202 corresponds to light reflecting from the physical credential 202 in a first manner (e.g., as detected/received by a sensor) and a second, different orientation of the physical credential 202 corresponds to light reflecting from the physical credential in a second, different manner (e.g., as detected/received by the sensor).

The computing device 214 may determine a type of the physical credential 202 based on at least one received image of the physical credential 202. In various embodiments, the computing device 214 may determine the type of the physical credential 202 based on one of the first received image or the second received image of the physical credential 202. As an example, the computing device 214 may determine that the physical credential 202 is a Virginia driver's license based on a received image of the physical credential 202 in the first orientation 280. In various embodiments, multiple images of the physical credential 202 may be used to determine the type of the physical credential 202. In various embodiments, the computing device 214 may determine the type of the physical credential 202 based a determined shape or size of the physical credential 202 and/or based on any determined feature included on the physical credential (e.g., any textual or graphical feature identifying the type of the physical credential 202).

Based on a determined type of the physical credential 202, the computing device 214 may determine one or more secure features associated with the physical credential 202 such as, for example, the secure feature 208. As an example, after determining that the physical credential 202 is a Virginia driver's license, the computing device 214 may determine that the physical credential 202 includes a secure feature 208 that is a hologram.

Light reflected off of the secure feature 208 (e.g., the reflected ray 212) may be used to determine a physical property of the secure feature 208 (and/or the physical credential 202) such as, for example and as discussed above, an optical characteristic of the secure feature 208. In various embodiments, the optical characteristic of the secure feature 208 may be determined based on the reflected light in two or more images of the physical credential 202 and/or the secure feature 208 in two or more orientations including, for example, a video sequence of the physical credential 202 and/or the secure feature 208 in a variety of orientations. Based on the reflected light in two or more images, more reliable and more accurate determinations of one or more optical characteristics of the secure feature 208 may be determined, thereby improving detection of authentic physical credentials and inauthentic physical credentials.

The optical characteristics of the secure feature 208 that may be determined may include, for example, color, refractive index, transparency, an intensity of reflected light, a scattering of light, or a movement of a glare reflected by the secure feature 208. As an example, the secure feature 208 may be a laminate covering the physical credential 202. The light reflected off of the laminate secure feature 208 (e.g., the reflect light ray 212) may be used to determine the refractive index of the laminate secure feature 208 and/or a transparency of the laminate secure feature 208.

As another example, a glare of the light reflected off of the laminate secure feature 208 (e.g., the reflected light ray 212) may be detected. In various embodiments, a movement of the glare between the first and second images may be determined. In various embodiments, a change in a size of the glare between the first and second images may be determined.

As a further example, a color of light reflected off of the secure feature 202, a scattering of light reflected off of the secure feature 208, and/or an intensity of light reflected off of the secure feature 208—or a change in any of these optical properties—may be determined.

In various embodiments, any determined physical and/or optical characteristic of the secure feature 208 may be compared to a corresponding expected physical and/or optical characteristic of the secure feature 208. In various embodiments, based on the determined type of the physical credential 202, known properties of the secure feature 208 may be determined. Based on the known properties of the secure feature 208 and/or the first and second orientations 280 and 282 of the physical credential 202, the computing device 214 may determine the expected physical and/or optical characteristic of the secure feature 208. In various embodiments, the computing device 214 may determine the first and second orientations 280 and 282 of the physical credential 202 based on the first and second images of the physical credential 202. In various embodiments, the expected physical and/or optical characteristic of the secure feature 208 may also be determined based on a lighting level of the environment of the physical credential 202.

In general, the computing device 214 may use the changing light reflected off of the secure feature 208—from different orientations of the secure feature 208—to measure and/or detect an optical characteristic of the secure feature 208. The computing device 214 may also use known information regarding the secure feature 208 and the physical credential 202 (e.g., the lighting of the physical credential 202, the orientation of the physical credential 202, etc.). The computing device 214 may then compare the determined optical characteristic of the secure feature 208 to the expected optical characteristic of the secure feature 208 to determine if the physical credential 202 is authentic.

For example, if the determined optical characteristic of the secure feature 208 matches the expected optical characteristic of the secure feature 208, then the computing device 214 may decide the physical credential 202 is authentic. If the determined optical characteristic of the secure feature 208 does not match the expected optical characteristic of the secure feature 208, then the computing device 214 may decide the physical credential 202 is inauthentic.

In various embodiments, the computing device may determine a first value (e.g., based on a determined metric) indicating the determined optical characteristic of the secure feature 208 and may determine a second value (e.g., based on the determined metric) indicating the expected optical characteristic of the secure feature 208. The first and second values may be compared. If the first value is within a predetermined range of the second value, the computing device 214 may decide the physical credential 202 is authentic. If the first value is outside of the predetermined range of the second value, the computing device 214 may decide the physical credential 202 is inauthentic.

In various embodiments, one or more additional features of the physical credential 202 may be detected and/or measured to contribute to any determination whether the physical credential 202 is authentic or inauthentic. In various embodiments, a shape, size, and/or thickness of the physical credential 202 may be determined and may be used to determine an authenticity of the physical credential. Any feature on the physical credential 202—such as any graphical, textual, or pictorial features—may also be used to determine an authenticity of the physical credential. In various embodiments, authentication may be based on a presence of such features and/or confirmation that any such feature matches an expected characteristic (e.g., if a watermark included on the physical credential 202 matches an expected design).

In various embodiments, machine learning may be used to facilitate authentication of a physical credential. In various embodiments, authentic and inauthentic physical credentials may be used to train a system that may use machine learning to apply training data to detection of authentic and inauthentic physical credentials.

In various embodiments, the computing device 214 may operate in conjunction with one or more computing devices including any local or remote computing device to perform any of the operations described herein. In various embodiments, information regarding known properties of secure features may be stored by a remote computing device that the computing device 214 interacts with to obtain such information or any information derived therefrom.

Having discussed an example arrangement for determining an authenticity of a physical credential, discussion will now turn to an illustrative method for determining an authenticity of a physical credential

FIG. 3 illustrates an example method 300 for determining a physical characteristic of a physical credential (e.g., the physical credential 202) in accordance with one or more aspects described herein. Method 300 may be implemented based on the example arrangement depicted in FIGS. 2A and 2B. Method 300 may be implemented by a suitable computing system, as described herein. For example, method 300 may be implemented in any suitable computing environment by a computing device and/or combination of computing devices, such as computing devices 101, 105, 107, and 109 of FIG. 1 and/or computing device 214 of FIGS. 2A and 2B. Method 300 may be implemented in suitable program instructions, such as in software 127, and may operate on data, such as data 129.

At step 302, a light level of an environment of the physical credential may be determined. The physical credential may be in any environment such as inside any physical space or may be outside in an open or non-enclosed area. The environment of the physical credential may be illuminated by any light source including natural light and/or artificial light. The light level of the environment may be determined by ambient light of the environment. A sensor may be used to determine the light level of the environment by, for example, determining an intensity of light received by a sensor.

At step 304, the determined light level of the environment of the physical credential may be compared to a light level threshold. The light level threshold may represent a minimum intensity of light for receiving images of the physical credential. If the determined light level of the environment is lower than the light level threshold, then at step 306 an action to increase the light level of the environment may occur. In various embodiments, at step 306, a supplemental light source may be automatically activated that may be used to illuminate the physical credential. For example, a light source on a computing device (e.g., a smartphone, a 3D reader/scanner) may be automatically turned on. In various embodiments, at step 306, a request may be issued to a user to increase an illumination of the physical credential (e.g., a request to hold the physical credential under a light, to turn a light on, or to move the physical credential to a space with a higher light level). As an example, a notification may be issued to the user that the determined light level is insufficient and that the light level of the environment is to be increased. If the determined light level of the environment is higher than the light level threshold, then the method 300 may proceed to step 308.

At step 308, multiple images of the physical credential may be received. In various embodiments, a video sequence of the physical credential may be received. The video sequence may include multiple images of the physical credential. In various embodiments, two or more images of the physical credential may be received. In various embodiments, each received image may correspond to a different orientation of the physical credential. In various embodiments, multiple images of the physical credential in the same orientation may be received provided at least one additional image of the physical credential in a different orientation is received.

The orientation of the physical credential may be changed by moving the physical credential (e.g., by a user holding the physical credential moving the physical credential or holding the physical credential differently). The orientation of the physical credential may be relative to a light source associated with the physical credential or a senor (e.g., an optical sensor) receiving the images of the physical credential such that moving one or more of the physical credential, the sensor, and/or the light source changes the relative orientation of the physical credential.

The images of the physical credential may be received by any type of sensor including a camera or video camera. In various embodiments, the sensor may be rotated or moved relative to the physical credential when the physical credential is held in a stationary position to receive the multiple images and/or video sequence of the physical credential.

In various embodiments, the received images and/or video sequence of the physical credential may be stored in a memory of a computing device. In various embodiments, at least a first image of the physical credential in a first orientation is received and a second image of the physical credential in a second orientation is received.

At step 310, a type of the physical credential may be determined. The type of the physical credential may be determined based on at least one received image of the physical credential. The type of the physical credential may be determined based on one or more physical features or components of the physical credential including the shape and size of the physical credential or any feature applied to the physical credential (e.g., any textual, graphical, pictorial feature or markings including watermarks, stamps, seals, etc.).

The physical credential may be any type of physical credential as described herein including, for example, an ID card, a document, or a financial instrument. A database or other memory (e.g., either remote or local to the physical credential) may be used to determine the type of the physical credential. In various embodiments, a local or remote database or memory component may store information regarding various characteristics and features of various physical credentials. The stored known characteristics of various physical credentials may be consulted to determine a specific type of the physical credential in the at least one received image of the physical credential.

At step 312, a secure feature or component of the physical credential may be determined (e.g., the secure feature 208). The secure feature may be determined based on the determined type of the physical credential. The secure feature may be a physical feature of the physical credential. The secure feature may be any type of secure feature as described herein including, for example, a hologram, a layer of plastic covering all or a portion of the physical credential, a stamp, a watermark, a graphical image, a textual image, or a photographic image.