Increasing dissimilarity of characters disposed on optically active articles

Abstract

The invention relates to increasing the dissimilarity of characters disposed on optically active articles. The inventors of the present application have developed novel retroreflective materials and articles. One example of the present application is a retroreflective article comprising a substrate and machine-readable information disposed on the substrate, where the machine-readable information corresponds to predetermined human-readable information.

Description

[0001] This application is a divisional application of Chinese patent application filed on August 19, 2016, with application number 201680048623.3 and title "Increasing the Dissimilarity of Characters Set on Optically Active Articles". Technical Field

[0002] This application relates in its entirety to novel retroreflective articles and systems in which such articles can be used. Background Technology

[0003] Automatic vehicle recognition (AVR) or automatic license plate recognition (ALPR) refers to the detection and identification of vehicles through electronic systems. Exemplary uses of AVR or ALPR include, for example, automated toll collection (e.g., electronic toll collection systems), traffic enforcement (e.g., red-light operation systems, speed enforcement systems), searching for vehicles associated with crime, access control systems, and facility access control. Currently used AVR systems may include systems that use RFID technology to read RFID tags attached to vehicles. ALPR systems may use cameras to capture license plate images.

[0004] Some AVR systems use RFID, but not all vehicles can include RFID tags. Furthermore, some tag readers may have difficulty accurately locating the exact position of unpowered RFID tags. Thus, these tag readers may only detect the presence or absence of tags within their sensitivity range, rather than the information included in the RFID tag. Some RFID tag readers may only operate at short distances, perform poorly in the presence of metal, and / or be obstructed by interference when many tagged objects are present.

[0005] The ALPR system uses an image capture device to read vehicle information, such as license plate numbers or other visual content of the license plate. In some instances, this information is attached to the license plate, printed on it, or adjacent to it. The ALPR system can be used in many environments because vehicles are required to have license plates with visually identifiable information in almost every part of the world. However, image capture and recognition of vehicle license plate information can be complex. For example, the accuracy of the ALPR system's reading can depend on the quality of the captured image as evaluated by the reader. Attached Figure Description

[0006] Figure 1 A block diagram illustrating an example system 1000 for decoding information included in an encoded region of an optically active article according to the technology of this disclosure.

[0007] Figure 2 A block diagram illustrating an example computing device according to one or more aspects of this disclosure.

[0008] Figure 3Exemplary characters with similar shapes are shown.

[0009] Figure 4 An exemplary embodiment of machine-readable information according to this application is shown.

[0010] Figure 5A An exemplary embodiment of the retroreflected article according to this application, as observed under diffuse visible light, is shown.

[0011] Figure 5B The image shown is an observation under infrared radiation. Figure 5A The retroreflected article shown.

[0012] Figure 6 An exemplary embodiment of the retroreflected article according to this application, as observed under diffuse visible light, is shown.

[0013] Figure 7 The image shown is an observation under infrared radiation. Figure 6 The retroreflected article shown.

[0014] Figure 8 A flowchart illustrating an exemplary method according to this application.

[0015] Figure 9 A flowchart illustrating an exemplary method according to this application.

[0016] Figure 10 A flowchart illustrating an exemplary method according to this application.

[0017] Figure 11A An exemplary embodiment of the retroreflected article according to this application, as observed under diffuse visible light, is shown.

[0018] Figure 11B The image shown is an observation under infrared radiation. Figure 11A The retroreflected article shown.

[0019] Figure 12 A flowchart illustrating example operation of a computing device configured to perform the techniques of this disclosure.

[0020] Figure 13 A flowchart illustrating example operation of a computing device configured to perform the techniques of this disclosure. Detailed Implementation

[0021] This disclosure relates to increasing the dissimilarity of characters disposed on optically active articles. For example, the technique may include placing one or more coded regions at or within the location of a character to increase the dissimilarity of the character's appearance relative to other characters in a character set. The technique definitively evaluates one or more locations of the character and selects locations that increase the dissimilarity of the character relative to other characters to meet a dissimilarity threshold, rather than arbitrarily placing coded regions at certain locations of the character. In this way, the technique of this disclosure can improve the dissimilarity of characters beyond simply placing markers at or within the character.

[0022] Figure 1 A block diagram illustrating an example system 1000 for including symbols on an optically active article according to the technology of this disclosure. Figure 1 As shown, system 1000 includes an image capture device 1002. Image capture device 1002 may include one or more image capture sensors 1006 and one or more light sources 1004. System 1010 may also include one or more optically active articles, such as license plates 1008, as described in this disclosure. License plates 1008 may be attached to or otherwise associated with vehicle 1010. In some examples, image capture device 1002 is communicatively coupled to computing device 1002 via network 1002 using one or more communication links. In other examples, as described in this disclosure, image capture device 1002 may be communicatively coupled to computing device 1002 via one or more forms of direct communication without network 1002 (such as via wired or wireless connections that do not require a network).

[0023] Out of Figure 1 For illustrative purposes, optically active article 1008 is shown as a license plate attached to vehicle 1010. Vehicle 1010 may be a car, motorcycle, aircraft, seagoing vessel, military equipment, bicycle, train, or any other transport vehicle. In other examples, optically active article 1008 may be attached to, included in, or embedded in an object, or otherwise integrated with or associated with an object, including but not limited to: documents, clothing, wearable devices, buildings, fixed equipment, or any other object. In some examples, optically active article 1008 may be a separate object not attached to vehicle 1010 but printed on vehicle 1010 or other suitable object.

[0024] Optically active article 1008 may include reflective, non-reflective, and / or retroreflective sheets applied to a base surface. In some examples, the optically active article may be a retroreflective article. Information such as, but not limited to, characters, images, and / or any other graphic content may be printed, formed, or otherwise embodied on the retroreflective sheet. The reflective, non-reflective, and / or retroreflective sheet may be applied to the base surface using one or more techniques or materials, including but not limited to mechanical bonding, thermal bonding, chemical bonding, or any other suitable technique for attaching the retroreflective sheet to the base surface. The base surface may include any surface to which the reflective, non-reflective, and / or retroreflective sheet may be attached (such as, as described above, an aluminum plate). Information may be printed, formed, or otherwise embodied on the sheet using any one or more films coated with ink, dye, thermal transfer tape, colorant, pigment, and / or adhesive. In some examples, the information is formed by a multilayer optical film, a material comprising optically active pigments or dyes, or a material comprising optically active pigments or dyes.

[0025] exist Figure 1 In the example, optically active article 1008 (e.g., a license plate) includes printed information 1026A to 1026F (“Information 1026”). Figure 1 In this context, each instance of information 1026 is a symbol from a symbol set. The symbol set can be an alphabet, a digit set, and / or any other glyph set. Figure 1 In this context, the symbol set includes at least the letters of the English alphabet and Arabic numerals.

[0026] exist Figure 1 In the examples, each symbol includes one or more coded regions 1028A to 1028C. The coded region may be a location, area, or region of an optically active article 1008 that can be printed selectively according to the techniques of this disclosure using (1) visually opaque infrared-transparent cyan, magenta, and yellow (CMY) inks or (2) visually opaque infrared-opaque inks (e.g., inks containing carbon black). In some examples, the coded region is embedded within instances of printed information, such embedded data units 1028A being included within the boundaries of printed information 1026A (e.g., the Arabic numeral "2"). The boundary or perimeter of a symbol may be an interface between a first set of pixel values ​​representing the symbol and a second set of pixel values ​​representing the space (e.g., blank space) surrounding the symbol or within the symbol's representation. In some examples, the boundary or perimeter of a symbol may have more than one interface, such as an "A" including an inner interface (e.g., a central blank space within an "A") and an outer interface (e.g., blank space surrounding an "A"). In some examples, the coded region may be "active" or "inactive". Figure 1In the examples, coded regions printed with visually opaque infrared-transparent ink that reflect light above a threshold intensity (e.g., 1028A) are active, while coded regions printed with visually opaque infrared-opaque ink that do not reflect light above a threshold intensity (e.g., 1028B) are inactive. In alternative examples, coded regions printed with visually opaque infrared-transparent ink that reflect light above a threshold intensity are inactive, while coded regions printed with visually opaque infrared-opaque ink that do not reflect light above a threshold intensity are active. For the purposes of this disclosure, active coded regions are generally described as regions printed with visually opaque infrared-transparent ink.

[0027] like Figure 1 As shown, coding areas 1028A and 1028C are printed using a combination of visually opaque infrared-transparent CMY inks (e.g., "processed black"), and coding area 1028B is printed using a visually opaque infrared-opaque ink. For illustrative purposes, coding area 1028B printed with visually opaque infrared-transparent CMY inks is shown as... Figure 1 The cross-hatching in the image, although visible to the human eye, allows the coded region 1028B to appear as other non-coded regions (e.g., matte black) of information 1026A. Figure 1 In addition to the location of the embedded data unit 1028, the printing information 1026A can be printed using a combination of visually opaque infrared opaque black ink.

[0028] When the printed information 1026 is exposed to infrared light 1027 from the light source 1004, the infrared light is reflected back to the image capturing device 1002 from the positions corresponding to the active coding areas 1028A and 1028C. Because the inactive coding area 1028B is printed with visually opaque infrared-opaque ink, the infrared light 1027 is absorbed at various points within the boundaries of the character "2" except for the active coding areas 1028A and 1028C. The infrared light will be reflected from the optically active article 1008 at the positions of the active coding areas 1028A and 1028B, as well as other active coding areas of the optically active article 1008 printed with visually opaque infrared-transparent ink (and not visually opaque infrared-opaque ink). Thus, the infrared image captured by the image capturing device 1002 will be as follows: Figure 1 As shown, there are blanks, gaps, or voids in the areas printed with visually opaque infrared transparent ink, while other areas printed with visually opaque infrared opaque ink will appear black or otherwise be visually distinguishable from the visually opaque infrared transparent ink.

[0029] In some examples, when printed with a visually opaque infrared-transparent ink, active coding regions 1028A and 1028C appear opaque or black to the image capture device 1002 in a first spectral range and transparent or white to the image capture device 1002 in a second spectral range. A portion of information 1026A (including inactive coding region 1028B) printed with a visually opaque infrared-opaque ink appears opaque or black to the image capture device 1002 in the second spectral range and also appears opaque or black to the image capture device 1002 in the first spectral range. In some examples, the first spectral range is from about 350 nm to about 750 nm (i.e., the visible light spectrum), and the second spectral range is from about 700 nm to about 1100 nm (i.e., the near-infrared spectrum). In some examples, the first spectral range is from about 700 nm to about 850 nm, and the second spectral range is between about 860 nm and about 1100 nm.

[0030] In some examples, active coding regions 1028A and 1028C appear opaque or black to the capture device 1002 under a first illumination condition and transparent or white to the capture device 1002 under a second illumination condition, while inactive coding region 1028B appears opaque or black to the capture device 1002 under both the first and second illumination conditions. In some examples, the first illumination condition is ambient visible (i.e., diffuse visible light), and the second illumination condition is visible retroreflection (i.e., coaxial visible light). In some examples, the positions of one or more light sources differ under the first and second illumination conditions.

[0031] In some examples, suitable printing technologies include screen printing, flexographic printing, thermal mass transfer printing, and digital printing such as laser printing and inkjet printing. One advantage of using digital printing is that information can be easily and quickly customized / changed to meet customer needs without producing new screens or flexographic sleeves.

[0032] In some examples, the printing of the coded and uncoded portions of a symbol is performed in alignment such that they completely overlap. In some examples, the active coded portions are printed first on the retroreflective substrate, followed by the printing of the uncoded portions of the symbol, or vice versa. In some examples, materials described in co-pending U.S. Patent Application 61 / 969889 (Attorney's File No. 75057US002) are used to print human-readable and / or machine-readable information, the entire disclosure of which is incorporated herein by reference, but other suitable materials may also be used.

[0033] In some examples, the coded region includes at least one of infrared reflective, infrared scattering, and infrared absorbing materials. The use of these materials produces contrast in the infrared spectrum and thus appears “dark” when viewed under such conditions. Exemplary materials that may be used include those listed in U.S. Patent 8,865,293 (Smithson et al.), the disclosure of which is incorporated herein by reference in its entirety.

[0034] like Figure 1 As shown, system 1000 may include image capture device 1002. Image capture device 1002 can convert light or electromagnetic radiation sensed by image capture sensor 1006 into information, such as a digital image or bitmap comprising a set of pixels. Each pixel may have chromaticity and / or luminance components representing the intensity and / or color of the light or electromagnetic radiation. Image capture device 1002 may include one or more image capture sensors 1006 and one or more light sources 1004. In some examples, image capture device 1002 may include, for example, Figure 1 The image capture sensor 1006 and light source 1004 are shown in a single integrated device. In other examples, the image capture sensor 1006 or light source 1004 may be separate from or otherwise not integrated into the image capture device 1002. Examples of the image capture sensor 1006 may include a semiconductor charge-coupled device (CCD) or active pixel sensor in complementary metal-oxide-semiconductor (CMOS) or N-type metal-oxide-semiconductor (NMOS, active MOS) technology. Digital sensors include flat panel detectors. In examples, the image capture device 1002 includes at least two different sensors for detecting light in two different wavelength spectra. In some examples, the first image capture sensor and the second image capture sensor detect the first wavelength and the second wavelength substantially simultaneously. "Substantially simultaneously" can mean detecting the first wavelength and the second wavelength within 10 milliseconds of each other, within 50 milliseconds of each other, or within 1000 milliseconds of each other, to name just a few examples.

[0035] In some examples, one or more light sources 1004 include a first radiation source and a second radiation source. In some examples, the first radiation source emits radiation in the visible spectrum, and the second radiation source emits radiation in the near-infrared spectrum. In other examples, both the first and second radiation sources emit radiation in the near-infrared spectrum. Figure 1 As shown, one or more light sources 1004 may emit radiation in the near-infrared spectrum (e.g., infrared light 1027).

[0036] In some examples, the image capture device 1002 includes a first lens and a second lens. In some examples, the image capture device 1002 captures frames at 50 frames per second (fps). Other exemplary frame capture rates include 60 fps, 30 fps, and 25 fps. It will be apparent to those skilled in the art that the frame capture rate depends on the application and different rates, such as, for example, 1000 fps or 200 fps, can be used. Factors affecting the desired frame rate include, for example, the application (e.g., parking fee collection), the vertical field of view (e.g., a lower frame rate can be used for a larger field of view, but depth of focus may be problematic), and vehicle speed (faster traffic requires a higher frame rate).

[0037] In some examples, the image capture device 1002 includes at least two channels. These channels may be optical channels. The two optical channels may pass through a lens to a single sensor. In some examples, each channel of the image capture device 1002 includes at least one sensor, a lens, and a bandpass filter. The bandpass filter allows multiple near-infrared wavelengths to be transmitted for reception by a single sensor. The at least two channels may be distinguished by one of the following: (a) bandwidth (e.g., narrowband or wideband, where narrowband illumination may be any wavelength from visible light to near-infrared); (b) different wavelengths (e.g., narrowband processing at different wavelengths may be used to enhance features of interest, such as license plates and their lettering (license plate identifiers), while suppressing other features (e.g., other objects, sunlight, headlights); (c) wavelength region (e.g., wideband light in the visible spectrum and used with a color or monochrome sensor); (d) sensor type or characteristics; (e) exposure time; and (f) optical components (e.g., a lens).

[0038] exist Figure 1 In some examples, the image capturing device 1002 may be stationary or otherwise mounted in a fixed position, and the position of the optically active article 1008 may not be stationary. When the vehicle 1010 approaches or passes by the image capturing device 1002, the image capturing device 1002 may capture one or more images of the optically active article 1008. However, in other examples, the image capturing device 1002 may not be stationary. For example, the image capturing device 1002 may be in another vehicle or a moving object. In some examples, the image capturing device 1002 may be held by a human operator or a robotic arm device that changes the position of the image capturing device 1002 relative to the optically active article 1008.

[0039] exist Figure 1In one example, image capture device 1002 may be communicatively coupled to computing device 112 via one or more communication links 1030A and 1030B. Image capture device 1002 may transmit images of optically active article 1008 to computing device 1016. Communication links 1030A and 1030B may represent wired or wireless connections. For example, communication links 1030A and 1030B may be wireless Ethernet connections using the WiFi protocol and / or wired Ethernet connections using Category 5 or Category 6 cables. Any suitable communication link is possible. In some examples, image capture device 1002 may be communicatively coupled to computing device 1016 via network 1014. Network 1014 may represent any number or number of network connectivity devices, including but not limited to routers, switches, hubs, and interconnect communication links that provide packet and / or frame-based data forwarding. For example, network 1014 may represent the Internet, a service provider network, a customer network, or any other suitable network. In other examples, the image capture device 1002 is communicatively coupled to the computing device 1016 via a direct connection such as a Universal Serial Bus (USB) link.

[0040] Computing devices 1016 and 1032 represent any suitable computing system that can be located remotely from image capture device 1002, such as one or more desktop computers, laptop computers, mainframes, servers, cloud computing systems, etc., capable of sending and receiving information with image capture device 1002. In some examples, computing devices 1016 and 1032 implement the techniques of this disclosure.

[0041] exist Figure 1 In the example, the optically active article 1008 may initially include coding regions 1028A to 1028C and Figure 1 Examples of information 1026A to 1026F for other coded regions shown are generated. During manufacturing or printing, a retroreflective sheet is applied to an aluminum plate, and each instance of information 1026A to 1027 is printed based on a printing specification that can be provided by a human operator or generated by the machine. The printing specification may specify one or more locations to be printed with visually opaque infrared-opaque ink or visually opaque infrared-transparent ink; that is, the printing specification may specify one or more active and inactive coded regions. For example, the printing specification may specify one or more symbols from a symbol set, where a particular symbol includes one or more coded regions. The printing specification may indicate which coded regions in the embedded symbol are printed with visually opaque infrared-opaque ink (e.g., inactive), and which other coded regions in the embedded symbol are printed with visually opaque infrared-transparent ink (e.g., active).

[0042] As described above, printing specifications can be generated by a human operator or by a machine. For example, a printing press for printing visually opaque infrared-opaque ink and visually opaque infrared-transparent ink may include a user interface and / or communicatively coupled to a computing device 1032, which provides the user interface through which a human operator can input, specify, or otherwise provide printing specifications for optically active articles. For example, as... Figure 1 As shown, a human operator can specify one or more symbols to be printed as instances of information 1026A to 1026F. Each symbol may include one or more encoded regions. The encoded regions may be located at fixed, predefined positions.

[0043] In some instances, computing device 1016 may perform optical character recognition (OCR) on instances of information 1026A to 1026F. For example, computing device 1016 may perform OCR on one or more symbols represented by instances of information 1026A to 1026F. However, some symbols such as “2” and “Z” or “B” and “8” may be similar in spatial appearance. In some examples, spatial appearance may be a structurally visible representation of the symbol. As the similarity between two symbols increases, OCR processing may be more likely to incorrectly classify an image of one symbol (e.g., “2”) as another symbol (e.g., “Z”). The techniques disclosed herein can reduce incorrect classification of symbols in OCR processing by embedding an active coding region within the symbol at a specific location to increase the dissimilarity to a level greater than a dissimilarity threshold.

[0044] To embed active coding regions within symbols at specific locations to increase dissimilarity beyond a dissimilarity threshold, computing device 1032 may generate a symbol set in which one or more symbols may each include one or more active coding regions. Computing device 1032 may include a set generation component 1034 that may generate such symbol sets. For example, computing device 1032 may initially store a symbol set including the English alphabet and Arabic numerals.

[0045] Initially, the symbols in the symbol set may not include any active coding regions. The generation component 1034 may select specific symbols from the symbol set, such as the symbol "2". The generation component 1034 may compare the spatial appearance of the "2" symbol with one or more of the remaining symbols in the symbol set (e.g., the "Z" symbol). In some examples, the comparison may perform pairwise comparisons of the "2" and "Z" symbols to determine the dissimilarity between the two symbols. For example, each of the "Z" and "2" symbols may be represented as data comprising independent grayscale arrays of integers from 0 to 255. Each array position may correspond to a portion of the symbol or a portion of the blank space surrounding the symbol. Integers in the range of 0 to 255 may represent the intensity of brightness, where 0 may represent pure white and 255 represents pure black, and values ​​within this range represent different intensities of black. In some examples, the white-black spectrum may be inverted to values ​​from 255 to 0.

[0046] In some examples, the dissimilarity between the "Z" symbol and the "2" symbol can be expressed as the entry-based L2 norm between the corresponding grayscale arrays used for the "Z" and "2" symbols. Therefore, the generation component 1034 can determine the individual dissimilarity between the "2" symbol and the other symbols in the symbol set. Although the dissimilarity is described as... Figure 1 The entry-style L2 norm between the corresponding grayscale arrays can be used, but the set generation component 1034 can use any suitable comparison technique to calculate the dissimilarity between the two symbols.

[0047] Based on a comparison of the corresponding grayscale arrays for the "Z" and "2" symbols, the set generation component 1034 can determine the dissimilarity between the two symbols. The set generation component 1034 can determine if the dissimilarity meets a threshold (e.g., a first threshold) (e.g., the dissimilarity is less than or equal to the threshold). The threshold can be a hard-coded, user-defined, or machine-generated predefined value. For example, the threshold can be the minimum amount of dissimilarity between the two symbols.

[0048] If the similarity is less than or equal to a minimum dissimilarity, then the set generation component 1034 can modify the symbol to store data representing an active coding region embedded in a specific symbol at a particular location. Instead of embedding the active coding region at any possible location within the symbol, the set generation component 1034 evaluates a set of possible locations where an active coding region can be located within the symbol. For example, coding regions 1028A, 1028B, and 1028C represent three possible locations for embedding an active coding region, but many other coding regions may also be within the symbol “2” represented by the instance of information 1026A. The set generation component 1034 can select a location from this set of possible locations to embed the active coding region at a location that increases, for example, the dissimilarity between the symbols “2” and “Z”. In some examples, the set generation component 1034 selects a location from this set of possible locations to embed the active coding region at a location that increases the dissimilarity. The set generation component 1034 can embed multiple active coding regions at different locations that increase the dissimilarity. To embed an active coding region within a symbol, the set generation component 1034 can modify integer values ​​within the symbol's grayscale array. For example, at locations within the symbol's grayscale array (e.g., points / pixels, regions, or areas), the set generation component 1034 can modify integer values, such as changing a black value of 255 to a white value of 0. In this way, the symbol's grayscale array appears as if the active coding region is included within the symbol.

[0049] In some examples, a first set of pixel values ​​representing an image of one or more active coded regions within the one or more coded regions falls within a first pixel value range. In some examples, a second set of pixel values ​​representing the remainder of at least one symbol excluding the one or more active coded regions falls within a second pixel value range different from the first pixel value range. In some examples, the image is a first image, wherein a first image of an optically active article is captured within a first spectral range in the near-infrared spectrum, a second image of the optically active article is captured within a second spectral range in the visible spectrum, and a third set of pixel values ​​representing at least one symbol in the second image falls within a second pixel value range, wherein a first proportion of the third set of pixel values ​​representing at least one symbol is greater than a second proportion of the second set of pixel values ​​representing at least one symbol. In some examples, the symbol may be embodied in a vectorized representation. The computing device 116 may convert the symbol representation between a bitmap and a vectorized representation. For example, when the symbol includes one or more active coded regions, when the symbol is represented in bitmap form, the computing device 116 may convert the modified bitmap back to a vectorized representation for use in printing or manufacturing processes.

[0050] In some examples, when embedding one or more active coding regions into the symbol "2", the set generation component 1034 may again compare the spatial appearance of the modified "2" symbol (having one or more active coding regions) with one or more of the remaining symbols in the symbol set (e.g., the "Z" symbol). For example, the set generation component 1034 may compare the spatial appearance of the modified "2" with one or more other symbols in the symbol set that include the "Z" symbol. The set generation component 1034 may determine whether the similarity between the modified "2" and at least one other symbol in the symbol set satisfies (e.g., greater than or equal to) a threshold (e.g., a second threshold). In some examples, the set generation component 1034 may determine that any similarity between any two symbols in the symbol set satisfies a threshold. That is, all symbols are at least dissimilar to a degree greater than or equal to the threshold. For example, the set generation component may identify a specific location from multiple locations within a specific symbol to include active coding regions that increase the updated spatial appearance of the specific symbol and the dissimilarity between all symbols in the symbol set to satisfy the threshold. The set generation component 1034 may repeatedly perform the process described above to add active coding regions at symbol locations to increase the amount of dissimilarity to a level greater than a threshold. In some examples, the set generation component 1034 updates the symbol set by replacing a previous version of a symbol with a modified version that includes one or more different or additional active coding regions at one or more different locations of the symbol. In some examples, the set generation component 1034 may identify other locations within a symbol that provide less dissimilarity between two symbols than other locations within the symbol. For example, the set generation component 1034 may identify a second location from among multiple locations within a particular symbol that, when an active coding region is embedded, increases the dissimilarity between the updated spatial appearance of the particular symbol and the spatial appearance of at least one other symbol by an amount less than the similarity between the different updated spatial appearances and the spatial appearances of at least one other symbol when an active coding region is embedded at a first location of the symbol. In such instances, the set generation component 1034 may prevent the modification of the symbol to store data representing an active coding region embedded in the particular symbol at the second location. As described above, in some examples, the techniques of this disclosure can a priori identify symbols with dissimilarity less than a threshold and modify such symbols to increase dissimilarity to greater than the threshold, rather than a posteriori determining that a particular symbol is incorrectly classified as another symbol by the OCR system and adding distinguishing markers to that particular symbol. Thus, in some examples, the techniques of this disclosure can generate a set of symbols with dissimilarity satisfying the threshold without relying on misclassification of symbols in OCR to update the spatial appearance of symbols to increase dissimilarity, rather than using OCR for any classification or misclassification of such symbols.

[0051] In some examples, set generation component 1034 may generate an updated symbol set in which the dissimilarity between at least two symbols meets a threshold. The updated symbol set may be stored as OCR data at computing device 1016, and OCR component 1018 uses this OCR data to identify different symbols. In some examples, computing device 1032 may include construction component 136 for generating print specifications based on one or more user inputs. For example, construction component 136 may receive user input specifying the string "250 RZA". Construction component 136 may select a symbol representing "250 RZA" from the updated symbol set and include that symbol in the print specification. The print specification may include data indicating the location of active coding regions within each symbol based on the symbols selected from the symbol set. For example, the print specification may indicate the application of visually opaque infrared transparent ink to print the locations corresponding to active coding regions 1028A and 1028C, and the application of visually opaque infrared opaque ink to print the remaining area of ​​the symbol "2" (e.g., an instance of information 1026A).

[0052] In some examples, computing device 1032 may be communicatively coupled to a printing press or other manufacturing apparatus capable of setting symbols on an optically active article according to printing specifications. The printing press or other manufacturing apparatus may print, for example, visually opaque infrared-transparent ink at active coding regions 1028A and 1028C, and print the remaining area of ​​the symbol "2" (e.g., an example of information 1026A) with visually opaque infrared-opaque ink. In other examples, the printing press or other manufacturing apparatus may print, for example, visually opaque infrared-opaque ink at active coding regions 1028A and 1028C, and print the remaining area of ​​the symbol "2" (e.g., an example of information 1026A) with visually opaque infrared-transparent ink. In this way, the printing press may construct an optically active article having symbols having active coding region locations to provide dissimilarity that satisfies a predefined threshold. Although assembly generation component 1034 and construction component 136 are shown as being included in the same computing device 1032, in some examples, components 1034 and 138 may be included in separate computing devices. The computing device 1032 is communicatively coupled to the network 1014 via the communication link 1030C.

[0053] exist Figure 1In the example, computing device 1016 includes an optical character recognition component 1018 (or "OCR module 1018"), a service component 1022, and a user interface (UI) component 1024. Components 1018, 1022, and 1024 may perform the operations described herein using software, hardware, firmware, or a mixture of both hardware, software, and firmware residing in and executing on computing device 1016 and / or at one or more other remote computing devices. In some examples, components 1018 and 1022 may be implemented as hardware, software, and / or a combination of hardware and software. Computing device 1016 may execute components 1018 and 1022 using one or more processors. Computing device 1016 may execute any of components 1018 and 1022 as a virtual machine executing on the underlying hardware or within a virtual machine executing on the underlying hardware. Components 1018 and 1022 may be implemented in various ways. For example, any of components 1018, 1022 may be implemented as a downloadable or pre-installed application or "app." In another example, any of components 1018, 1022 may be implemented as part of the operating system of computing device 1016.

[0054] After the optically active article 1008 has been constructed, it can be attached to the vehicle 1010. Figure 1 In the example, vehicle 1010 may be traveling on a road and approaching image capture device 1002. Image capture device 1002 may cause light source 1004 to project infrared light 1027 in the direction of vehicle 1010. substantially simultaneously with the projection of infrared light 1027 by light source 1004, image capture device 1006 may capture one or more images of optically active article 1008, such as image 1009. "Substantially simultaneously" may mean simultaneously or within 10 milliseconds, 50 milliseconds, or 1000 milliseconds. Image 1009 may be a bitmap in which visually opaque infrared-opaque ink is rendered as black pixels and visually opaque infrared-transparent ink is rendered as white pixels. The resulting image 1009 may be stored as a bitmap, and image capture device 1002 may transmit the bitmap to computing device 1016 via network 1014.

[0055] OCR component 1018 initially receives a bitmap representing image 1009, which represents symbols as instances of information 1026A to 1026F. For example... Figure 1As shown, image 1009 includes at least one symbol (e.g., "2"), which includes a set of one or more active coding regions 1028A, 1028C embedded with the symbol. In response to receiving image 1009, OCR component 1018 performs optical character recognition on image region 1029 including an instance of information 1026A. Based on the optical character recognition performed by OCR component 1018, OCR component 1018 determines that image region 1029 of image 1009 represents at least one symbol "2". OCR component 1018 can implement any one or more OCR techniques, including but not limited to matrix matching and feature matching. Matrix matching can perform a pixel-by-pixel comparison of one or more portions of image region 1029 with a set of one or more stored glyphs, one of which corresponds to the symbol "2". Feature matching decomposes various features of an instance of information 1026A (e.g., lines, loops, line directions, intersections, etc.), which are compared with glyph features of a corresponding symbol set to identify the symbol "2". In any case, the OCR component 1018 can use the symbol set generated by the set generation component 1034 to perform optical character recognition.

[0056] Because instances of information 1026A include active coding regions 1028A and 1028C, OCR component 1018 is unlikely to incorrectly classify the symbol “2” represented by an instance of information 1026A as “Z” represented by an instance of information 1026E. As described above, the arrangement of one or more active coding regions 1028A and 1028C within at least one symbol provides character dissimilarity between at least one symbol and another in a symbol set that satisfies a predefined threshold (e.g., a second threshold). In some examples, the arrangement may refer to the location of one or more active coding regions.

[0057] Service component 1022 may receive one or more values ​​from OCR component 1018, such as the string "250 RZA". Service component 1022 may provide any number of services by performing one or more operations. For example, service component 1022 may provide toll collection services, vehicle registration verification services, security services, or any other suitable services. The toll collection service may identify a vehicle and collect tolls from the vehicle owner's payment account. For example, the values ​​received by service component 1022 may include information that can be used to collect tolls. The vehicle registration verification service may verify that the vehicle registration is current based on this value by accessing the registration database. The security service may determine whether a specific license plate associated with a vehicle is counterfeit based on this value or the absence of this value. Service component 1022 may also provide any other suitable services. Figure 1In the example, service component 1022 can determine whether a license plate associated with a vehicle is counterfeit based on the license plate string. Service component 1022 can send data to UI component 1024, which indicates whether the license plate is counterfeit.

[0058] The computing device 1000 may also include a UI component 1006. The UI component 1006 of the computing device 1002 may receive user input instructions from input devices such as a touchscreen, keyboard, mouse, camera, sensor, or any other input device. The UI component 1006 may transmit the user input instructions as output to other modules and / or components, enabling those other components to perform operations. The UI component 1006 may act as an intermediary between various components and modules of the computing device 1016 to process input detected by input devices and send it to other components and modules, and to generate output from other components and modules that may appear at one or more output devices. For example, the UI component 1006 may generate one or more user interfaces for display. The user interface may correspond to a service provided by one or more of the service components 112. For example, if the license plate for vehicle 1010 has expired, the UI component 1006 may generate an alarm to be output for display in a graphical user interface. In some examples, the UI component 1006 may generate one or more alarms, reports, or other communications to be sent to one or more other computing devices. Such alerts may include, but are not limited to, emails, text messages, lists, phone calls, or any other suitable communication.

[0059] Figure 2 A block diagram illustrating an example computing device according to one or more aspects of this disclosure. Figure 2 Only those shown are shown Figure 1 The computing device 1016 shown is a specific example. Many other examples of the computing device 1016 may be used in other instances and may include a subset of the components included in the example computing device 1016, or may include components in… Figure 2 Additional components not shown in the example computing device 1016. In some examples, computing device 1016 may be a server, tablet computing device, smartphone, wrist-worn or head-mounted computing device, laptop computer, desktop computing device, or any other computing device capable of running a set, subset, or superset of functions included in application 2028.

[0060] like Figure 2As illustrated in the example, computing device 1016 may be logically divided into user space 2002, kernel space 2004, and hardware 2006. Hardware 2006 may include one or more hardware components that provide an operating environment for components executing in user space 2002 and kernel space 2004. User space 2002 and kernel space 2004 may represent different portions or partitions of memory, wherein kernel space 2004 provides higher privileges to processes and threads than user space 2002. For example, kernel space 2004 may include operating system 2020, which operates with higher privileges than components executing in user space 2002.

[0061] like Figure 2 As shown, hardware 2006 includes one or more processors 2008, input units 2010, storage devices 2012, communication units 2014, and output units 2016. The processors 2008, input units 2010, storage devices 2012, communication units 2014, and output units 2016 may be interconnected via one or more communication channels 2018. Communication channels 2018 may interconnect each of units 2008, 2010, 2012, 2014, and 2016 for inter-unit communication (physically, communicatively, and / or operatively). In some examples, communication channels 2018 may include a hardware bus, a network connection, one or more inter-process communication data structures, or any other components for transferring data between hardware and / or software.

[0062] One or more processors 2008 may implement functions within computing device 1016 and / or execute instructions within computing device 1016. For example, processor 2008 on computing device 1016 may receive and execute instructions stored in storage device 2012, which provide functionality for components included in kernel space 2004 and user space 2002. These instructions executed by processor 2008 may cause computing device 1016 to store information in storage device 2012 and / or modify information in storage device 1012 during program execution. Processor 2008 may execute instructions for components in kernel space 2004 and user space 2002 to perform one or more operations according to the techniques of this disclosure. That is, components included in user space 2002 and kernel space 2004 may be operated by processor 2008 to perform the various functions described herein.

[0063] One or more input components 242 of the computing device 1016 may receive input. Examples of input include tactile input, audio input, motion input, and optical input, to name just a few. In one example, the input component 242 of the computing device 1016 includes a mouse, keyboard, voice response system, camera, button, control panel, microphone, or any other type of device for detecting input from a human or machine. In some examples, the input component 242 may be a presence-sensitive input component, which may include a presence-sensitive screen, touch-sensitive screen, etc.

[0064] One or more output components 2016 of the computing device 1016 may generate output. Examples of output are tactile output, audio output, and video output. In some examples, the output component 2016 of the computing device 1016 includes a presence-sensitive screen, a sound card, a video graphics adapter card, a speaker, a cathode ray tube (CRT) monitor, a liquid crystal display (LCD), or any other type of device for generating output for humans or machines. The output component may include a display component, such as a cathode ray tube (CRT) monitor, a liquid crystal display (LCD), a light-emitting diode (LED), or any other type of device for generating tactile output, audio output, and / or visual output. In some examples, the output component 2016 may be integrated with the computing device 1016. In other examples, the output component 2016 may be physically external to and separate from the computing device 1016, but may be operatively coupled to the computing device 1016 via wired or wireless communication. The output component may be a built-in component of the computing device 1016, located within and physically connected to the external package of the computing device 1016 (e.g., a screen on a mobile phone). In another example, the output component may be an external component of the computing device 1016, located outside and physically separate from the package of the computing device 1016 (e.g., a monitor, projector, etc., sharing a wired and / or wireless data path with a tablet computer).

[0065] One or more communication units 2014 of computing device 1016 can communicate with external devices by transmitting and / or receiving data. For example, computing device 1016 can use communication units 2014 to transmit and / or receive radio signals over a radio network such as a cellular radio network. In some examples, communication units 2014 can transmit and / or receive satellite signals over a satellite network such as a Global Positioning System (GPS) network. Examples of communication units 2014 include network interface cards (e.g., such as Ethernet cards), optical transceivers, radio frequency transceivers, GPS receivers, or any other type of device capable of transmitting and / or receiving information. Other examples of communication units 2014 may include Bluetooth, which is present in mobile devices. ®GPS, 3G, 4G and Wi-Fi ® Radio receivers and Universal Serial Bus (USB) controllers, etc.

[0066] One or more storage devices 2012 within the computing device 1016 may store information for processing during operation of the computing device 1016. In some examples, the storage device 2012 is temporary memory, meaning that the primary purpose of the storage device 2012 is not long-term storage. The storage device 2012 on the computing device 1016 may be configured as volatile memory for short-term storage of information, and therefore the stored contents are not retained when the device is deactivated. Examples of volatile memory include random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), and other forms of volatile memory known in the art.

[0067] In some examples, storage device 2012 also includes one or more computer-readable media. Storage device 2012 may be configured to store a larger amount of information than volatile memory. Storage device 2012 may also be configured as a non-volatile memory space for long-term storage of information, retaining information after an activation / deactivation cycle. Examples of non-volatile memory include magnetic hard disks, optical disks, floppy disks, flash memory, or electrically programmable memory (EPROM) or electrically erasable programmable memory (EEPROM). Storage device 2012 may store program instructions and / or data associated with components included in user space 2002 and / or kernel space 2004.

[0068] like Figure 2 As shown, application 2028 executes in user space 2002 of computing device 1016. Application 2028 can be logically divided into presentation layer 2022, application layer 2024, and data layer 2026. Presentation layer 2022 may include user interface (UI) component 2028, which generates and renders the user interface of application 2028. Application 2028 may include, but is not limited to, UI component 1024, OCR component 1018, and one or more service components 1022. For example, application layer 2024 may include OCR component 1018 and service components 1022. Presentation layer 2022 may include UI component 1024.

[0069] Data layer 2026 may include one or more data repositories. Data repositories may store data in structured or unstructured form. Example data repositories may be any one or more of a relational database management system, an online analytical processing database, a table, or any other suitable structure for storing data. OCR data repository 2030 may include matrix matching data to perform pixel-by-pixel comparisons, such as stored glyphs. OCR data repository 2030 may include feature matching data to perform feature identification, such as glyph features of a set of corresponding symbols. Service database 2032 may include any data that provides services to service component 1022 and / or is generated by the services provided by service component 1022. For example, service data may include vehicle registration information, security information (e.g., cyclic redundancy code or cyclic redundancy check), user information, or any other information. Image data 2031 may include one or more images received from one or more image capture devices. In some examples, the images are bitmaps, Joint Photographic Experts Group images (JPEG), Portable Web Graphics images (PNG), or any other suitable graphic file format.

[0070] exist Figure 2 In some examples, one or more of the communication units 2014 may receive from the image capture device an image of an optically active article comprising a set of one or more symbols. In some examples, any one or more components of the UI component 1024 or the application layer 2024 may receive an image of the optically active article and store the image in image data 2031. At least one of the set of one or more symbols represented in the image includes a set of one or more coded regions embedded with the symbol.

[0071] In response to receiving an image, the OCR component 1018 can compare a first spatial appearance of a specific symbol with a second spatial appearance of another symbol in the symbol set, wherein the first and second spatial appearances are based on specific lighting conditions. Figure 2 In the example, the lighting conditions can be infrared lighting conditions. For example, in... Figure 1 As described above, by applying one or more OCR techniques to an image region, OCR component 1018 can determine that the image region represents the symbol "2". As an example, OCR component 1018 can compare the image region with OCR data 2030 to identify a match. Because the symbol includes active coding regions 1028A and 1028C, OCR component 1018 is unlikely to incorrectly classify the symbol "2" as the symbol "Z". OCR component 1018 can determine one or more symbols included within an optically active article. In some examples, one or more symbols represented in the optically active article are referred to as one or more values.

[0072] Service component 1022 can perform one or more operations based on one or more specific values, such as performing a security check to determine whether an optically active article (e.g., a license plate) is counterfeit. Service component 1022 can, for example, query service data 2032 to select a CRC code or CRC check data. Service component 1022 can use service data 2032 to confirm whether the optically active article is counterfeit. For example, in response to determining that the optically active article is counterfeit, service component 1022 can send data to UI component 1024, which causes UI component 1024 to generate an alarm for display. UI component 1024 can send data to the output component of output component 2016, which causes the output component to display the alarm.

[0073] In some examples, computing device 1016 can receive multiple images of an optically active article within a specific time period. For example, an image capturing device can send multiple images, each captured within a specific time period, to computing device 1016. In some examples, the time period can be 50 milliseconds, 500 milliseconds, 1 second, or 5 seconds. In some examples, the time period can be any value between 10 milliseconds and 10 seconds. Application 2028 can perform the above... Figure 2 The technique described herein is performed repeatedly for each image captured within a specific time period. The OCR component 1018 can evaluate the value of each image and provide the service component 1022 with the value of the most frequent or most likely value.

[0074] In some examples, the term "human-readable information" refers to information that can be read and understood by humans without additional machine translation. Examples of human-readable information include, but are not limited to, alphanumeric characters, drawings, geometric shapes, symbols, and Asian or Arabic language characters. In some examples, human-readable information does not include coded information that humans cannot understand without additional machine translation, such as, for example, barcodes. In some examples, under certain conditions, human-readable information may be visible or invisible to appropriate optical detectors (e.g., the human eye).

[0075] In some examples, the term "machine-readable information" refers to information encoded in a form that can be optically imaged by a machine or computer and is readable by the hardware and software of the machine or computer rather than by a human. In some examples, under certain conditions, machine-readable information may be visible or invisible to appropriate optical detectors (e.g., cameras).

[0076] Existing ALPR systems or machine vision systems may include at least one optical detector (e.g., a camera) and software associated with that detector. The optical detector captures at least one image of an object of interest under a first condition. The object of interest may be, for example, a vehicle license plate, vehicle certification sticker, sign, security document, or conspicuous sheet material. The associated software then analyzes the image and extracts useful information from it. In some cases, extracting information from an image is also referred to as reading the object or reading information set on the object.

[0077] Some techniques disclosed herein relate to machine-readable information including embedded markers placed to maximize (or increase to a level greater than a threshold) the differences between similar human-readable information (e.g., characters), thereby improving character differentiation. In one aspect, embedded markers are placed at predetermined locations to produce a font or set of machine-readable characters that maximize the differences between similar characters (or increase to a level greater than a threshold). Strategic and predetermined gap placement is performed using a method that selectively selects the positions of embedded markers that result in the highest possible character differentiation. In some examples, the embedded markers include at least one gap or discontinuous portion. In some examples, the method of the invention includes a gap placement algorithm that seeks to select gap positions that produce the maximum number (exceeding a threshold) of differences among similar characters. Thus, the embedded markers (e.g., gaps) form machine-readable information (e.g., machine-readable characters) that can be more easily distinguished by an OCR engine.

[0078] Furthermore, in the articles and methods of the present invention, it is not necessary to read both human-readable and machine-readable information for accurate detection and reading of human-readable information. For example, the currently disclosed machine-readable information corresponds to predetermined human-readable information. Thus, in some examples, only one image of the retroreflected article needs to be captured, provided that machine-readable information is allowed to become visible and detectable.

[0079] In one aspect, this application relates to a retroreflective article comprising a retroreflective substrate and machine-readable information disposed on at least a portion of the retroreflective substrate, wherein the machine-readable information corresponds to predetermined human-readable information. In some examples, the article is one of a sign, certification label, conspicuous sheet, security document, and license plate. "Machine-readable information corresponds to predetermined human-readable information" can mean that the machine-readable information is essentially the same as the human-readable portion, except that the machine-readable information may require additional machine translation to be understood by a human. That is, for an article comprising both human-readable and machine-readable information, a machine vision system that reads the machine-readable information obtains the same information as a human reading the human-readable information.

[0080] In some examples, this application relates to methods and systems for reading articles such as, for example, license plates. The machine-readable information provided herein can be used to accurately identify and / or read license plates without relying on independent reading of human-readable information or images.

[0081] In some cases, human-readable information set on the substrate includes connected characters. In some languages ​​or fonts, the characters are cursive (i.e., flowing or connected), meaning there are no visible gaps or spaces between characters. Exemplary free-flowing languages ​​include Arabic, Persian, and Urdu. Similarly, other languages ​​or fonts may have independent characters (i.e., not connected to each other), but may include connecting elements set above or below a given string or non-collinear characters. Exemplary languages ​​of this type include Siamese. These fonts and / or languages ​​are particularly difficult for machine vision systems to read because they lack clear character segmentation.

[0082] In some examples, the method of the present invention can be used to provide character segmentation in machine-readable information corresponding to human-readable information including connected characters.

[0083] In some examples, the machine-readable information includes at least one embedded marker. The embedded marker may include at least one discontinuity or gap within at least one human-readable character. The gap is strategically selected and placed in a predetermined position such that the machine-readable information corresponds to predetermined human-readable information. Specific gap placement helps the machine vision system clearly distinguish information from similar shapes or forms, such as… Figure 3 Those shapes or forms shown.

[0084] Figure 3 The human-readable characters 110A, 110B, 110C, and 110D are shown. These characters are similar in shape and difficult to distinguish by the current ALPR system, such as “0” (zero), “O” (the letter “O”), “8”, and “B”.

[0085] Figure 4 An example of machine-readable information 200 according to this application is shown. The machine-readable information includes several independent (i.e., unconnected) machine-readable information 2010A, 2010B, 2010C, 2010D, wherein each machine-readable information 2010A, 2010B, 210C, 210D includes an embedded marker. The embedded marker shown includes at least one continuous portion 220 and discontinuous portions or gaps 230. Character segmentation 240 between adjacent machine-readable information is also shown.

[0086] An example of embedded markers includes discontinuous portions or gaps present in machine-readable information. Another example of embedded markers includes geometric shapes or symbols present in machine-readable information, such as at least one of, for example, a dot or circle (hollow or solid), a triangle, a square, a star, an asterisk, a character outline, etc.

[0087] Such embedded markers are strategically placed and their size is set. In some examples, this application relates to methods for encoding machine-readable information that is subsequently decryptable. In one aspect, the method of the invention includes an encoder that takes any basic symbol (e.g., star, dot, character outline, geometry) or character as input to produce machine-readable information including embedded markers as output. In some examples, the encoder relies on maximizing the minimum pairwise character dissimilarity over a predetermined set of constraints (or increasing that dissimilarity to a value greater than a specific threshold). In this particular example, character dissimilarity is measured by the L1 norm between any two characters. The embedded marker placement method of the invention allows the use of any pairwise distance function (e.g., pairwise difference), where “similar” characters are those with the minimum pairwise distance. Machine-readable embedded markers are placed within characters to increase the pairwise distance between the most similar characters. With the method of the invention, the embedded marker placement strategy can obtain the maximum (or greater than a threshold) distance between the most similar characters. For example, the method of the invention can provide at least 80% dissimilarity greater than or equal to the maximum dissimilarity within a dissimilarity range.

[0088] The exemplary set of constraints includes, but is not limited to, at least one of the following: the size of the embedded marker, the number of embedded markers, the percentage of the remaining original characters, the orientation of the embedded marker, the position of the embedded marker between characters, and the shape of the embedded marker.

[0089] In some examples, the term "human-readable information" refers to information that can be read and understood by humans without additional machine translation. Examples of human-readable information include, but are not limited to, alphanumeric characters, drawings, geometric shapes, symbols, and Asian or Arabic language characters. In some examples, human-readable information does not include coded information that humans cannot understand without additional machine translation, such as, for example, barcodes. In some examples, under certain conditions, human-readable information may be visible or invisible to appropriate optical detectors (e.g., the human eye).

[0090] In some examples, the term "machine-readable information" refers to information encoded in a form that can be optically imaged by a machine or computer and is readable by the hardware and software of the machine or computer rather than by a human. In some examples, under certain conditions, machine-readable information may be visible or invisible to appropriate optical detectors (e.g., cameras).

[0091] In some examples, human-readable information is visible to an optical detector in a first spectral range but invisible to the detector in a second spectral range, while machine-readable information is visible to the optical detector in the second spectral range but invisible to the first spectral range. In some examples, the machine-readable information is embedded machine-readable information. In some examples, the first spectral range is from about 350 nm to about 750 nm (i.e., the visible light spectrum), and the second spectral range is from about 700 nm to about 1100 nm (i.e., the near-infrared spectrum). In some examples, the first spectral range is from about 700 nm to about 850 nm, and the second spectral range is between about 860 nm and about 1100 nm. In some examples, human-readable information is visible to the detector under a first illumination condition but invisible to the detector under a second illumination condition, while machine-readable information is visible to the detector under the second illumination condition but invisible to the first illumination condition. In some examples, the first illumination condition is an ambient visible condition (i.e., diffuse visible light), and the second illumination condition is a visible retroreflection condition (i.e., coaxial visible light). In some examples, the position of one or more light sources differs under the first lighting condition from that under the second lighting condition.

[0092] In some examples, machine-readable information is presented in the form of binary optical codes. In binary optical codes, the code can be divided into known regions of a predetermined number and geometry. Regions in the image can then be classified as bright or dark areas. Bright areas or pixels represent values ​​(e.g., 0 (zero)), and dark areas represent another value (e.g., 1). The large contrast (i.e., brightness difference) between bright and dark areas makes it easier to interpret the binary optical codes.

[0093] In some examples, human-readable and / or machine-readable information is printed on a retroreflective substrate. Suitable printing techniques include screen printing, flexographic printing, thermal mass transfer printing, and digital printing such as, for example, laser printing and inkjet printing. One advantage of using digital printing is that information can be easily and quickly customized / changed to meet customer needs without the need to produce new screens or flexographic sleeves.

[0094] In some examples, the machine-readable information is embedded machine-readable information. In these examples, the printing of the human-readable information and the machine-readable information is performed in alignment such that they completely overlap. In some examples, the human-readable information is printed first on the retroreflective substrate, followed by the embedded machine-readable information. In other examples, the machine-readable information is printed before the human-readable information. In some examples, the human-readable information is printed with a visually opaque infrared-transparent ink (e.g., CMY ink), which makes the information visible in the visible spectrum but invisible in the infrared spectrum. In some examples, the machine-readable information is printed using a visually opaque infrared-opaque ink (e.g., an ink containing carbon black), which makes the information visible in both the visible and infrared spectra. In some examples, the machine-readable information is printed on the substrate using a visually transparent infrared-opaque ink. In some examples, the materials described in co-pending U.S. Patent Application 61 / 969889 (Attorney's File No. 75057US002) are used to print the human-readable information and / or the machine-readable information, the entire disclosure of which is incorporated herein by reference.

[0095] In some examples, the machine-readable information includes at least one of infrared reflective, infrared scattering, and infrared absorbing materials. The use of these materials produces contrast in the infrared spectrum and thus appears “dark” when viewed under such conditions. Exemplary materials that may be used include those listed in U.S. Patent 8,865,293 (Smithson et al.), the disclosure of which is incorporated herein by reference in its entirety.

[0096] Figure 5A An illustrative example of this application under diffuse visible light is shown. In this case, the retroreflected article 300 and human-readable information including the characters “EXPLORE Minnesota.com” 310 and “485JBP” 315 are visible. Figure 5B Shown under infrared radiation Figure 5A The retroreflected article 300 is shown. In this case, human-readable information including the characters “EXPLOREMinnesota.com” 310 is undetectable, and the characters “485JBP” are invisible to the detector. Under infrared radiation, machine-readable information 320 including embedded markers 330 (e.g., gaps) is visible.

[0097] Figure 6 Another illustrative example of this application under diffuse visible light is shown. In this case, the retroreflected article 400 and human-readable information including the characters “EXPLORE Minnesota.com” 410, the characters “485” 415a, and “485JBP” 415b are visible. Figure 7It shows the infrared radiation. Figure 6 The reflective article 400 is shown. In this case, a portion of the human-readable information, specifically the character "485", is substantially invisible, while the embedded machine-readable information 420, including embedded markers 430a, 430b, and 430c, is visible.

[0098] In some examples, such as Figures 5A to 7 As shown, machine-readable information occupies at least some of the same space as human-readable information and is described as embedded machine-readable information. The embedded machine-readable information is completely contained within the boundaries of at least a portion of the human-readable information. In some examples, the embedded machine-readable information is occluded by the human-readable information under a first condition.

[0099] In some examples, the machine-readable information includes embedded markers comprising at least one continuous portion and at least one discontinuous portion, wherein the continuous and discontinuous portions together correspond to human-readable information. For example, Figure 4 The embedded markers for character 210A include continuous portions 220 and discontinuous portions or gaps 230, which are read as the digit "0" (zero) by the machine vision system in combination. As described above, the software used in the machine vision system of this application includes algorithms that identify strategically placed embedded markers in machine-readable information and match the machine-readable information with known human-readable information (such as, for example, human-readable characters). Therefore, the machine vision system is allowed to correctly read machine-readable information and associate it with predetermined human-readable information.

[0100] In currently available ALPR systems, the rotation, alignment, scaling, and skew of retroreflected articles detected by optical detectors are calculated using assumed knowledge about the size and shape of the retroreflected articles. These methods do not take into account whether the retroreflected articles to be detected and read are vertical or properly aligned. Therefore, existing systems may incorrectly miss or fail to detect articles that have rotated, for example, due to environmental conditions such as rain and wind.

[0101] Figure 8 This is a flowchart illustrating an exemplary method for reading information provided on a retroreflected article of manufacture according to this application. In the illustrated method, after capturing an image, the retroreflected article of manufacture is positioned within the image, and then the image of the retroreflected article of manufacture is normalized. After the image of the article of manufacture of manufacture is normalized, the disclosed method uses an optional character segmentation step, followed by selecting a machine-readable information dictionary such that the machine-readable information corresponds to predetermined human-readable information. Thus, the machine-readable information is classified and read.

[0102] In some examples, the method of the present invention utilizes a preprocessor to provide a normalized and ready-to-process image as output, such as... Figure 9 As shown. The preprocessor uses an image captured by an optical detector and determines whether high-contrast regions exist in the image. This process is called image filtering. In some examples, high contrast is achieved by using an optically active substrate (e.g., a reflective or retroreflective substrate). Images without high-contrast regions are discarded. The remaining images are then moved to a second preprocessing step where high-contrast regions are separated and normalized. Normalization includes, for example, deskipation, rotation, resizing, and scaling of the high-contrast regions. After normalization, machine-readable information corresponds to predetermined human-readable information.

[0103] like Figure 5B and Figure 7 As shown, the method of the present invention can also provide additional inputs to the preprocessor, increasing the likelihood of generating a properly normalized image of the retroreflected article. Exemplary additional inputs include, but are not limited to, alignment markers comprising vertical and / or horizontal segments parallel to the license plate edge, and embedded markers comprising a predetermined space between the markers. In an example where the predetermined space between the markers is used, this known distance can be used to determine whether the retroreflected article is rotated or skewed. For example, if the distance measured in the image is shorter than the known predetermined distance, the system may assume that the retroreflected article is skewed.

[0104] In some examples, the method of the present invention also includes an OCR (Optical Character Recognition) engine that binarizes the normalized image. The image is then segmented into individual characters using a character separator. The symbols are compared with any active dictionary loaded into the engine, and the most likely match is selected. Finally, each of these symbols is classified according to the dictionary, resulting in a set of symbols.

[0105] Figure 10 This flowchart illustrates an exemplary example of how a machine-readable dictionary can be used to map machine-readable information to predetermined human-readable information. First, a set of constraints is generated based on rules governing the OCR engine and camera system (e.g., resolution, minimum embedding tag size, maximum number of embedding tags per character, allowed embedding tag shapes, etc.). This set of constraints, along with the input dictionary containing human-readable fonts or information, is included in an iterative embedding tag placement algorithm. After each iteration, the convergence of the solution is tested. If convergence has not yet been achieved, the algorithm is iterated again. Once convergence is achieved, the output is a new dictionary containing the new maximum set of dissimilar characters for the embedded machine-readable information.

[0106] Figure 11AAn exemplary license plate 800 according to this application is shown. The license plate 800 includes human-readable information 815, 817, and 819 visible under a first condition (i.e., diffuse ambient light), and machine-readable information including embedded markers 830 and 831 visible under a second condition (i.e., infrared light). The embedded markers 831 are strategically configured to produce character segmentation for additionally connected characters. The embedded markers 831 create “vertical lines” in the characters, making such markers usable for determining the orientation and rotation of the license plate.

[0107] The retroreflective article chosen for any particular implementation will depend on the desired optical, structural, and durability characteristics. Thus, the desired retroreflective article and materials will vary based on the intended application. Retroreflective articles and materials include reflective and retroreflective substrates. As used herein, the term "retroreflective" refers to a property that an object possesses such that the direction of reflection of obliquely incident light rays is antiparallel to or approximately antiparallel to their incident direction, such that the obliquely incident light rays return to a position at or immediately adjacent to the light source. Two known types of retroreflective sheets are microsphere-based sheets and solid-angle sheets (commonly referred to as prism sheets).

[0108] Microsphere-based sheets are often referred to as “beaded” sheets. They consist of a large number of microspheres that are typically at least partially embedded in an adhesive layer and have associated specular or diffuse reflective materials (e.g., pigment particles, metal flakes, vapor-deposited layers) to reflect incident light back.

[0109] Solid angle retroreflective sheets, often referred to as "prism" sheets, include a body portion that typically has a substantially flat front surface and a structured rear surface comprising multiple solid angle elements. Each solid angle element includes three generally mutually perpendicular optical surfaces. A sealing layer can be applied to the structured surfaces to keep the individual solid angles away from contaminants. Flexible solid angle sheets may also be incorporated into the examples or embodiments of this patent application. The retroreflective sheet used in conjunction with this application can be, for example, rough or smooth.

[0110] The retroreflective articles described herein are generally constructed as sheets applicable to a given object or substrate. The article is generally unilaterally optical. That is, one side (designated as the front) is typically adapted to receive incident light from a light source and emit reflected or retroreflected light toward a detector (such as an observer's eye), and the other side (designated as the back) is typically adapted to be applied to an object, such as via an adhesive layer. The front faces both the light source and the detector. The article generally does not allow a large amount of light to be transmitted from the front to the back, and vice versa, at least in part due to the presence of substances or layers on the retroreflector, such as metal vapor deposits, sealing films, and / or adhesive layers.

[0111] One application of the retroreflective article described herein is for license plates detected by a license plate detection or recognition system. An exemplary license plate detection system uses a camera and lighting system to capture an image of the license plate. The image of the scene including the license plate can be captured under conditions of ambient visibility and additional light provided by a designated light source (e.g., a coaxial illumination device that shines light directly onto the license plate when the camera is ready to record an image). The light emitted by the coaxial illumination device, combined with the retroreflective properties of the license plate, produces a strong bright signal from the license plate location in an additional large image scene. This bright signal is used to identify the location of the license plate. Automatic license plate recognition (ALPR) then focuses on the region of interest (the bright region) and searches for a match between the expected human-readable information or embedded machine-readable information by looking for a recognizable contrast pattern. In this application, only machine-readable information or symbols need to be read by the ALPR system.

[0112] In some examples, light in driving and ALPR environments can be divided into the following spectral regions: visible light in the region between approximately 350 nm and approximately 700 nm, and infrared light in the region between approximately 700 nm and approximately 1100 nm. Typical cameras have sensitivity encompassing both ranges, but the sensitivity of standard photographic systems decreases significantly for wavelengths longer than 1100 nm. Various light-emitting diodes (LEDs) can emit light across this entire wavelength range, and most LEDs are typically characterized by a central wavelength and a narrow distribution around that wavelength. For example, in a system comprising LEDs emitting light at a wavelength of 830 nm ± 20 nm, a suitably equipped camera can detect license plates in the near-infrared spectrum using light invisible to the vehicle driver. Therefore, the driver will not see the "flickering" effect of the LED and will not be distracted by it.

[0113] In some examples, the camera and light are typically mounted at an angle to the direction of vehicle movement to observe the license plate. Exemplary mounting locations include positions above traffic flow or from the side of the road. Images can be collected at angles ranging from 20 to 45 degrees to the vertical angle of incidence (front) of the license plate. Detectors sensitive to infrared or ultraviolet light may be used, as appropriate, to detect reflected light outside the visible spectrum. Exemplary detectors include cameras, including those sold by 3M Company, (St. Paul, MN), of St. Paul, Minnesota, including but not limited to the P372.

[0114] The retroreflective article described herein can be used to improve the capture efficiency of these license plate detection or recognition systems. ALPR capture can be described as the process of correctly locating and identifying license plate data, including but not limited to signage, license plate type, and license plate origin. Applications of these automated systems include, but are not limited to, electronic toll collection systems, red-light violation systems, speed enforcement systems, vehicle tracking systems, trip timing systems, automatic identification and alarm systems, and vehicle access control systems. As mentioned above, current vehicle license plate recognition systems achieve lower capture efficiency than desired due to factors such as low or inconsistent contrast of signs and blurred or distracting contrast of illustrations and / or markings on the license plate.

[0115] In some examples, the retroreflective article of this disclosure can also be used for signage. As used herein, the term "signage" refers to an article that typically uses alphanumeric characters, symbols, graphics, or other markings to convey information. Specific examples of signage include, but are not limited to, signs, road signs, conspicuous sheets, window stickers, identification materials (e.g., licenses), and vehicle license plates for traffic control purposes. In some examples, it may be advantageous to use the article of this application to achieve the desired performance of viewing machine-readable barcodes under visible light without altering the appearance of the signage. Such retroreflective articles will make the reading of specific information on the signage generally consumer-meaning, while avoiding distraction of drivers or sign readers by "hidden" markings and / or detection of unwanted "hidden" markings, such as variable information in barcodes. Such developments facilitate the use of invisible markings and / or signaling in articles of article for safety purposes, identification, and inventory control. For example, hidden markings may contain sign-specific information, such as, for example, signage material batch numbers, installation dates, reorder information, or product lifespan.

[0116] In another aspect, this application relates to an optical character recognition process, comprising the following steps: providing a retroreflective substrate including human-readable information and embedded machine-readable information including at least one embedded marker; detecting the machine-readable information and reading the machine-readable information using an optical character recognition engine; wherein the machine-readable information corresponds to human-readable information.

[0117] In another aspect, this application relates to a system for reading an article of literature, the article of literature comprising a retroreflected article containing human-readable information and machine-readable information, the machine-readable information including at least one gap and at least one solid portion; an optical detector that detects and generates an image of the retroreflected article; a processor that preprocesses the image to locate the retroreflected article on the image and normalize the image, and locates machine-readable information on the retroreflected article; and an optical character recognition (OCR) engine that associates the machine-readable information with the human-readable information.

[0118] In another aspect, this application relates to a method of preparing a retroreflective article, comprising providing a retroreflective substrate; applying human-readable information having a periphery to the retroreflective substrate; and applying machine-readable information including embedded markers and solid portions, wherein the machine-readable information is disposed on the retroreflective substrate and contained within the periphery of the human-readable information. In some examples, at least one of the steps of applying human-readable information and applying machine-readable information includes printing.

[0119] The objectives and advantages of this application are further illustrated by the following examples, but the specific materials and quantities thereof referenced in the examples, as well as other conditions and details, should not be construed as undue limitation of the invention. Those skilled in the art will recognize that other parameters, materials, and apparatus may be used.

[0120] Example 1 :

[0121] Preparation such as Figure 5A and Figure 5B The license plate 300 shown is provided. A first retroreflective sheet (available under the trade name "DigitalLicense Plate Sheeting 3750" from 3M Company, (St. Paul, MN)) is provided. On the retroreflective sheet, a background graphic 310 including an image of trees and lakes and the text "Explore Minnesota.com" is printed on the upper portion of the retroreflective sheet, and human-readable information 315 including the text "485 JBP" is printed in the center. The background graphic 310 is printed using visually opaque infrared transparent cyan, magenta, and yellow (CMY) inks (available under the trade names "3M UV Inkjet Ink 1504 Cyan", "3M UV Inkjet Ink 1504 Magenta", and "3M UV Inkjet Ink 1506 Yellow" from 3M Company (3M)). Visually opaque infrared-opaque black ink (available from 3M Corporation under the trade name "3M UV Inkjet Ink 1503 Black") is used to print human-readable information 315. Both background graphics 310 and human-readable information 315 are printed using a single-pass UV inkjet printer (available from 3M Corporation under the trade name "3M High Definition Printer").

[0122] A second retroreflective sheet similar to the first retroreflective sheet is provided. Embedded machine-readable information 320, including embedded marker 330, is printed in black on the second retroreflective sheet using a combination of visually opaque and infrared-transparent CMY inks. The embedded machine-readable information 320 is then cut out and placed on human-readable information 315 such that the machine-readable information completely overlaps with and is contained within the boundaries of human-readable information 315. The printed retroreflective sheet is then adhered to an aluminum substrate (not shown).

[0123] Figure 5A Image of license plate 300 taken under diffused ambient lighting. Background graphics 310 and human-readable information 315 are visible under the ambient lighting. Figure 5B This is an image of the same license plate taken under retro-infrared (IR) illumination. The image was captured using a digital SLR camera (model D5500, commercially available from Nikon Corporation (Melville, NY)). For the IR image, the camera features an IR cutoff filter removed from the sensor and a Hoya R72 filter for filtering visible light (commercially available under the Hoya name from Kenko Tokina, Tokyo, Japan).

[0124] Under retroreflected IR illumination, human-readable information 315 is essentially invisible, while machine-readable information 320, including embedded markers 330, becomes visible to the detector (i.e., the camera). The embedded markers 330 of the machine-readable information 320 are strategically placed as gaps to maximize character differentiation, as previously described. Furthermore, some of these gaps are selected to aid in license plate alignment, which occurs during the preprocessing steps of the OCR system. Specifically, gaps similar to those within the characters “5”, “J”, and “P” are selected for license plate alignment.

[0125] Example 2 :

[0126] like Figure 6 and Figure 7 The license plate 400 shown is prepared as generally described in Example 1, except that the machine-readable information 420 includes embedded markers 430a, 430b and 430c prepared as follows.

[0127] As described in Example 1, machine-readable information 420 is prepared on a second retroreflective sheet. The character "485" is printed using the visually opaque infrared-transparent ink of Example 1. The character "4" is cut out from the second retroreflective sheet, and then seven holes, each with a diameter of 0.25 inches (0.635 cm), are cut out from the character "4" to produce an embedded marker 430a. The embedded marker 430a is placed on the character "4" of the human-readable information 415 such that it completely overlaps with and is contained within the boundaries of the human-readable information 415. Thus, the embedded marker 430a comprises a circle that becomes visible under retroreflective infrared light.

[0128] The character "8" printed on the second retroreflective sheet is cut out and trimmed into a thinner, narrower version of the same character, resulting in an embedded marker 430b. The embedded marker 430b is then placed on top of and overlaps with the character "8" of the human-readable information 415.

[0129] Similarly, the character "5" printed on the second retroreflective sheet is cut out from the second retroreflective sheet. Then, five holes, each with a diameter of 0.25, are cut out from the character "5" and manually set onto the character "5" of human-readable information 415, thereby producing an embedded marker 430c.

[0130] Figure 6 Image of license plate 400 taken under diffused ambient lighting. Background graphics 410 and human-readable information 415 are visible under the ambient lighting. Figure 7 This is an image of the same license plate 400 taken under retro-infrared (IR) illumination. Under retro-infrared IR illumination, the background graphics 410 and human-readable information 415a are essentially invisible, while the machine-readable information 420, including embedded markers 430a, 430b, and 430c, becomes visible to the detector (i.e., the camera).

[0131] Example 3 :

[0132] like Figure 11A and Figure 11B The license plate 800 shown is prepared as generally described in Example 1, except as mentioned below.

[0133] A retroreflective sheet (commercially available from 3M Company under the trade name "High Definition License Plate Sheeting 6700") is provided. A patterned background graphic 810 is printed on the entire area of ​​the retroreflective sheet using CMY inks and a UV inkjet printer as described in Example 1. A blue rectangle 812 is printed on the upper portion of the retroreflective sheet using CMY inks. Human-readable information containing the characters "Dubai 5 ADUAE" 815 and Arabic numerals 817 is concentrated and printed on the blue rectangle using white CMY inks. Human-readable information 819 containing black Arabic numerals is concentrated and printed on the retroreflective sheet using black infrared transparent ink as described in Example 1.

[0134] Using the black, visually opaque, infrared-opaque ink of Example 1, the embedded machine-readable information 820, including embedded markers 830 and 831, is printed in a concentrated manner aligned with human-readable information 819.

[0135] Apply a vinyl coating (available commercially from 3M, "9097 Vinyl Protective Overlay Film") to the printed reflective sheet.

[0136] Figure 11A Image of license plate 800 taken under diffused ambient lighting. Background graphics 810 and human-readable information 815, 817, and 819 are visible under the ambient lighting. Figure 11B This is an image of the same license plate 800 taken under retro-infrared (IR) illumination. Under retro-infrared IR illumination, the background graphics 810 and human-readable information 815, 817 and 819 are essentially invisible, while the machine-readable information 820, including embedded markers 830, 831, becomes visible to the detector (i.e., the camera).

[0137] Embedded markers 831 are strategically positioned to create character segmentation for additional connected characters. Furthermore, embedded markers 831 create "vertical lines" within the characters, making such markers usable for determining the orientation and rotation of the license plate.

[0138] Figure 12 A flowchart illustrating example operation of a computing device configured to perform the techniques of this disclosure is provided below. For illustrative purposes only. Figure 1Example operations are described within the context of computing device 1032. Computing device 1032 may select a specific symbol from a symbol set (1200). Computing device 1032 compares a first spatial appearance of the specific symbol with a second spatial appearance of another symbol in the symbol set, wherein the first and second spatial appearances are based on specific lighting conditions (1202). Computing device 1032 may determine, at least partially based on this comparison, that the dissimilarity between the first and second spatial appearances satisfies a first threshold (1204). Computing device 1032 may identify a specific location from a plurality of locations within the specific symbol that, when an active coding region is embedded, increases the dissimilarity between the updated spatial appearance of the specific symbol and at least the second spatial appearance to satisfy a second threshold (1206). Computing device 1032 may modify the symbol to store data representing an active coding region embedded in the specific symbol at the specific location (1208). Computing device 1032 may update the symbol set to include a specific symbol that includes data representing an active coding region (1210). In some examples, the IR version of the image can be used alone without requiring visual information (human-readable and / or machine-readable information) to determine framing, and then additionally using IR (machine-readable) to extract the embedded data. In some examples, computing device 116 can use only the machine-readable information used for OCR and then derive human-readable information from the OCR output (e.g., recognized symbols). In some examples, it is not necessary to identify human-readable information before performing the techniques of this disclosure.

[0139] Figure 13 A flowchart illustrating example operation of a computing device configured to perform the techniques of this disclosure is provided below. For illustrative purposes only. Figure 1 Example operations are described within the context of computing device 1016. Computing device 1016 may receive (1300) an image (of an optically active article) from an image capturing device, comprising a set of one or more symbols of a symbol set. At least one symbol of the set of one or more symbols represented in the image may include a set of one or more coded regions embedded within the symbol. Computing device 1016 may perform optical character recognition on a specific image region of the image comprising at least one symbol to determine, at least in part, that the specific image region represents at least one symbol (1302). The arrangement of the set of one or more active coded regions within the at least one symbol provides character dissimilarity between at least one symbol and another symbol in a symbol set that satisfies a predefined threshold. Computing device 1016 may perform one or more operations (1304) at least in part based on determining that the specific image region represents at least one symbol.

[0140] In one or more examples, the described functionality may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, then these functions may be stored or transmitted as one or more instructions or code on or through a computer-readable medium and executed by a hardware-based processing unit. A computer-readable medium may include a computer-readable storage medium corresponding to a tangible medium such as a data storage medium, or a communication medium that includes any medium facilitating the transfer of a computer program from one place to another, for example, according to a communication protocol. In this way, a computer-readable medium may generally correspond to (1) a tangible computer-readable storage medium that is non-transitory, or (2) a communication medium such as a signal or carrier wave. A data storage medium may be any available medium accessible by one or more computers or one or more processors to retrieve instructions, code, and / or data structures for implementing the techniques described in this disclosure. Computer program products may include computer-readable media.

[0141] By way of example, and not limitation, such computer-readable storage media may include RAM, ROM, EEPROM, CD-ROM or other optical disc storage devices, magnetic disk storage devices or other magnetic storage devices, flash memory, or any other medium that can be used to store desired program code in the form of instructions or data structures and that is accessible by a computer. Furthermore, any connection is appropriately referred to as a computer-readable medium. For example, if instructions are transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. However, it should be understood that while computer-readable storage media and data storage media do not include connections, carrier waves, signals, or other transient media, they instead involve non-transient tangible storage media. Disks and platters used include compact discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, where discs typically reproduce data magnetically, while platters optically reproduce data using lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0142] The instructions can be executed by one or more processors, such as one or more digital signal processors (DSPs), general-purpose microprocessors, application-specific integrated circuits (ASICs), field-programmable arrays (FPGAs), or other equivalent integrated or discrete logic circuit systems. Therefore, the term "processor" as used may refer to any or any other of the aforementioned structures suitable for implementing the technology. Furthermore, in some aspects, the functionality may be provided within dedicated hardware and / or software modules. Additionally, the technology may be fully implemented in one or more circuit or logic elements.

[0143] The techniques disclosed herein can be implemented in a wide variety of devices or apparatuses, including wireless telephone handsets, integrated circuits (ICs), or IC sets (e.g., chipsets). Various components, modules, or units are described in this disclosure to enhance functional aspects of a device configured to perform the disclosed techniques, but they do not necessarily need to be implemented by different hardware units. Rather, as described above, various units may be combined in hardware units along with suitable software and / or firmware, or provided by collecting a set of interoperable hardware units including one or more processors as described above.

[0144] It should be recognized that, depending on the examples, certain actions or events in any of the methods described herein may be executed in a different order, or may be added together, combined, or omitted (e.g., not all of the described actions or events are necessary for the practice of the method). Furthermore, in some examples, actions or events may be executed simultaneously rather than sequentially, for example, through multithreaded processing, interrupt handling, or multiple processors.

[0145] In some examples, computer-readable storage media include non-transitory media. In some examples, the term "non-transitory" indicates that the storage medium is not embodied in a carrier or propagating signal. In some examples, non-transitory storage media stores data that may change over time (e.g., in RAM or cache).

[0146] Various examples have been described. These examples, and others, are all within the scope of the following claims.

Claims

1. A method comprising: The image of an optically active article is received by a computing device and from an image capturing device, the image comprising a set of one or more symbols of a symbol set, wherein at least one of the one or more symbols represented in the image comprises a set of one or more active coding regions embedded within the symbol; Optical character recognition is performed on a specific image region of the image that includes the at least one character, so that the computing device determines that the specific image region represents the at least one symbol based at least in part on the set of one or more active coding regions, wherein the arrangement structure of the set of one or more active coding regions within the at least one symbol provides character dissimilarity between the at least one symbol and another symbol in the symbol set that satisfies a predefined threshold; as well as The computing device performs one or more operations based at least in part on determining that the specific image region represents the at least one symbol.

2. The method of claim 1, wherein the active coding region of the particular symbol is printed with a visually opaque infrared-transparent ink, and the remaining area of ​​the particular symbol excluding the active coding region is printed with a visually opaque infrared-opaque ink.

3. The method of claim 1, wherein the active coding region of the particular symbol is printed with a visually opaque infrared-opaque ink, and the remaining area of ​​the particular symbol excluding the active coding region is printed with a visually opaque infrared-transparent ink.

4. The method according to claim 1, Wherein, the first set of pixel values ​​representing one or more active coding regions within the one or more coding regions is within the range of the first pixel value. The second set of pixel values ​​representing the at least one symbol, excluding the remaining portion of the one or more active coding regions, falls within a second pixel value range different from the first pixel value range.

5. The method according to claim 4, The image is a first image, and the first image of the optically active article is captured in a first spectral range within the near-infrared spectrum. The second image of the optically active article is captured in a second spectral range within the visible spectrum. The third set of pixel values ​​representing the at least one symbol in the second image is within the range of the second pixel values, and the first proportion of the third set of pixel values ​​representing the at least one symbol is greater than the second proportion of the second set of pixel values ​​representing the at least one symbol.

6. The method of claim 1, wherein the image is captured under illumination conditions within the near-infrared spectrum.

7. The method of claim 1, wherein at least one symbol represents human-readable information in a first spectral range, wherein the first spectral range includes wavelengths between 350 nm and 750 nm.

8. A retroreflective article, comprising: retroreflective substrate; A set of one or more symbols disposed on the retroreflective substrate, wherein the arrangement of one or more active coding regions at one or more predetermined positions within at least one of the one or more symbols in the set of one or more symbols provides character dissimilarity between at least one symbol and another symbol in the set based on the arrangement of the one or more active coding regions at the one or more predetermined positions, satisfying a predefined threshold.

9. The method of claim 8, wherein visually opaque infrared-transparent ink is disposed at the set of one or more active coding regions, and visually opaque infrared-opaque ink is disposed on the remaining area of ​​the particular symbol excluding the set of one or more active coding regions.

10. The method of claim 8, wherein visually opaque infrared-opaque ink is disposed at the set of one or more active coding regions, and visually opaque infrared-transparent ink is disposed on the remaining area of ​​the particular symbol excluding the set of one or more active coding regions.

11. A method for preparing an optically active article, comprising: Receive user input for specified information; At least in part based on the user input, a printing specification is generated including a specific symbol having one or more active coding regions, wherein the printing specification specifies a set of one or more symbols included in an optically active article, wherein the one or more active coding regions embedded in the specific symbol provide a dissimilarity between a first spatial appearance of the specific symbol and at least a second spatial appearance of a second symbol based on an arrangement structure of the one or more active coding regions at one or more predetermined positions, satisfying a predefined threshold. as well as The optically active article having the specific symbol is constructed at least in part based on the printing specifications.

12. The method of claim 11, wherein constructing the optically active article with the specific symbol comprises: The one or more active coding regions of the specific symbol are printed with a visually opaque infrared transparent ink, and the remaining areas of the specific symbol excluding the one or more active coding regions are printed with a visually opaque infrared opaque ink.

13. The method of claim 11, wherein producing the optically active article comprises: The one or more active coding regions of the specific symbol are printed with a visually opaque infrared-opaque ink, and the remaining areas of the specific symbol excluding the one or more active coding regions are printed with a visually opaque infrared-transparent ink.

14. The method of claim 11, wherein the threshold is the maximum dissimilarity within the dissimilarity range.

15. The method of claim 11, wherein the second threshold is at least 80% of the maximum dissimilarity within the dissimilarity range.

16. The method of claim 11, wherein the second threshold is at least one of a hard-coded value, a user-defined value, or a machine-generated value.

17. The method according to claim 11, Wherein, the first set of pixel values ​​representing the one or more active coding regions within one or more coding regions is within the range of the first pixel value. The second set of pixel values ​​representing the at least one symbol, excluding the remaining portion of the one or more active coding regions, falls within a second pixel value range different from the first pixel value range.

Images