29. The system of claim 24, wherein, The shared frame of reference is either the first frame of reference or the second frame of reference.

30. The system of claim 24, wherein, The shared frame of reference is a third frame of reference that is different from the first frame of reference and the second frame of reference.

Images