Method and system for deriving a digital representation of an unfolded blank and for cost estimation based thereon

Abstract

A computer-implemented method for creating a computer-aided design (CAD) corresponding to a 2-dimensional rendering of an unfolded blank configured for manipulation into a 3-dimensional shape. The method includes obtaining a first digital non-CAD design file containing information related to the unfolded blank geometry, but lacking metadata defining cut lines or crease lines separately from surrounding content, and deriving, using a computer processor, a digital representation of the unfolded blank geometry based on the first digital non-CAD design file. The digital representation includes defined data corresponding to a shape having one or more defined cut lines and / or crease lines. A system for performing the method includes a computer processor and a machine-readable medium accessible by the computer processor, the machine-readable medium including non-transitory instructions readable by the computer processor for performing the method steps of defining the digital non-CAD design file and deriving the digital representation therefrom.

Description

Background Art

[0001] In the packaging industry, folding carton packaging manufacturing companies (commonly referred to as "converters") often receive quote requests for producing a specified number of boxes. In order to be able to determine a price, converters need to estimate the production cost of the specified number of boxes. The production cost needs to be relatively accurate because a quote that is too low may result in financial losses, while a quote that is too high may result in business losses. Estimates of production costs can include, but are not limited to, material costs (folding carton board, ink, foil, varnish, and the like), tooling costs (printing plates, molds, and the like), manual labor, machine costs for printing and finishing equipment, and shipping costs. Material costs, and more specifically the cost of folding carton board, are one of the main elements in such production cost estimates.

[0002] To determine the cost of folding carton board, the converter needs to determine how many sheets of what size stock are needed to produce a specified number of boxes. For example, the flat, unfolded shape 100 depicted in FIG1 represents the outline geometry of the unfolded "blank" to be cut from a sheet of carton stock and then folded into a box. If the converter plans to use a 750mm x 350mm sheet of stock to produce boxes, then he or she must determine how many blanks will fit on the sheet to minimize waste. This requires generating a layout such as the layout 200 depicted in FIG2 below, in which two blanks 100 fit on a sheet 210.

[0003] Given the order quantity for boxes, the converter can now determine that he needs a number of sheets equal to half the order quantity, plus setup waste on each machine.

[0004] For simplicity, a single quantity of a single blank geometry on a single sheet size is referenced in the above examples, but in practice, a customer may request prices for multiple quantities of multiple blank geometries (e.g., staggered pricing, such as 1000, 10,000, 100,000 units). The converter may want to determine production costs for different sheet sizes. These different sheet sizes may be sheet sizes available in stock, or there may be different sheet sizes for different production methods (e.g., a sheet size for printing on a large format offset press plus trimming on a large format die cutter, and an additional sheet size for printing on a smaller digital press plus trimming on a small format die cutter).

[0005] Responding to such a quote request requires generating, evaluating, and comparing many layouts. Ideally, the translator has a CAD design file format that is supported (e.g. ArtiosCAD TM ARD file, CFF2 file, or other files known to those skilled in the art, although as described herein, other file types with isolated or isolable information corresponding to the geometry of the blank may be considered suitable CAD design files for the purpose of easy import into a CAD system). Using the CAD design of the blank, a CAD system such as EskoArtiosCAD TM ArtiosCAD software to create layouts. TM The software has functionality to automatically determine the optimal layout (ie, the one with the highest number of blanks on the sheet).

[0006] However, in many cases, the prospective customer does not have access to the digital data. In some cases, the geometry of the blank may be embedded as a contour in a graphic file (e.g., PDF), from which the blank geometry must be extracted and converted from other surrounding objects into the CAD design of the blank. For PDFs with the blank geometry embedded as discrete layers, this extraction may be straightforward. However, sometimes the blank geometry is contained in the same layer or separation as other information, and time-consuming manual processing is required to extract the blank geometry from the cluttered graphic file in a form usable by the CAD software.

[0007] In other cases, the prospective customer only has a physical sample. In this case, the converter now has multiple options:

[0008] • A rough "guesstimate" can be made of the number of blanks that should fit into the sheet based on a visual assessment. The risk of error with this approach is high.

[0009] • The physical sample may be manually positioned on the physical sheet by a human operator and shifted around to find a good layout. Again, the risk of error is high and this is time consuming.

[0010] A physical sample can be given to a CAD designer, who measures the sample and creates a digital CAD file representing the physical sample. One advantage of this method is that after the CAD file is created, the layout tools described above can be used. While this method produces accurate results, it is time-consuming.

[0011] The workflow for the physical sample method can be characterized as having the following steps:

[0012] 1. A Customer Service Representative (CSR) at the converter receives a request for quotation from a print buyer, such as, for example, along with a physical sample of a box.

[0013] 2. CSR sends a design request to the CAD designer.

[0014] 3. CAD designers use measurements of the dimensions of the physical sample to create the CAD drawing.

[0015] 4. The CAD designer creates a set of CAD drawing layouts for the set of alternative sheet sizes.

[0016] 5. The CAD designer provides the results back to the CSR.

[0017] 6. CSR calculates the price of materials needed to produce the required number of boxes.

[0018] 7.CSR adds other production costs and finalizes the price quoted to the customer.

[0019] Therefore, there is a need in the art for systems and methods that create CAD designs of shapes such as packaging blanks from non-CAD files, such as images of physical samples or cluttered graphic files, and for systems and methods that use such CAD designs in a rapid and automated manner to quickly determine the quantity of sheet material (and other costs) required to produce a given quantity of packaging blanks. Summary of the Invention

[0020] One aspect of the present invention includes a computer-implemented method for creating a computer-aided design (CAD) file corresponding to a 2-dimensional rendering of an unfolded blank configured to be manipulated into a 3-dimensional shape, such as the unfolded blank configured to be manipulated into the 3-dimensional shape of a packaging box along cut lines and crease lines. The method includes the steps of obtaining, in a computer processor, a first digital non-CAD design file containing information related to the geometry of the unfolded blank, the non-CAD design file not having metadata defining cut lines or crease lines separately from surrounding content, and deriving, using the computer processor, a digital representation of the geometry of the unfolded blank based on the first digital non-CAD design file. The digital representation includes defined data corresponding to a shape having one or more defined cut lines and / or crease lines.

[0021] In some embodiments, the first digital non-CAD design file comprises an image file, and the image file may include markings having known geometry and dimensions, wherein deriving the digital representation comprises using the markings to determine the dimensions of the unfolded blank. The step of deriving the digital representation may also include using the markings to identify and compensate for artifacts in the image, such as rotation relative to a preferred rotation, perspective relative to a plan view, and scaling. The markings may include a plurality of fiducials, including one or more QR codes distributed in a pattern, such as three QR codes positioned at positions corresponding to vertices of a right triangle. The markings may embody machine-readable encoded information as size and position information about the plurality of fiducials, or the encoded information may be machine-readable as identification information of an address on a machine-accessible network, wherein machine-readable size and position information about the plurality of fiducials is stored.

[0022] A method of deriving a digital representation from an image file may include, after compensating for artifacts in an image to create a straightened image, a processor performing the step of generating a CAD design mask from the straightened image, the mask including a first set of one or more cut lines forming a boundary. The mask may then be applied to the straightened image, a line detection algorithm may be performed within the portion of the image bounded by the mask, and lines connected on one side to the first set of cut lines are identified as a second set of one or more cut lines, and lines connecting two lines from either the first or second set of cut lines are identified as crease lines.

[0023] In other embodiments, the information in the non-CAD design file associated with the unfolded blank geometry includes one or more lines in a scrambled graphics file, such as a PDF file lacking metadata defining cut lines or crease lines. In such embodiments, the step of deriving the digital representation may include a processor executing substeps comprising first collecting lines and curves from the scrambled graphics file and associating each of the collected lines and curves into one or more groups according to one or more grouping rules. Each of the one or more groups is then cleaned to remove dangling lines and arrows and to close any unwanted gaps between lines, and the group corresponding to the blank is identified and converted into an isolated CAD design mask comprising a first set of one or more cut lines forming a boundary. The isolated CAD design mask may be further processed by the processor, which executes the following substeps: identifying all straight lines within the mask boundary, removing arrows, identifying lines connected on one side to the first set of cut lines as one or more additional cut lines from a second set of cut lines, and identifying lines connecting two lines from either the first set or the second set of cut lines as crease lines.

[0024] In some embodiments, the step of deriving the digital representation may include matching, using a computer processor, the geometry of the CAD design derived from the non-CAD design file with the stored digital representation in the digital memory.

[0025] The digital representation derived by any of the aforementioned methods can be used to define a plurality of candidate layouts on a collection of one or more sheets; and a material consumption cost associated with each of the plurality of candidate layouts can be calculated. A mold manufacturing cost and production time can also be calculated for each candidate layout, along with an optimal cost solution from the plurality of candidate layouts. Defining the plurality of candidate layouts can include searching a digital repository for one or more pre-existing stored candidate layouts, and can also include identifying, from information stored in the digital repository, the existence of prefabricated molds corresponding to the one or more pre-existing stored candidate layouts.

[0026] Another aspect of the present invention includes a system for creating a computer-aided design (CAD) file corresponding to a 2-dimensional rendering of an unfolded blank configured to be manipulated into a 3-dimensional shape. The system includes a computer processor and a machine-readable medium accessible by the computer processor, the machine-readable medium including non-transitory instructions readable by the computer processor. The instructions include instructions defining a first digital non-CAD design file, the first digital non-CAD design file containing information related to the unfolded blank, the unfolded blank having no metadata defining cut lines or crease lines separately from surrounding content; and instructions for causing the processor to perform steps for deriving a digital representation based on the first digital non-CAD design file, the digital representation including defined data corresponding to a shape having one or more defined cut lines or crease lines of the unfolded blank.

[0027] In some embodiments, the first digital non-CAD design file can be an image file, in which case the system can further include an image capture device, such as a scanner or a digital camera, configured to create the image file and connected to a public network having a machine-readable medium accessible to the computer processor. The image file can include markings having known geometry and dimensions, wherein the computer processor is programmed with instructions to use the markings to determine dimensions of the unfolded blank for the digital representation.

[0028] In some embodiments, the first digital non-CAD design file may be a scratch graphic file. Some embodiments may include a digital memory, wherein the computer processor is configured to match the geometry of the CAD design derived from the non-CAD design file with a stored digital representation in the digital memory, search the digital memory for one or more pre-existing stored candidate layouts, or a combination thereof. In some embodiments, the computer processor may also be configured with machine-readable instructions for defining a plurality of candidate layouts on a collection of one or more sheets using the digital representations; and calculating a material consumption cost associated with each of the plurality of candidate layouts.

[0029] Yet another aspect of the present invention includes a marker configured for inclusion in a digitally captured image file of an unfolded 2-dimensional blank configured for manipulation into a 3-dimensional shape, the image file being for use in a method for creating a CAD design file, the CAD design file including one or more defined cut lines corresponding to the blank, the marker comprising a plurality of fiducials distributed in a pattern, one or more of the fiducials comprising a QR code having a predetermined geometric shape and size. The marker may comprise three QR codes positioned at positions corresponding to vertices of a right triangle. Each QR code may embody machine-readable encoded information as dimension and position information about the QR code, or as identification information of an address on a machine-accessible network, wherein the machine-readable dimension and position information about the QR code is stored.

[0030] Additional aspects of the present invention include a graphics file containing such indicia, and an image file containing machine-readable language corresponding to a captured image of the indicia and a captured image of the blank. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] FIG. 1 is a schematic diagram corresponding to a CAD design of a 2D blank configured for folding into a 3D shape.

[0032] 2 is a schematic diagram of an exemplary layout including a plurality of blanks of FIG. 1 arranged on a sheet.

[0033] Figure 3 is a scanned image of an exemplary blank.

[0034] Figure 4 is a photographic image of an exemplary blank along with markings having fiducials.

[0035] Figure 5 is an exemplary blank derived by a computer processor according to an embodiment of the present invention.

[0036] Figure 6 It's on the sheet Figure 5 An exemplary layout of the blank.

[0037] Figure 7 is a flow chart corresponding to an embodiment of one aspect of the present invention for converting an image of a physical sample into a CAD design.

[0038] Figure 8 is a schematic diagram of an exemplary system embodiment of the present invention.

[0039] Figure 9A is a flow chart corresponding to an embodiment of one aspect of the present invention for processing a camera image with markers into a suitable CAD design mask. Figure 9B is a flow chart corresponding to an embodiment of one aspect of the present invention for deriving cut lines and / or crease lines from a CAD design mask derived from an image of a physical sample.

[0040] Figure 10A Depicted is an exemplary graphic rendering from an exemplary PDF file for processing into a CAD design.

[0041] Figure 10B Depicts from Figure 10A Example separated layers of a file showing die-line information that is mixed with other information and requires further processing to be converted into a CAD design.

[0042] Figure 10C Depicts from Figure 10B An exemplary CAD design mask derived from the information depicted in FIG.

[0043] Figure 11A is a flow chart corresponding to an embodiment of one aspect of the present invention for converting a messy drawing file lacking metadata into a CAD design mask.

[0044] Figure 11B is a flow chart corresponding to an embodiment of one aspect of the present invention for deriving cut lines and / or crease lines from a CAD design mask derived from a haphazard graphic file. DETAILED DESCRIPTION

[0045] As used herein, the term "blank geometry" refers to the set of lines and paths that indicate how a flat substrate, such as a folding carton board, will be cut, creased, or folded. The term "CAD design" refers to any data structure in computer memory that represents an accurate digital representation of the blank geometry, wherein the data structure representing the blank geometry is isolated or isolable from other data structures. In practice, an accuracy within 0.2 mm has been found to be a suitable accuracy for the digital representation to produce reliable estimates, but the present invention is not limited to any particular accuracy. The term "CAD design" is not intended to be limited to any particular machine-readable format, and in particular, not to be limited to only formats traditionally associated with industrial CAD systems. For example, a PDF can be used as a CAD design, with the data structure representing the blank geometry rendered entirely in a specific spot color so that it can be easily isolated from other information in the same layer. The term "computer memory" refers to any medium in any form in which machine-readable code can be stored, including but not limited to portable media (e.g., optical, magnetic, flash memory), local memory, and memory accessible via a network, such as "in the cloud." The term "CAD design mask" refers to the outline of the blank geometry defined by the CAD design, but lacking any cut and / or crease lines within the boundaries of that outline. "CAD design files" as referenced herein include CAD designs in files stored in any format in any type of computer memory. "Non-CAD design files" are files that do not have metadata that defines cut lines or crease lines separately from the surrounding content. Exemplary non-CAD design files include, but are not limited to, image files, such as those captured by a camera as described herein, or "messy" graphic files that contain many other objects in addition to the outline of the blank, such as measurements, bleed lines, and graphic content for printing on the resulting package. "Layout" refers to the positioning of one or more CAD designs on a sheet that is subsequently cut along defined cut lines and / or crease lines to form a blank, such as a blank configured for assembly into a packaging box.

[0046] Figure 7 An exemplary workflow 700 according to the present invention is shown in FIG. Figure 8. As described above, in some cases, a CSR may receive a request for quote from a print buyer along with a physical sample of the box to be created. If not already in a flat form, the box may need to be first disassembled into a flat form by prying open glue seams and folding any flaps so that the blank lies on a single plane. In step 710, the CSR uses an image capture device 810, such as, for example, a digital camera (such as may be present on a mobile device such as a phone or tablet computer) or a scanner to capture a digital image of the physical sample. The image captured by the scanner may have a desired degree of accuracy based on the known size of the scanner itself, such as Figure 3 The image 300 depicted in FIG. An image captured with a digital camera, such as Figure 4 The image 400 depicted in , preferably also includes a mark 410 of known geometry and size captured in the same image as the blank geometry 420. Figure 4 The exemplary marking depicted in includes a set of three QR bar codes 412, 414, and 416 having known sizes and spacing from one another. Each QR code may provide information about the size and absolute position of each code.

[0047] In step 720, the CSR then transmits the captured image of the physical sample along with the order information (required quantity, inventory type, etc.) to an input 822 of the estimation system 820, such as by uploading the information using a preconfigured user interface. The image may then be stored in a memory 826. As is well known to those skilled in the art, the step of capturing the image may include storing the image locally in a digital memory, which may reside locally in a camera or phone and / or remotely "in the cloud." The step of inputting the image into the estimation system may include, for example, using a user interface of the estimation system, which may be accessed via a network (e.g., the Internet) to select a digital image from a storage device (e.g., by typing in a file name or "browsing" through a list of files or icons and selecting the desired file) and instructing upload. In other embodiments, the image capture device may be directly connected to the estimation system (via a network or hardwired) so that the captured image resides in a storage device associated with the estimation system (which may be local or cloud-based). It is worth noting that an image capture device (scanner, phone) can be considered to be connected to the computer processor of the evaluation system, as long as the capture device to which the computer processor is connected is also connected to a network (e.g., the Internet), even though other intermediary devices may be required to transmit the image to the processor. For example, an email from a scanner can be received via the Internet by a user on a first computer, who saves the image to a digital storage device, and then uploads the image to the processor, which may reside on a second computer. In other embodiments, a scanner or camera can be connected to a digital storage device on a network, and the computer processor can be configured to retrieve the image directly from the digital storage device.

[0048] The estimation system 820 comprises at least a processor programmed with instructions for performing the method steps of generating an estimate based on the image and information input received via input 822. The estimation system is programmed to perform step 730 to derive an accurate digital representation of the relevant portion of the blank shape (CAD design), as described in more detail herein. For estimation purposes, the relevant portion may only need to include the periphery (or hull) of the blank, such as Figure 5 , which has length L and width W dimensions that are accurate to within a certain degree of accuracy (e.g., plus or minus 0.2 mm or less). However, to create a production CAD design, more details, such as the location of crease lines, may be relevant. In step 740, the digital representation of the blank 500 is used to generate one or more layouts for a collection of one or more sheets, such as Figure 66. The set of one or more sheets can include a single sheet having a fixed set size, a plurality of sheets each having a different fixed set size, or a plurality of sheets having a variable set size, including at least one of a width or a length having a range of sizes with a minimum and a maximum value.

[0049] In step 750, the estimating system 820 then automatically derives the folding carton board cost for the required quantity of boxes for each sheet size, and the data is output by the estimating system via output 824 to a receiver 830. The receiver 830 may include, for example, a management information system (MIS) used by the CSR or any type of information viewer accessible to the CSR. The receiver 830 may include a display, such as a screen of a computer or mobile phone or tablet computer, for viewing the information on a web page accessible to the CSR, or embodied in an electronic communication sent to an address associated with the CSR. If appropriate predetermined information (e.g., mold making cost, production time, printing cost, shipping cost, profit) is pre-programmed into the estimating system, the remainder of the production cost and the final provision of a quote to the customer may be performed fully automatically, or in some embodiments, the CSR may manually or using other well-known processes integrate the folding carton board costs provided by the system into a more complete cost estimate. In particular, the system can be programmed to calculate the mold manufacturing cost based on linear units of mold material required and the production time for each candidate layout based on each of a plurality of layouts, or can be programmed to provide an estimate based on matching the CAD design with a database of reference CAD designs, as further described herein. The processor can also be programmed to select the best cost solution from the plurality of candidate layouts.

[0050] Deriving an accurate digital representation in step 730 may include any number of processing steps, such as compensating for rotation, perspective, and scaling to obtain an accurate digital representation of the blank (CAD design file). Figure 9AThe exemplary derivation process 900 depicted in FIG includes, in step 910, retrieving information from a QR code captured in an image, including information regarding the absolute position and size of the QR code. The QR code information may be encoded directly in the QR code itself, or the QR code may identify a predetermined location in a database where information regarding the position and size is located (e.g., a URL corresponding to a network address and a data storage location on a network accessible to the processor). It should be understood that while described in an embodiment in which the fiducials used to compensate the image include three QR codes, embodiments with fewer than three such codes may be provided, as long as information related to the marker is otherwise accessible to the system. For example, a suitable marker may include a single QR code and two non-encoded shapes, wherein the single QR code (or a data storage device linked to the code) provides data regarding the QR code and the other two fiducials sufficient to derive the desired information using any distortion in the geometry of the respective fiducials. In yet another embodiment, the relevant information may be encoded in another manner other than a QR code, such as in computer (and human) readable text, or using any other unique identifier corresponding to known information. Furthermore, while the three QR codes are positioned to correspond to the vertices of a right triangle, the geometric arrangement of the QR codes (or other fiducials) is not limited to any particular geometry, so long as the geometry is known and can be used by the processor to calculate dimensions and otherwise compensate for artifacts introduced by the image capture process. Finally, while extensively discussed herein in the context of embodiments using QR codes as fiducials, it should be understood that the present invention is not limited to the use of any particular type of fiducial, nor is it limited to the use of fiducials that embody, provide access to, or are otherwise presented in combination with coded information. Any predetermined geometry, size, and position of fiducial may be used, so long as the predetermined geometry, position, and size are known to the processor when performing the processing steps.

[0051] In step 920, the processor uses the four corner points of each QR code (or other reference) to derive a homography, which is applied to straighten the image (e.g., to remove distortion caused by camera angle, zoom, or rotation). It is worth noting that even the most basic digital cameras now include various compensations for camera lens distortion, and therefore the images captured and provided to the processor are typically already compensated accordingly. In embodiments where the images are not compensated, lens distortion compensation can optionally be performed prior to step 920 using any of a number of techniques known in the art. The image's resolution in pixels per unit size (e.g., dots per inch (DPI)) is also recorded based on the QR code size. In step 930, the processor segments the image and identifies the segments corresponding to the physical sample of the blank. This identification can be based on known characteristics of the marking, the location of the QR code, and information about the background that can be derived from known information about the marking. In step 940, the processor fills the holes and smoothes the edges of the blank identified in the previous step. In step 950, the system determines the rotation and compensates for it. For example, parallel straight edges on opposite sides of the blank, or ninety-degree angles between adjacent edges, can be identified and snapped to a horizontal / vertical grid. The foregoing steps can be performed on a raster image or a vectorized image, but in embodiments using a raster image, the system vectorizes segments of the image corresponding to a physical sample of the blank and formats this vector information into a CAD design for storage as a CAD design file.

[0052] It should be noted that although described above in conjunction with embodiments where the image is captured by a CSR, embodiments where the image is captured by a customer of interest may be implemented. For example, a graphic file (e.g., a PDF) with the markings may be sent to the customer, printed by the customer, and placed by the customer near the blank when capturing the image of the blank. Similarly, the customer may use a scanner to capture the scanned image. In some cases, the customer may be required to use an approved printer to print the markings or an approved scanner, or to communicate information about the printer / scanner to the CSR so that any irregularities introduced by the customer's printer or scanner are known and compensated for when processing the image. In other cases, the CSR may send pre-printed markings to customers (particularly repeat or otherwise known customers), or may pre-register scanners so that images sent by customers to the CSR can be reliably verified.

[0053] In the first stage of processing, after the blank geometry has been isolated from the captured image (sometimes called a "mask") and defined as a CAD design, a second stage of processing may be performed to determine cut and / or crease lines within the boundaries of the mask. Figure 9BAn exemplary processing method 970 for this second stage is outlined in

[0065] . After the mask and cut lines are calculated in the first stage, the mask is applied to the straightened image in step 980. Line detection is then performed in step 985 within the portion of the image bounded by the mask using one of many line detection algorithms known in the art. Exemplary line detection algorithms include, but are not limited to, the Hough Line Transform, as described in U.S. Patent 3,069,654 and Duda and Hart, "Use of the Hough Transformation to Detect Lines and Curves in Pictures," Comm.ACM 15, No. 1, pp. 11-15 (January 1972), both of which are incorporated herein by reference. In step 990, all lines that are connected on one side to the cutting line detected in the first stage are interpreted as additional cutting lines. Then, in step 995, all lines connecting two cutting lines (as identified in the first processing stage or in step 990) are interpreted as crease lines.

[0054] As described above in the background, sometimes a request for a quote may instead be based on a cluttered graphic file that does not have accurate embedded dimensional information (e.g., a PDF that lacks metadata, such as optional content groups or the use of colorants for fills or strokes that define cut lines or crease lines separately from the surrounding content). An exemplary embodiment for exporting a CAD design from such a PDF will now be described. Figure 10A Depicted in is an image corresponding to a graphics file (although shown in monochrome, the graphics may also include color). Figure 10B Describes the Figure 10A , showing that such separation contains many other objects besides the outline of the blank, such as measurements, diffusion lines and graphic content for printing on the resulting packaging. Figure 10C Depicts the Figure 10B Further processing may then be performed to identify cut and / or crease lines within the outline of the blank geometry, as further described herein.

[0055] Figure 11AAn exemplary algorithm 1100 for processing a drawing file is shown in FIG, and the algorithm may include first collecting all lines and curves from a correct separation (in step 1110, and then grouping the objects according to one or more rules in step 1120. Exemplary rules may include grouping objects together if and only if: (a) one of their endpoints touches another object, or (b) they use the same stroke style. Applying the rules provides multiple groups representing different parts, such as diffusion lines, legends, or the box itself. In step 1130, the groups are cleaned up using known algorithms to remove dangling lines and arrows and to close any gaps where lines are within a sufficient distance of each other to be considered touching but are not actually touching. In step 1140, a group corresponding to the blank can be identified from the multiple groups using a number of different heuristics, such as, for example, (a) the group with the most parts; (b) the group with the largest surface area; and (c) the group closest to a predefined length and width.

[0056] The identified groups are then processed in step 1150 to convert them into a mask, such as by, for example, performing the following steps: (1) rendering the groups at a high DPI; (2) applying a flood fill to the outside of the box, and (3) inverting the flood-filled image. The foregoing steps result in a mask that may potentially still contain a few lines pointing outward. To remove the additional lines in step 1160, the processor may apply a 1-pixel erosion followed by a 1-pixel dilation. The above operations may be performed on either a raster image or a vector image, but if performed on a raster image, the raster image may be vectorized, such as by any algorithm known in the art for such a purpose, including but not limited to the polygon-based tracing algorithm known as Potrace (Potrace: a polygon-based tracing algorithm, Peter Selinger, September 20, 2003; http: / / potrace.sourceforge.net / potrace.pdf), which is incorporated herein by reference.

[0057] After the mask has been isolated from the cluttered graphic file and defined as a CAD design, additional processing may be performed to determine cut lines and / or crease lines within the boundaries of the mask. Figure 11BThe exemplary processing method 1170 depicted in FIG. 1 may include first identifying all straight lines within the mask in step 1175 and then filtering out arrows in step 1180 by any means known in the art, such as by finding large lines connected to smaller lines at a specific angle (which form open arrows) or by finding small triangles connected to larger lines (closed arrows). In step 1185, all lines connected on one side to a cut line detected in the first stage are interpreted as additional cut lines. In step 1190, all lines connecting two cut lines (as identified in the first stage or step 1185) are interpreted as crease lines.

[0058] It has been found that the above-described method successfully finds die lines in PDF or other cluttered graphic files where data corresponding to the blank's geometry is included alongside other data such as dimension lines, legends, additional text boxes, or labels, making it difficult to automatically distinguish CAD data from the other data. Exemplary embodiments can successfully derive the actual die line by providing only two pieces of data: blank width and blank height, which are typically available to CSRs when an estimate is requested. Therefore, the estimation processor can be programmed with instructions for deriving CAD data from images and from PDFs.

[0059] While the methods and systems described herein have particular benefits for estimating and quoting (i.e., determining production costs in order to provide quotes to prospective customers), they are not limited to such uses. CAD designs exported from images or PDFs as described herein can be used for any purpose, such as, for example, for creating CAD designs and layouts for production use after a customer places an order. Thus, the systems and methods for automatically generating CAD designs discussed herein can also be used by CAD designers as a starting point for creating further CAD designs and layouts for use in production. Thus, the methods described herein can significantly accelerate the design process.

[0060] When used for estimation, additional processing steps and system modules can help streamline the estimation process and improve its accuracy. For example, once a CAD design is generated from an image, the exemplary method can include searching a database of reference CAD designs for CAD designs that are identical or very similar to the CAD design generated from the image. The database can reside in digital storage, such as a computer memory library, a computer library on disk, or a computer library in a local database or cloud database.

[0061] like Figure 8As depicted in the figure, computer memory 826 schematically represents any type of digital memory accessible by any means to a computer processor, which memory may store any machine-readable instructions for performing the methods of the present invention, including but not limited to instructions executable by a computer for performing method steps, non-CAD files and CAD files as described herein, databases as referenced herein in any context, including databases of CAD designs, databases of information corresponding to QR codes, and other databases referenced herein later. Although shown as a single unit, the digital storage device may include many units of digital storage media, including local or remote, portable or non-portable RAM or ROM.

[0062] When a match or near match is found after the search, the reference CAD design can be used as a substitute for the derived CAD design for estimation or production purposes. In some embodiments, the method can include searching a database of existing parametric template CAD designs (template CAD designs with dimensional variables such as length, width, depth, etc.). If a match or near match exists between a CAD design generated from a non-CAD file and a template CAD design for a particular set of dimensional values, then the existing template CAD design instantiated using that set of dimensional values ​​can be considered a parametrically equivalent design and can be used as a substitute for the derived CAD design for estimation or production.

[0063] Additional embodiments may also include searching a database of existing layouts for a specific CAD design derived from a non-CAD file. If the database search results in a match, then no new layout is required for estimation or production. For a specific existing layout, a database of die-cutter tooling may be searched for tooling made for that specific layout. If a die-cutter tooling already exists for the desired layout, then no new die tooling is required, which may result in a lower cost estimate or may save the step of manufacturing a new die-cutter tooling during production.

[0064] While the foregoing description of the need is from the perspective of folding carton production, this is merely one example of a desired embodiment of the present invention as described herein. The present invention is not limited to any particular use and may be applied without restriction to any process that would benefit from automatically deriving dimensionally accurate CAD information from an image or PDF, including in other parts of the packaging industry, including but not limited to the manufacture of labels, flexible parts, POS displays, and the like.

[0065] Although various methods have been described herein for deriving a CAD design from a non-CAD design file, including a captured image or a sparse graphic file, or by matching information in a non-CAD design file with information in a searchable database, it should be understood that an ideal system can be configured to perform more than one, if not all, of the aforementioned methods. Thus, a user of such a system will have multiple ways to potentially minimize the time and effort required to convert non-CAD design information into a CAD design, such as for the purposes of estimation as described herein. It should also be understood that the methods described herein may not provide a 100% accurate CAD design in 100% of cases, and in some cases, additional manipulation of the CAD design by the CAD operator may be required to achieve an ideal CAD design for estimation purposes.

[0066] While the invention has been illustrated and described herein with reference to specific embodiments, it is not intended to be limited to the details shown. Rather, various modifications may be made to the details within the scope and range of equivalents of the claims and without departing from the invention.

Claims

1. A computer-implemented method for creating a computer-aided design (CAD) corresponding to a 2-dimensional rendering of an unfolded blank, the unfolded blank being configured for manipulation into a 3-dimensional shape, the method comprising the steps of: (a) obtaining in a computer processor a first digital non-CAD design file comprising an image file containing information related to the geometry of the unfolded blank, the non-CAD design file being devoid of metadata defining cut lines or crease lines separately from surrounding content; (b) deriving, using a computer processor, a digital representation of the unfolded blank geometry based on the first digital non-CAD design file, the digital representation including defined data corresponding to the shape having one or more defined cut lines and / or crease lines, wherein deriving the digital representation comprises the processor performing the following steps: generating a CAD design mask, the mask including a first set of one or more cut lines forming a boundary; and A line detection algorithm is performed within the portion of the image bounded by the mask, the line detection algorithm being configured to identify one or more of the following: a second set of one or more cutting lines that are one or more lines connected on one side to the first set of cutting lines; and One or more crease lines being one or more lines connecting two lines selected from the first set of cutting lines and the second set of cutting lines.

2. The method according to claim 1, wherein The image file includes markings having known geometry and dimensions, wherein the step of deriving the digital representation includes using the markings to determine dimensions of the unfolded blank.

3. The method according to claim 2, wherein: The step of deriving the digital representation also includes using the marker to identify and compensate for artifacts in the image.

4. The method according to claim 3, wherein: The artifacts in the image are selected from the group consisting of: rotation relative to a preferred rotation, perspective relative to a plan view, and scaling.

5. The method according to any one of claims 2 to 4, wherein The marking includes a plurality of fiducials distributed in a pattern.

6. The method according to any one of claims 2 to 4, wherein: The marking includes one or more QR codes.

7. The method according to any one of claims 2 to 4, wherein The label includes three QR codes positioned at positions corresponding to the vertices of a right triangle.

8. The method according to any one of claims 2 to 4, wherein The marking includes one or more of a plurality of fiducials distributed in a pattern, one or more QR codes, and three QR codes positioned at positions corresponding to vertices of a right triangle, and wherein the marking embodies machine-readable encoded information as size and position information about the plurality of fiducials.

9. The method according to any one of claims 2 to 4, wherein: The tag includes one or more of a plurality of fiducials distributed in a pattern, one or more QR codes, and three QR codes positioned at positions corresponding to vertices of a right triangle, and machine-readable encoded information as size and position information about the plurality of fiducials, and wherein the tag encoded information is machine-readable as identification information of an address on a machine-accessible network, wherein the machine-readable size and position information about the plurality of fiducials is stored.

10. The method of any one of claims 2 to 4, further comprising compensating for artifacts in the image to produce a straightened image, and applying the mask to the straightened image.

11. The method according to any one of claims 1 to 4, wherein The information includes one or more lines in a messy drawing file.

12. The method according to claim 11, wherein Messy graphics files include PDF files that are missing metadata that defines cut lines or crease lines.

13. The method according to claim 12, wherein: The PDF file lacks an optional content group that defines cut lines or crease lines and lacks the use of colorants that allow identification of the fill or stroke of a PDF path corresponding to a cut line or crease line.

14. A computer-implemented method for creating a computer-aided design (CAD) corresponding to a 2-dimensional rendering of an unfolded blank, the unfolded blank configured for manipulation into a 3-dimensional shape, the method comprising the steps of: (a) obtaining, in a computer processor, a first digital non-CAD design file containing information relating to an unfolded blank geometry, the non-CAD design file comprising a graphics file devoid of metadata defining cut lines or crease lines, the information comprising one or more lines in a scrambled graphics file; (b) deriving, using a computer processor, a digital representation of the unfolded blank geometry based on the first digital non-CAD design file, the digital representation including defined data corresponding to the shape having one or more defined cut lines and / or crease lines, wherein deriving the digital representation comprises the processor performing the following sub-steps: collecting lines and curves from a messy graphic file and associating each of the collected lines and curves into one or more groups according to one or more grouping rules; clearing each of the one or more groups to remove dangling lines and arrowheads and any unwanted gaps between closed lines; identifying which of the one or more groups corresponds to the blank; as well as The identified group is converted into an isolated CAD design mask comprising a first set of one or more cut lines forming a boundary.

15. The method of claim 14, further comprising the processor further processing the isolated CAD design mask by performing the following sub-steps: Identify all lines within the mask boundary; Remove the arrow; identifying lines connected on one side to the first set of cutting lines as one or more additional cutting lines in the second set of cutting lines; and A line connecting two lines from the first set or the second set of cutting lines is identified as a crease line.

16. The method according to any one of claims 1-4 and 14-15, further comprising a processor performing the following steps: (c) searching the digital representation derived from the non-CAD design file in step (b) for a stored reference digital representation that matches the derived digital representation in the digital memory, and using the stored reference digital representation as a replacement digital representation.

17. The method according to any one of claims 1-4 and 14-15, further comprising a processor performing the following steps: (c) searching the digital memory for the digital representation derived from the non-CAD design file in step (b) for a stored template CAD design that matches the derived digital representation when instantiated with a particular set of parameter values, and using the template CAD design instantiated with the set of parameter values ​​as an alternative digital representation.

18. A computer-implemented method for creating a computer-aided design (CAD) corresponding to a 2-dimensional rendering of an unfolded blank configured for manipulation into a 3-dimensional shape, the method comprising the steps of: (a) obtaining in a computer processor a first digital non-CAD design file containing information relating to the geometry of an unfolded blank, the non-CAD design file being devoid of metadata defining cut lines or crease lines separately from surrounding content; (b) deriving, using a computer processor, a digital representation of the unfolded blank geometry based on the first digital non-CAD design file, the digital representation including defined data corresponding to the shape having one or more defined cut lines and / or crease lines; (c) defining, with a computer processor, a plurality of candidate layouts on a collection of one or more sheets using the digital representation or an alternative digital representation, including searching a digital repository for one or more pre-existing stored candidate layouts, the digital repository containing the same blank geometry as the derived digital representation; as well as (d) Calculating a material consumption cost associated with each of the plurality of candidate layouts.

19. The method of claim 18, further comprising calculating mold manufacturing cost and production time for each candidate layout; and selecting an optimal cost solution from the plurality of candidate layouts.

20. The method according to claim 18 or 19, wherein The collection of one or more sheets includes a single sheet having a fixed collection size, multiple sheets each having a different fixed collection size, or multiple sheets having a variable collection size, including at least one of a width or length having a range of sizes with a minimum and a maximum value.

21. The method of claim 18 or 19, further comprising identifying, from information stored in the digital repository, the existence of pre-made stencils corresponding to one or more of the pre-existing stored candidate layouts.

22. The method according to any one of claims 1-4, 14-15 and 18-19, wherein The unfolded blank is configured to be manipulated into the 3-dimensional shape of the box along the cut lines and crease lines.

23. A system for creating a computer-aided design (CAD) file corresponding to a 2-dimensional rendering of an unfolded blank configured for manipulation into a 3-dimensional shape, the system comprising: Computer processors; A non-transitory machine-readable medium accessible by a computer processor and comprising instructions readable by the computer processor for: a) defining a first digital non-CAD design, the first digital non-CAD design comprising an image file containing information related to an unfolded blank without metadata defining cut lines or crease lines separately from surrounding content; as well as b) causing the processor to perform steps for deriving a digital representation of data including a definition of a shape corresponding to the one or more defined cut lines or crease lines of the unfolded blank based on the first digital non-CAD design file, wherein deriving the digital representation includes generating a CAD design mask, the mask including a first set of one or more cut lines forming a boundary; and executing a line detection algorithm within the portion of the image defined by the mask, the line detection algorithm configured to identify one or more of the following: the second set of one or more cut lines that are connected on one side to the one or more lines of the first set of cut lines, and the one or more crease lines that are one or more lines connecting two lines selected from the first set of cut lines and the second set of cut lines.

24. The system of claim 23, further comprising an image capture device configured to create the image file and connected to a public network having a machine-readable medium accessible to the computer processor.

25. The system of claim 24, wherein: Image capture devices include scanners.

26. The system according to claim 24 or 25, wherein: Image capture devices include digital cameras.

27. The system according to claim 24 or 25, wherein: The image file includes markings having known geometry and dimensions, wherein the computer processor is programmed with instructions for using the markings to determine dimensions of the unfolded blank for digital representation.

28. The system of claim 23, further comprising a digital memory, wherein The computer processor is configured to match information from the non-CAD design file with the stored digital representation in the digital memory, search the digital memory for one or more pre-existing stored candidate layouts, or a combination thereof.

29. The system of claim 28, wherein: Digital storage includes a library in a computer repository, a library on disk, a local database, or a cloud database.

30. The system of claim 23, wherein: The digital representation includes a computer-aided design (CAD) file.

31. The system of claim 27, wherein: The marking includes a plurality of fiducials distributed in a pattern, one or more of the fiducials including a QR code having a predetermined geometric shape and size.

32. A system for creating a computer-aided design (CAD) file corresponding to a 2-dimensional rendering of an unfolded blank configured for manipulation into a 3-dimensional shape, the system comprising: Computer processors; digital memory; A non-transitory machine-readable medium accessible by a computer processor and comprising instructions readable by the computer processor for: a) defining a first digital non-CAD design, the first digital non-CAD design comprising a graphics file containing information related to an unfolded blank without metadata defining cut lines or crease lines separately from surrounding content; as well as b) causing the processor to perform steps for deriving a digital representation of data including a definition of a shape corresponding to one or more defined cut lines or crease lines of the unfolded blank based on the first digital non-CAD design file; (c) using the digital representation to define a plurality of candidate layouts on a collection of one or more sheets, including searching a library stored in a digital memory for one or more pre-existing stored candidate layouts containing the same blank geometry as the derived digital representation; and (d) Calculating a material consumption cost associated with each of the plurality of candidate layouts.

33. A system for creating a computer-aided design (CAD) file corresponding to a 2-dimensional rendering of an unfolded blank configured for manipulation into a 3-dimensional shape, the system comprising: Computer processors; A non-transitory machine-readable medium accessible by a computer processor and comprising instructions readable by the computer processor for: (a) defining a first digital non-CAD design, the first digital non-CAD design comprising a PDF file lacking metadata defining cut lines or crease lines, the PDF file containing information related to an unfolded blank including one or more lines in a scrambled graphics file; as well as (b) causing the processor to perform steps for deriving a digital representation of data including a definition of a shape corresponding to one or more defined cut lines or crease lines of the unfolded blank based on a first digital non-CAD design file, the non-CAD design file including information, wherein deriving the digital representation comprises the processor performing the following sub-steps: collecting lines and curves from a messy graphic file and associating each of the collected lines and curves into one or more groups according to one or more grouping rules; clearing each of the one or more groups to remove dangling lines and arrowheads and any unwanted gaps between closed lines; identifying which of the one or more groups corresponds to the blank; as well as The identified group is converted into an isolated CAD design mask comprising a first set of one or more cut lines forming a boundary.

34. The system of claim 33, further comprising instructions for causing a processor to: The digital representation derived from the non-CAD design file in step (b) is searched in the digital memory for a stored reference digital representation that matches the derived digital representation, and the stored reference digital representation is used as a replacement digital representation.

35. The system of claim 33, further comprising instructions for causing a processor to: Searching the digital representation derived from the non-CAD design file in step (b) in a digital memory for a stored template CAD design that matches the derived digital representation when instantiated with a particular set of parameter values, and using the template CAD design instantiated with the set of parameter values ​​as an alternative digital representation.

Images