JP5826774B2 - Method for authenticating contents and special label for authenticating contents - Google Patents

Description

Claim of priority This application is the priority of Indian Patent Provisional Application No. 625 / CHE / 2010 (subject “X-ray image based item counter”) filed with the Chennai Patent Office on March 10, 2010. Claims the interests of rights.

Specification The following specification describes the essence of the present invention.

The present invention belongs to the field of performing contents inspection inside a sealed container regarding the number of contents or whether the contents are genuine from the viewpoint of proof of alteration and prevention of counterfeiting . Modern customers can keep track of their consignments, but the present invention allows the contents to be monitored and inspected at each transit point in transit. Direct benefits, and if the item is stolen, shape, size, and when the weight was swap with fake attached genuine label the same, a Rukoto intrusion location are identified, thus is to provide the described functionality in each passing point.

  As a method for preventing the theft of goods, a method such as an RF-ID tag is known. RF-ID tags are an excellent solution for inventory management and tracking, but are not a reliable technology for preventing theft / replacement and hidden loss of items in the supply chain. The main limitation is the fact that the RF-ID reader only counts the number of RF-ID labels, not the actual number of items. Therefore, if someone pulls out an item from the box and returns only the RF-ID label to the box, the number remains correct. If a genuine RF-ID label is affixed to a fake product that has been replaced with a real product, there is no distinction and no mutual check is performed. Furthermore, the RF-ID tag does not automatically acquire its own credential every time it is reused.

  The main objective of the present invention is to detect the type of intrusion that becomes hidden only as a result, remains undetectable and untraceable, and becomes apparent only when the box is opened at the delivery location. The method not only relates to the number of items but also detects whether the item has been reordered or replaced with a fake of the same shape, size and weight. Disclosed is an improved and more powerful method of article theft / replacement detection technology that is also protected from insider threats.

  The following description illustrates the method and design of the present invention.

The present invention, internal layering of radiopaque material are arranged in random, we propose a special purpose label embedded therein. These labels are tagged at designated locations on a product basis or on a box basis. Although visually all tags from the outside look the same, they are, inside, Propelled by one only spread radiopaque layering in a small fraction of the total label area, the radiopaque Sex layering automatically acquires a spatially random orientation after application to the package. The orientation can be measured at the time of imaging relative to the intersection of the box edges or any fixed packaging location. After tagging, the items are stacked in a plurality of layers arranged in rows and columns in a box. By performing X-ray imaging of the box, a row and column tagging matrix can be seen. If any item is stolen, the composite imaging result changes. When replaced with a fake, the orientation changes even if the designated place is tagged with a real label, and the imaging result comparison module detects this change. In another exemplary embodiment intended for tamper proof and anti-counterfeiting applications, instead of performing internal imaging such as x-rays, the label is imaged externally. That is, these labels have an imprinted pattern for machine vision to calculate an automatically acquired spatial orientation relative to a fixed external reference point of the package. In both embodiments, if manufactured exactly the same, the label should be visually symmetric regardless of the orientation of the package, and the circular shape is an example.

In another embodiment, these labels can have a pattern printed randomly with invisible ink and can be scanned with a special purpose camera. If it is difficult to use invisible inks, it is possible to make these labels invisible to humans by placing labels with black patterns under the red and blue screens. This embodiment merely proves that the box has not been opened in the passage channel and has not been repackaged with real-looking tape. The term “pattern” should not be understood as referring to a particular shape, color and size. This is because the pattern can be manufactured in an arbitrary shape, character shape, or symbol shape and various dimensions. The pattern on the label can be printed, punched on the label, or another layer applied to the label surface. The label can be attached by any method from manual to fully automatic attaching machines. The label may be hybrid. That is, it is possible to provide both a visible pattern and an invisible pattern at the same place. Similarly, the spatial orientation of one label can be placed on another similar label affixed to the same package relative to a reference point on the package. Although infrared has been referred to for the display of invisible patterns, those skilled in the art will be able to use other spectra as well.

The niche attributes are listed below.
・Inspect the contents inside the sealed box at each passing point , detect alteration, and counterfeit inspection・ Provide explanation and audit functions in all directions for internal parties ・ Unauthorized unpacking and rearrangement of items • Detection of replacement of original goods and counterfeit • Enables content monitoring in normal customer satisfaction tracking while transporting valuable or fragile consignments.
・ Because the evidence is concrete, it can be accepted as legal evidence

The invention described herein is illustrated by way of example and not limitation in the accompanying drawings. For simplicity and clarity of illustration, the components shown in the figures are not necessarily drawn to scale. For example, for clarity, the dimensions of some components may be exaggerated over other components.
FIG. 6 is a tagging diagram of internal layering of radiopaque material. FIG. 5 is an internal tagging diagram showing radiopaque layering in a random orientation, although all tags appear to be symmetrical when viewed from the outside. FIG. 3 is an illustration of an item tagged on two different sides of the box. It illustrates tagged X-ray imaging result of goods stacked in a box with a label having a different orientation of the radiopaque Reiyarin grayed. (004) and (005) are X-ray images from two different faces of the box. It is a figure which shows the X-ray imaging result after one item was extracted. It is a figure which shows the X-ray imaging result after one item is replaced with the fake and the equivalent label is tagged. As a result, the orientation of the radiopaque pattern of the label tagged on the replaced item changes. FIG. 5 is a diagram showing the top surface of an item tagged with a symmetrical label in an imaging result with all items intact in the container. FIG. 6 is a diagram illustrating the top surface of an item tagged with a symmetrical label in an imaging result indicating that the pattern spacing has changed due to one item being removed from the container. In a given row and column, five items are stacked in the depth direction, and each item is tagged with a label, but because one pattern overlaps and hides, four patterns It is a figure which shows the X-ray image of the side surface which is displayed only. In a given row and column, 5 items are stacked in the depth direction, each item is labeled with a label, and the angled X-ray reveals the hidden pattern. It is a figure which shows the X-ray image of the side surface by which all five patterns are displayed. FIG. 4 shows two labels having an invisible ink pattern. One label has a reference pattern, the other has an actual comparison pattern, and both the reference line and the comparison pattern are printed on the label with ink. These are invisible to the human eye and can only be digitally scanned by a suitable infrared camera. FIG. 6 shows two labels of another type. Two circles are printed with invisible ink on the reference circular adhesive label. Another circle is on the base tape. The angle with respect to the base adhesive tape is calculated according to the orientation of the circular adhesive tape by image processing. These adhesive tapes are self-destructable so that they cannot be reused. FIG. 6 illustrates one embodiment of a radiopaque label with random orientation. Invisible radiopaque material layering is in the form of small circles randomly embedded anywhere in the larger circular radiolucent label, depending on the orientation of the larger circular label Thus, a unique imaging signature is automatically acquired. It is a figure which shows the X-ray imaging result of the box which has a stack | stuck of a banknote. There are 4 bills in the length direction, 5 bundles in the height direction, and 5 bundles in the depth direction, and each bundle is tagged with a smart label. It is a figure which shows the X-ray imaging result of the box stuffed with the thermocoll block so that a radiation transparent medium may enter into the path | route of an X-ray beam. There are two circular labels under the plastic straps, and in order to steal, you need to mess with the plastic straps, which will change the direction of the labels. FIG. 5 shows an omnidirectional scan of a trapezoidal box including all faces and edges by using an image acquisition platform with two cameras.

  The following description describes a method and apparatus of technology based on X-ray imaging of lost / stolen / replaced items. In the following description, numerous specific details are set forth such as system component implementations, types, interrelationships, etc., in order to provide a thorough understanding of the present invention. However, as will be appreciated by one skilled in the art, the present invention may be practiced without such specific details. In order to avoid obscuring the present invention, various low level details and complete instruction sequences not directly related to the present invention have not been described in detail. Those skilled in the art will be able to implement the appropriate functionality without undue experimentation according to the description contained herein.

  References to “embodiments” herein may include particular features, structures, or characteristics of the described embodiments, but for all embodiments, the particular features, structures, or characteristics are It shows that it does not necessarily need to be included. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with one embodiment, this is within the scope of the knowledge of those skilled in the art, regardless of whether it is explicitly described or not. It shall affect such relevant features, structures, or characteristics.

  In the following description and claims, the term “radiopaque layering” inside the tag is shown to refer to a straight, triangular, or circular shape, but this may be any random shape. However, it should be understood that the shape is not limited because it can be manufactured. How the labels are applied on the inside “or” outside of the box depends entirely on the specific deployment requirements of each customer. Similarly, although the present invention proposes the use of a circular label as the best security level is obtained, the label may be any other shape.

  Embodiments of the present invention will be described below.

Tagging with labels with embedded radiopaque layering in a random orientation Each item must be tagged with a special purpose label. FIG. 1 shows a human visible label (001) and a radiopaque layer (002) embedded therein. Each tag looks symmetrical when viewed from the outside, but inside each tag has a radiopaque layering oriented in a random direction, as shown in FIG. Another type of label is shown in FIG. Here, a circular label that appears symmetrical has an inner layer (018) of radiopaque material. In the usual case, each box is tagged with one label. However, as shown in FIG. 3, a plurality of labels (003) can be tagged on the box. These labels obtain their own credential only at the time of application, and automatically obtain their new credential each time they are reused. The second embodiment is an external scanning solution that does not use X-ray labels, has an invisible pattern, can be digitally scanned by a special purpose camera, and is determined by the orientation of the invisible pattern relative to the pattern of the reference label Obtain your own credential automatically.

Reference X-Ray Imaging This imaging must be done at the packaging site in the high physical security surveillance area. Images of the sealed box are taken from two different sides of the box. That is, as shown in FIG. 4, shooting is performed in the direction in which the items overlap (004) and the direction in which they do not overlap (005). After the box is sealed, internal imaging (004) and (005) is performed. The reference imaging indicates an imaging result in which radiopaque layering of random orientations of a plurality of tags having different stacked items are overlapped. The reference image is indexed with the unique ID of the box, digitally signed, and transmitted to the central server.

X-ray imaging at passing points and delivery locations By comparing the X-ray imaging results of the boxes at each passing point and delivery location with the reference image, the radio-opaque layering X-ray imaging results of random orientation can be obtained. An overlapping matrix of labels is shown. When someone rearranges, steals, or replaces an item with a fake, the imaging result comparison module detects this by comparing it with the reference image of that particular box.

Item Loss / Theft Detection If any item is lost or stolen, the stacked imaging results of the stacked items are significantly different in that particular row or column (006) and ( 007), the image comparison module detects this and generates an alert.

Detecting replacement of items If any item is replaced with a fake that is similar in shape, size, and weight, the imaging results that overlap for the stacked items in that particular row or column (008) And (009) change. The image comparison module detects this and generates an alert.

Detecting the reordering of items When the positions of two similar items are swapped, the overlapping imaging results change for stacked items in that particular row or column. The image comparison module detects this and generates an alert.

Reduction of false alarms Even if one of the items is not tagged due to an unintentional human error, this will not cause false alarms because the reference image captures it.

In rare cases, the direction of the labels of the stacked items will be the same. If the direction of two items is completely the same by some chance, they may not be detected even if they are lost or stolen. However, if it is replaced, it is detected. This is because it is unlikely that the direction of the replaced item will be the same again. However, in order to further reduce this rare possibility, the use of two tags and combining mathematical probabilities that completely overlap effectively eliminates the case where the orientation of two items is exactly the same. It is possible to exclude.

Indexing images at the central server At the time of reference imaging, each box is assigned a randomly generated unique ID. The unique ID is affixed to the box, preferably on the same side as the reference image. A digital signature is applied to the unique reference imaging result of each box, and for comparison, a reference image indexed with the unique ID of the box is fetched. All subsequent imaging needs to be performed from the same side as the reference imaging. In order to enhance the protection of the network channel, the fetch of the reference imaging result can be cryptographically protected by a known protocol such as IPSEC / SSL.

Image normalization For a perfect comparison, it is necessary to clean up the X-ray imaging results. Wooden boxes and cardboard boxes may contain metal parts such as nails and coffins. Therefore, every time X-ray imaging is performed, it is necessary to subtract the X-ray imaging result of the reference empty box from the imaging result.

Detecting horizontal movement of an item If the image capture module analyzes that there is a difference between the normalized reference image and the normalized current hop captured image, the image capture module further detects the difference. Investigate whether this is due to slight horizontal movement of the goods in transit. The full imaging result of the box is cut out (trimmed) at a plurality of portions in the length direction and the height direction. Each cut-out portion represents an article stacked in the depth direction. To perform horizontal motion detection, the clipped portion is subtracted from the corresponding clipped portion of the reference image, and then the clipped portion of the reference image is replaced with the corresponding clipped portion of the recently captured image. Subtract from the marked part. After these two subtractions, each pixel can be examined for horizontal shift. If this is not real horizontal movement, the number of pixels shifting in the horizontal direction is negligible.

Size-based analysis of the number of items This mode is used to count the number of items. In this case, all labels have a symmetrical and identical layering of radiopaque material and no random orientation layering. For example, if the tag is circular, the radiopaque layering will be the same size circle. These tags are affixed on the top surface of each item at a location that is radiolucent in the X-ray beam direction. In X-ray imaging in which the X-ray beam is upward from downward, the size of the label closest to the flat panel detector is larger. In the unlikely event that an item is extracted, the number of items extracted from the box can be calculated based on the maximum size of the circle in the X-ray imaging pattern. When the item has not been extracted, as shown in FIG. 7, all the circles (010) in the X-ray imaging result have the same size. The amount of size change (011) as shown in FIG. 8 is determined by the number of extracted items. This method works best for X-ray equipment using flat panel detectors, where all items in the box are standard in size and shape, and some benchmarks are known in advance. This is the case.

Alternative absolute counting method to solve counting omission due to pattern overlap without using dual beam X-ray imaging In the size-based analysis described above, additional X-ray imaging from the top and further label orientation Regardless, it is necessary to use an additional label that is symmetrical during imaging. One alternative method proposed here does not require the addition of X-ray imaging from the top, and similarly utilizes the same label for both image comparison and absolute item counting. These labels have a pattern of radiopaque material, such as small circular or small rectangular patterns (for example only). The imaging software simply counts the number of small patterns in each row and column in X-ray imaging from the side. The number of small patterns in a given row and a given column gives the number of items stacked in the depth direction. However, counting conflicts can occur when one or more patterns completely or partially overlap each other. This counting conflict can be solved by angled x-rays, which can be achieved by simply rotating the box to a predetermined angle on the conveyor belt of the x-ray machine. Hiding the X-rays reveals hidden patterns. For example, assuming that five items are stacked in the depth direction, in FIG. 9, since one pattern is hidden by overlapping, small circles are “A” and “B” in (012). , “C”, and “D” are only displayed. When the box is rotated, as shown in FIG. 10, the fifth pattern “E” is exposed in the angled X-ray (013). Shown is a 1 inch × 1 inch region cut from the full imaging result of the box, which represents the imaging results of 5 items stacked in depth for a given row and column. ing. Within the angled X-ray, the 1 inch x 1.5 inch stretched area is cut out, revealing the hidden pattern "E", and this information is digitally recorded during the reference imaging itself And can be verified at the delivery location.

Counting partially overlapping circular patterns during X-ray imaging This method is applicable only if they are partially overlapping and the pattern is circular. This method ensures that when a circle is traversed along its circumference, the absolute value of the absolute difference (whether positive or negative) between the x and y coordinates of the two equidistant points on the circumference is always It takes advantage of the fact that they are the same (constant). If two or more circles partially overlap, the perimeter will not be geometrically circular, in which case the absolute value of the absolute difference between the x and y coordinates will be along the edges of the overlapping pattern. It does not remain constant for the same distance analyzed. Each time the total absolute difference between the x and y coordinates changes, the number of overlapping patterns increases and this traversal continues until the first starting point on the pattern circumference is reached.

Hidden features that prevent counterfeiting related to obvious features Hidden features that prevent counterfeiting are often required. In this embodiment of the invention , box authentication is described. The open edge of the box is sealed with a label printed with invisible patterns (014A) and (014B). Notably, the label used in this embodiment, the internal layering of the radiopaque material is not used is that of using an invisible pattern. A special purpose scanning device that can scan in a range invisible to the human eye, and scans the pattern digitally from the outside. As can be seen in the scanned imaging result of the label after application to the package, the pattern can be any shape / size, or simply a single line in a random orientation. In the case of a line, the spatial orientation of the imprinted pattern relative to a fixed reference point (014C) on the package can be an imaging credential. Imaging credential may be associated with the printed random packaging, therefore, even the insiders that can not be re-cloned and swap, ing unique signature piecewise. In production lines, special purpose image acquisition system scans each package from the outside, associate imaging credentials of each packaged in unique alphanumeric credential. This association is recorded in the database. The random pattern may be any shape and size. The reference imaging result is captured by a special scanner, and the spatial orientation extracted from the image is associated with a unique ID in piece units. During validation in the field, the extracted spatial orientation from the imaging results, according to the program, again, compared to the spatial orientation of the criteria associated with unique ID.

A two-label approach for direction-independent external scanning This embodiment is probably useful in the mass transit industry. In such an industry, it is not possible to tag smart labels with internal random radiopaque layering on an item-by-item basis, so only box-by-box authentication is required, and box sizes vary. It is decided according to the order. If the shape of the box is trapezoidal, it is possible to scan all surfaces and edges by scanning the upper and lower surfaces. In this embodiment, two labels are used as shown in FIG. One is a base label or a reference label that provides a reference pattern (014). The other is a comparison label (015) having a comparison pattern (016). The imaging result of the reference pattern is used to specify the direction of the comparison pattern image. Therefore, image analysis is performed based on the spatial orientation of the comparison image with respect to the reference image regardless of the scanning direction of the box. Both patterns are invisible to the naked eye (preferably printed with ink invisible to the human eye), but can be digitally scanned by a suitable camera. Preferably, these labels should be self-destructible to increase security. Comparison labels should look visually symmetric in shape, size, and application, differing only in pattern orientation, pattern shape, and pattern size, but these are not visible to the human eye Printed with ink. These labels must be applied so that the opening is sealed. These labels may be two concentric labels, and a small diameter base label may be applied over a large diameter reference label. In order to index the images in the database, a unique ID may also be printed on these labels with invisible ink. The unique ID may be in the form of any tracking technology, such as barcode, QR code, RFID tag, or magnetic stripe, to integrate security and tracking, and a copy of the integrated tracking credentials Detected based on the spatial orientation of the label. In another embodiment, there may be two invisible circles on the reference label (017), as shown in FIG. 12, and the angle and distance to the circle on the base label are calculated. When it is impossible to create a pattern using invisible ink, a combination of a red screen and a blue screen can be used as an adhesive tape for a label. In this case, the pattern is printed in black. For example, when a pattern is printed in black on a reference label with a red background and a semi-transparent dark blue tape is used, the pattern becomes invisible, but digital scanning is possible using an infrared camera. Similarly, as will be appreciated by those skilled in the art, various other color combinations can be used to render the imprinted pattern invisible.

Internal monitoring on box labels
In content inspection, which is an internal scanning embodiment, tagging an item with a label may not be feasible from a transportation perspective, but internal monitoring is still necessary. The challenge is that the items that can be housed in the box may be metallic or non-metallic. The box is padded with a thermocol block or any other radiolucent material and strapped along these. Thermocol block stuffing is suitable because the X-ray imaging results of the metal contents inside the box partially overlap with the smart label imaging results, and the metal contents inside are rearranged slightly May be necessary. Prior to applying the plastic or elastic strap, two labels are placed on the thermocol block and when the strap is applied, these labels automatically obtain their imaging credentials. For example, as shown in FIG. 15, the credential may have a relative orientation (019) and (020), but is not limited thereto. The box is placed in a larger container, and the credential is inspected by X-ray imaging. If there is an intrusion that cannot be detected, the strap must be cut, in which case the label will change its orientation and the imaging credentials will be changed automatically. In the case of external scanning embodiments with tamper detection, the adhesive surface covers both the elastic strap and the box surface when applying the label. The spatial orientation of the label is calculated and combined with the markings on the elastic belt to form an omnidirectional detection detection credential. It is impossible to make modifications from the other side / edge of the box, at least without moving the elastic strap, and if so, it will be detected based on changes in the omnidirectional modification detection credentials.
Utilization of spatial orientation
By setting the angle increment to 0.05 degrees, one circular label can have 7200 unique combinations. To increase the number of unique combinations, using three adjacent labels, the total number of unique combinations exceeds 300 billion (7200 × 7200 × 7200). If invisible patterns are not available, the label can have multiple visible patterns that have the same shape and color or dimensions that are mixed and narrow in a narrow range, so that only the scanner can distinguish and select individual patterns. By making it possible to calculate the spatial orientation, it is possible to make replication very difficult even when using a visible pattern.

Claims (13)

A label containing at least one spatial orientation of the imprint pattern for the calculations in at least one differential scanning capable area,
A database configured to store the spatial orientation of the label as a unique credential for the object in association with unique identification data of the labeled object;
The spatial orientation of the label attached to the object is identified, and an alert is generated based on a comparison result between the identified spatial orientation and a credential associated with the object stored in the database. A comparison module configured to generate;
A system comprising: The label is a visually symmetrical with respect to the orientation in the plane of the object which the label is assigned, according to claim 1 system. The at least one imprinted pattern of the label is a specific color pattern, a first layer of a first color different from the specific color, and different from the specific color and the first color Arranged between a second layer of a second color, so that the at least one imprinted pattern is invisible, but a special illumination camera is used to allow digital scanning. The system of claim 1. The system of claim 1, wherein the label is circular. The system of claim 1, wherein the at least one imprinted pattern comprises an invisible pattern. The system of claim 1, wherein the at least one imprinted pattern includes a radiopaque material disposed between layers of the label. The system of claim 1, wherein the spatial orientation includes an angle and a distance of the imprinted pattern relative to at least one external reference point on the object. A method performed in a system using a label that includes at least one imprinted pattern for spatial orientation calculation in an area capable of at least one differential scan, comprising:
Assigning the label to the object;
Storing the spatial orientation of the label relative to the object as a credential in a database in association with the unique identification data of the labeled object;
Identifying the spatial orientation of a label attached to the object;
Generating an alert based on a comparison between the identified spatial orientation and a credential associated with the object stored in the database. 9. The method of claim 8, wherein the label is visually symmetric with respect to orientation in the surface of the object to which the label is applied . A method that allows object authentication,
Affixing a security label on the object, the security label comprising a pattern;
Scanning the security label to obtain a first spatial orientation, wherein the first spatial orientation is between the pattern and a reference point on the object outside the security label. Scanning based on the spatial relationship of
Identifying a second spatial orientation, wherein the second spatial orientation is based on a spatial relationship that includes the reference point on the object outside the security label;
Authenticating the object in response to an image match between the first and second spatial orientations;
Not performing authentication of the object in response to an image mismatch between the first and second spatial orientations, the image mismatch removing the security label from the object; and the security label on the object Re-orienting, not implementing by changing the object, changing the position of the object, or replacing the security label on the object with another security label. The method of claim 10, wherein applying the security label on the object comprises applying the security label on a package of the object. Affixing the security label on the object includes affixing a security label having a plurality of layers, and the pattern is placed between the first layer of the security label and the second layer of the security label. The method of claim 10, wherein the method is in a layer to be peeled. The method of claim 10, wherein the security label is visually symmetric with respect to orientation in a surface of the object to which the security label is applied .

Images