CN117751378A - Digital watermarking system and method - Google Patents

Abstract

A digital watermarking method includes receiving, by an electronic processor, an original image signal comprising a series of original visual images, wherein the original image signal is encoded using a Perceptual Quantizer (PQ) luminance level coding transfer function to produce PQ luminance steps within the original image signal, and wherein the PQ luminance steps have a magnitude that varies across a luminance range. The method further includes receiving, by the electronic processor, a watermark image signal including the watermark and adjusting the strength of the watermark by at least a first weighting factor, the first weighting factor being a predetermined first number of PQ brightness steps.

Description

Digital watermarking system and method

Cross Reference to Related Applications

The present application claims priority from U.S. provisional patent application Ser. No. 63/228,220, filed 8/2 of 2021, and European patent application Ser. No. 21189088.4, filed 8/2 of 2021, each of which is incorporated herein by reference in its entirety.

Technical Field

Embodiments disclosed herein relate to visual anti-theft protection. More specifically, embodiments disclosed herein provide systems and methods of creating digital watermarks to identify illegal copies of visual works.

Background

Pirates are a well known problem to illegally record content such as movies, concerts, and proprietary activities. Typically, these recorded copies are sold by pirates to obtain profits or free distribution (e.g., via the internet), thereby depriving the content legal owners of revenue. Existing methods of preventing unauthorized copying of copyrighted works operate at the machine level to prevent the recording machine from creating unauthorized copies. However, these methods do not prevent live copying of copyrighted works. Furthermore, existing methods of preventing live copying of copyrighted works are not universally effective for all video signals.

Disclosure of Invention

The present disclosure provides systems and methods for creating digital watermarks. Embodiments provided herein include a digital watermarking method comprising receiving, by an electronic processor, an original image (video) signal comprising a series of original visual images, wherein the encoded original image (video) signal uses a Perceptual Quantizer (PQ) luminance level encoding transfer function to produce PQ luminance steps within the original image (video) signal, and wherein the PQ luminance steps have a magnitude that varies across a luminance range. The method further includes receiving, by the electronic processor, a watermark image signal including the watermark and adjusting the strength of the watermark by at least a first weighting factor, the first weighting factor being a predetermined first number of PQ brightness steps.

Embodiments provided herein include a system for creating a digital watermark. The system includes a memory and a controller coupled to the memory and including a processor configured to: receiving an original image signal comprising a series of original visual images, the original image signal being encoded using a Perceptual Quantizer (PQ) luminance level encoding transfer function to produce PQ luminance steps that vary across a luminance range; receiving a watermark image signal comprising a watermark; and adjusting the strength of the watermark by at least one weighting factor, the at least one weighting factor being a predetermined first number of PQ brightness steps.

Other aspects will become apparent by consideration of the detailed description and accompanying drawings.

Drawings

Fig. 1A is a diagram showing a difference in brightness perception between a human eye and a camera.

FIG. 1B illustrates the effectiveness of an exemplary luma level coding function in creating a uniform distribution of "just noticeable differences" (JND) steps across a luma range.

Fig. 2 is a diagram schematically illustrating the operation of a Perceptual Quantizer (PQ) level coding function across a luminance range and illustrating force components for determining the strength of a watermark.

Fig. 3A provides an exemplary watermark image according to a first polarity polarization.

Fig. 3B shows the watermark image of fig. 3A polarized according to a second polarity.

Fig. 4 is a flowchart of a method of creating a digital watermark according to one embodiment of the present disclosure.

Fig. 5 is a flowchart of a method of creating a digital watermark according to one embodiment of the present disclosure.

Fig. 6 is a schematic diagram of a system for creating a digital watermark according to one embodiment of the present disclosure.

Fig. 7 is a schematic diagram of another system for creating a digital watermark according to one embodiment of the disclosure.

Fig. 8 is a schematic diagram of another system for creating a digital watermark according to one embodiment of the disclosure.

Fig. 9 is a schematic diagram of another system for creating a digital watermark according to one embodiment of the disclosure.

Detailed Description

It should be noted that the embodiments described herein, or portions thereof, may be implemented using a number of hardware and software based devices, as well as a number of different structural components. Furthermore, it should be understood that the embodiments described herein may include hardware, software, and electronic components or modules that, for discussion purposes, may be shown and described as if most of the components were implemented solely in hardware. However, one of ordinary skill in the art will recognize, based on a reading of this detailed description, that in at least one embodiment, the electronic-based aspects described herein can be implemented in software (stored on a non-transitory computer-readable medium) executable by one or more processors. Thus, it should be noted that the embodiments described herein may be implemented using a number of hardware and software based devices as well as a number of different structural components. For example, the "controller," "control unit," and "control component" described in the specification can include one or more processors, one or more memory modules including non-transitory computer readable media, one or more input/output interfaces, and various connections (e.g., a system bus) connecting the components. Furthermore, as used herein, the terms "movie," "video signal," and the like are intended to refer to image data or an image signal that includes a series of visual images.

Preventing unauthorized copying via digital watermarking

The present application addresses the problem of pirates or other malicious actors creating illegal copies of videos, movies, and other visual works having a series of visual images. A system and method for embedding or overlaying a digital watermark signal into a video signal to detect and/or prevent unauthorized copies of a visual work is provided herein. Previous approaches to detecting and preventing unauthorized copies of such works operate at the machine level by preventing the recording machine (e.g., DVD player or video recorder) from creating copies, or at least preventing the creation of undistorted copies. However, these methods do not prevent live copying of video images on an image capture device (e.g., a camera). For example, these methods do not prevent a person from recording a movie in a movie theater using a camera or a cell phone.

Thus, newer methods of preventing unauthorized copying involve embedding a digital watermark into a movie so that the watermark can be identified even when a pirate records a live movie. An important consideration in creating a digital watermark is the strength of the watermark, where strength is generally understood to refer to the magnitude of the change in brightness of the movie (or any other original image to which the watermark is applied) due to the embedding of the digital watermark. In particular, an effective digital watermark should generally be invisible or not easily noticeable to the human eye, but should be strong enough to be recovered and/or detected by the camera. The embedded watermark is so strong as to be visible to the human eye and reduce enjoyment of the movie by authorized viewers (e.g., patrons in a movie theatre) is undesirable. Also, watermarks that are too weak to be detected by an electronic processor to identify illegally obtained copies of visual works are undesirable.

Existing methods of creating digital watermarks have not generally been successful in achieving a watermark of appropriate strength on all types of video signals. In particular, some existing methods of creating digital watermarks to prevent live unauthorized copying of video images are not ideal for video images of all brightness ranges. Also, existing methods of creating digital watermarks are not ideal for video images that are luma coded using some transfer function. For example, some existing digital watermarking techniques are not optimized for High Dynamic Range (HDR) video images. Some playback devices (e.g., projectors) insert watermarks into movies at presentation. These devices are designed to embed watermarks in movies with limited luminance ranges, such as Standard Dynamic Range (SDR) video images. Existing methods of embedding watermarks in motion pictures operate on motion pictures that are DCI gamma coded according to DCI digital cinema system specifications. However, existing methods of embedding watermarks in video are not adequate for HDR video images. Similarly, some existing methods of embedding watermarks in motion pictures are not ideal when applied to video images encoded using a luma coding function other than a gamma coding function. For example, some existing digital watermarking techniques are not ideal for video images that are luma coded using a Perceptual Quantizer (PQ) transfer function. In particular, some of the existing digital watermarking methods produce watermarks that are either too strong to make the watermark visible to the human eye and interfere with authorized viewers or too weak to make the watermark difficult to recover.

Accordingly, provided herein is a system and method of embedding a watermark signal in a video signal or other signal comprising a sequence of visual images such that the watermark (signal) is not visible to the human eye but is recoverable by an electronic processor and/or an image capture device. By adjusting the strength of the watermark to fall within this range, the method exploits the difference between the ability of the human eye to perceive an image and the ability of the image recording device to capture an image. The disclosed watermarking method may operate on HDR video images. The disclosed watermarking system may also operate on video images that are luma coded using quantization transfer functions (e.g., PQ transfer functions). Examples of PQ mapping functions are described in SMPTE ST 2084:2014, high dynamic range EOTF (High Dynamic Range EOTF of Mastering Reference Displays) of master reference display (hereinafter "SMPTE"), which is incorporated herein by reference in its entirety. In some cases, the watermark signal may include identification information regarding the provider and/or location at which the original authorized video signal was transmitted.

Luminance level coding

The basic theory behind luminance level coding stems from the difference between "luminance" (a physically measurable quantity) and "brightness" (a person's perception of luminance). In short, the human eye perceives light differently than a camera. As shown in fig. 1A, the output signal of the camera varies linearly with the variation of the brightness. For example, when twice as many photons hit the sensor, the output signal of the camera will double. However, humans do not perceive a change in brightness as having a linear relationship with a change in brightness. Instead, as perceived by humans, brightness is similar to a square root function of brightness. Thus, as shown in FIG. 1A, when twice as many photons hit the person's eyes, the person will not detect twice as much brightness. Instead, the person will detect a brightness increase of less than twice the brightness.

The sensitivity of the human visual system to increased or decreased brightness variations is significantly higher at low brightness levels than at high brightness levels. Thus, the human visual system may be able to perceive a difference between a pair of relatively low luminance values that differ by a certain nit (nits) amount, but not a difference between a pair of relatively high luminance values that differ by the same nit amount. In general, human vision can only perceive differences if the luminance values differ by at least a so-called "minimal perceived difference" (JND). Because of the non-linear perception of human vision between brightness and brightness, the amplitude of JNDs varies over a range of light levels, and is typically smaller at lower brightness levels and larger at higher brightness levels.

Thus, to better represent how humans perceive light, video images may be encoded via luminance level encoding. Referring to fig. 1B, encoded image data with equal size or linearly scaled luminance quantization steps does not match the perceived nonlinearity of human vision. Encoded image data having fixed-size luminance quantization steps over the luminance range also does not match the perceived nonlinearity of human vision. The result of the linear luma coding function is shown as element 20 in fig. 1B. Under these techniques, when codewords are assigned to represent quantized luminance values, too many codewords may be distributed in a particular region of the light level range (e.g., a bright region), while too few codewords may be distributed in a different region of the light level range (e.g., a dark region). In the over-distributed area, a large number of codewords may not produce perceptual differences and are therefore wasteful for all practical purposes. In the underdistributed region, two adjacent codewords may produce a perceived difference that is much larger than JND and may produce contour distortion (also referred to as banding) visual artifacts.

However, when the luminance quantization step changes to match the perceptual nonlinearity of human vision, the codeword may be adjusted to have a 1-to-1 correlation with the JND step. In other words, each codeword produces a perceptual difference equal to (or slightly below) JND. Different types of luminance level coding formulas attempt to achieve this goal. The result of the nonlinear luma coding function is shown as element 25 in fig. 1B. Depending on whether the scene light is input or the video signal is input, these luminance coding functions may be commonly referred to as electro-optical transfer functions (EOTF), optical-electrical transfer functions (OETF), or optical-optical transfer functions (OOTF). For example, some video images are encoded using a gamma function (e.g., DCI gamma encoding) or a logarithmic function to encode video signals. Other video images are encoded using the PQ transfer function to encode the video signal.

Fig. 2 depicts various functions across a range of brightness. The x-axis represents the luminance range from 0.0001 nit to 10000 nit. The y-axis represents quantization step size. As previously mentioned, to better represent human perception of light, each quantization step is not equal to one another. Instead, the brightness variation between each step may be varied to match the human perception of light. Thus, the y-axis indicates the increase or decrease (i.e., change) in brightness between each step at a particular brightness level (i.e., on the x-axis). The brightness variation between each step is provided in percent. For example, at higher brightness levels, and for the Perceptual Quantizer (PQ) function 35, the luminance percentage change between steps is about 0.25%. This means that at higher brightness levels, a person will be able to recognize a brightness change that increases or decreases by 0.25%. At lower brightness levels, and for the PQ function 35, the percent change in brightness between steps is large and approaches or exceeds the 10% change in brightness between steps. Although the luminance percentage change of the first JND at low luminance levels is greater than the luminance percentage change of the second JND at high luminance levels, it should be remembered that the luminance absolute change of the first JND is less than the luminance absolute change of the second JND (because the first JND is at a lower luminance level than the second JND).

With continued reference to fig. 2, the luminance level coding function is not universally applicable to all luminance ranges. For example, the gamma encoding function operates only on the luminance of a Standard Dynamic Range (SDR), which is in the range of about 0.001 nit to 100 nit. The gamma coding function does not represent JND for luminance with a High Dynamic Range (HDR) in the range of about 0.0001 nit to 10000 nit. Instead, other types of luminance level encoding functions may be used to encode HDR video. For example, HDR may be encoded using a perceptual quantizer PQ function 35. As shown in fig. 2, the PQ function 35 is applied on HDR luminance.

Existing methods of creating digital watermarks are not optimized for video signals encoded by some brightness level encoding processes to produce the strongest invisible watermark. For example, some existing digital watermarking techniques are not optimized for video signals encoded using quantized luminance level encoding functions (e.g., PQ luminance level encoding function 35). Also, some existing digital watermarking techniques are not optimized for video signals with high dynamic range. In particular, some existing digital watermarking techniques have failed in creating watermarks that are both invisible to the human eye and strong enough to be reliably recovered by an electronic processor. Rather, when applied to video signals encoded by certain luminance level encoding functions, the watermark is either visible to the human eye or cannot be reliably recovered and identified. Accordingly, a method of creating a digital watermark that is effective for an HDR video signal is provided herein. Further, a method of creating a digital watermark effective for video signals encoded using a quantized luminance level encoding function (e.g., PQ luminance encoding function 35) is provided herein. The method described herein provides unexpected results in terms of consistency in watermark quality and luminance ranges that were not possible using previous watermarking methods. Furthermore, the methods provided herein simplify the process of creating digital watermarks and provide a clean formulation method that can be easily applied to HDR video images.

Method for creating digital watermark

As previously described, the watermarking system may add invisible watermark images to the video signal to help identify unauthorized copies of visual works (e.g., movies). Fig. 3A and 3B provide exemplary watermark images that may be used to mark visual works. A watermark image signal comprising a watermark image may be embedded in the video signal such that data may be recovered from the copy of the video. The presence of the watermark alone may indicate that the video is an unauthorized copy. In addition, the watermark may include additional information indicating the location from which the movie was copied (e.g., from which theater the movie was copied) or the time frame in which the movie was copied.

By varying the watermark image in accordance with the binary bit stream, a digital watermark may be embedded as binary data in the video signal, thereby creating a sequence of watermark images (e.g., a sequence of interleaved watermark images 55a and 55 b). The watermark image may be modulated by the bit stream. Each frame of the watermark image sequence may represent one or more data bits, where watermark image 55a represents 0 and watermark image 55b represents 1 (or vice versa). Time modulating the watermark image provides a method of extracting the watermark from the copy of the movie.

In some embodiments, the watermark image may be a fixed image that is modulated by changing its polarity (as shown via watermark images 55a and 55 b). For example, fig. 3A and 3B illustrate two polarities of an exemplary watermark (watermark images 55a and 55B). Specifically, fig. 3A shows watermark 55a having a first polarity (e.g., positive polarity), and fig. 3B shows watermark 55B having a second polarity (e.g., negative polarity). The watermark image signal may modulate the watermark image by changing polarity between a first polarity and a second polarity for each frame in the watermark image sequence. For example, in the first frame of the movie, data bit '1' may add some positive strength (watermark image 55 a) to the movie frame. In the second frame of the movie, data bit '0' may add some negative strength (watermark image 55 b). This process may be repeated for a movie sequence, embedding the data bit stream by adding positive and negative intensities to the watermark images 55a and 55b on a frame-by-frame basis according to the data bits. In other embodiments, the watermark may be a sequence of moving images, possibly related to the actual motion picture image, modulated by changing polarity or modulated by some other method. Furthermore, in other embodiments, the watermark may be a combination of a fixed image and a moving image.

As previously mentioned, an important attribute of an effective watermark is the intensity of the watermark image (e.g., the brightness change of a movie or other image due to embedding the modulated watermark image). If the watermark is too strong it will be visible to the viewer or will produce a flickering effect (because the polarity of the watermark is modulated, the brightness of the movie is increased and decreased in successive frames). If the watermark is too weak, it will be difficult to recover and obtain the data needed to identify the unauthorized copy. It is therefore desirable to create watermarks having an intensity just below the ability of a person to see the watermark or flickering effect. It should be noted that motion in a sequence of motion images that are watermarked may reduce the ability of humans to perceive the watermark or flickering effects. The level at which a person can begin to see the watermark may be referred to as the critical level, or as the critical modulation level of the modulated watermark. Critical modulation levels are an important consideration in designing watermarks. An effective watermark should have an intensity close to but still below a critical level.

In some embodiments, the critical level of the watermark across the luminance range follows a curve similar to the quantized luminance level encoding function (in this case PQ luminance level encoding function 35). Thus, the quantized luminance level coding function 35 may be used as a guide to determine the strength of the watermark such that the watermark signal approaches an appropriate critical level. In particular, the strength of the watermark may be adjusted in proportion to the quantized luminance level encoding function 35. For example, the watermark strength may be adjusted by at least one weighting factor, which is a predetermined number of quantization steps. By applying a function proportional to the quantized luminance level encoding function 35 to adjust the strength of the watermark signal, a simpler system is obtained that produces watermark levels close to the critical level over a larger range of luminance values. In other words, adjusting the strength of the watermark signal in proportion to the step size of the PQ luminance function results in a more efficient watermark as well as a watermark that is efficient over a larger luminance range.

Fig. 4 illustrates a method 100 of creating a digital watermark for use with an original image (video) signal comprising a sequence of original visual images. The method 100 may be performed by one or more controllers having electronic processors. The method 100 includes receiving an original image (video) signal having a series of original visual images, wherein the original image (video) signal is encoded using a luminance level encoding transfer function (step 110). The original image (video) signal may be a movie or a video comprising a series of original visual images perceivable by the human eye. The original image (video) signal may be luma coded using a PQ transfer function. The method 100 further comprises receiving a watermark image signal having a watermark image (step 115). The watermark image may be a geometric shape, symbol, word, or other identifiable marking that may aid in identifying the visual work. In some embodiments, the watermark image is a fixed image that maintains a constant shape. In other embodiments, the watermark image is a sequence of moving images in which the watermark shifts relative position or changes shape. In other embodiments, the watermark image is a sequence of images related to the original sequence of images. The method 100 further comprises adjusting the intensity of the watermark image in proportion to a predetermined number of quantized luminance level encoding steps (120). In some embodiments, the intensity of the watermark image may be adjusted in proportion to a predetermined number of PQ steps. The strength of the watermark may be adjusted by adjusting the amplitude of the watermark signal.

In some embodiments, the watermark may be adjusted and/or modulated at a local level. For example, the strength of the watermark may be adjusted on a pixel-by-pixel level such that each pixel is adjusted independently. As another example, the polarity of the watermark may be adjusted on a pixel-by-pixel level.

Referring back to fig. 2, the strength of the watermark may be adjusted based on a weighting factor operating on the PQ brightness level encoding function 35 such that the strength of the watermark over the brightness range is proportional to the PQ brightness level encoding function 35. For example, the weighting factor may be set to a predetermined number of PQ steps. Since the PQ luminance level coding function follows a curve similar to the critical level, the weighting factor will also be approximately proportional to the critical level over the luminance range, thus achieving the goal of approaching the critical level over the luminance range. Furthermore, by basing the weighting factors on the PQ luminance level encoding function 35, the watermark can be effectively tuned over a larger luminance range (e.g., HDR image). For example, in some embodiments, the watermark strength may be adjusted in proportion to a predetermined number of PQ steps in the luminance range of 0.001 nit to 1000 nit. As another example, in some other embodiments, the watermark strength may be adjusted by a predetermined number of PQ steps in the luminance range of 0.0001 nit to 10000 nit. In other embodiments, the watermark strength may be adjusted by a predetermined number of PQ steps in the luminance range of 0.01 nit to 50 nit. In some other embodiments, the watermark strength may be adjusted by a predetermined number of PQ steps in a luminance range of about 0.0 nit to 100 nit. On the other hand, a weighting factor based on the coding function (e.g., gamma function) used for the SDR image will not produce a watermark whose intensity is proportional to the PQ step size or critical level over the luminance range. Thus, weighting factors based on functions such as gamma transfer functions may produce watermark intensities that are too strong over one portion of the luminance range or too weak over another portion of the luminance range.

In addition, the strength of the watermark may be adjusted by a weighting factor comprising a static force component 45 and a motion force component 50. For example, the strength of the watermark may be adjusted by the following formula: f=sf+mf, where f is the modulation level (or force), sf is the static force component, and mf is the motion force component. Furthermore, in some embodiments, the weighting factor of the static force component 45 may be a predetermined number of quantized luminance level encoding steps. The static force component 45 of the watermark may be present at all times and may be independent of the level of motion in the motion picture image. For example, by maintaining the modulation level below a critical level, the static force component 45 may be invisible regardless of the level of motion in the motion picture image. In some embodiments, the static force component 45 may be set to a constant, where the modulation level of the watermark over the luminance range is proportional to the number of PQ steps and thus the critical level.

The motion force component 50 of the weighting factor depends on the frame-by-frame motion in the motion picture image. A stronger watermark may be allowed in areas of the image with motion because the motion prevents the visibility and flickering effect of the watermark. Thus, the more motion that occurs in a region, the stronger the watermark in that region may be. In order to apply a stronger watermark when a larger motion allows, the strength of the motion preventing the visibility of the watermark should be measured and the strength of the watermark adjusted accordingly. Since the motion preventing the visibility of the watermark has a visual property closely related to the visibility of the watermark, the motion intensity can be measured in units proportional to the critical level and the PQ step size over the luminance range. For the same reason, the strength of the watermark may be adjusted in units proportional to the critical level and the PQ step size over the luminance range. The unit size over the luminance range is notable. The quantity is being adjusted and measured in units of a size over the luminance range proportional to the sensitivity of the attribute of human vision over the luminance range. This property is the visibility of the watermark image sequence, represented by the amplitude of the critical modulation level over the luminance range. This unit size helps and simplifies the creation of the strongest watermark that is not visible.

The motion force component 50 of the weighting factor may relate to spatial and temporal processing of the motion picture image. As with the static force component 45, the motion force component 50 may be adjusted based on the PQ brightness level encoding function. In particular, and depending on the measured motion level in a region of the original image sequence, the motion force component 50 may be proportional to the PQ quantized luminance level encoding step in that region. For example, motion may be measured as a change in PQ quantized luminance level encoding steps between a first image in a series of images and a second image in the series of images, and the motion force component 50 is adjusted by an amount proportional to the measured level. By scaling the measured motion to the PQ step size and scaling the motion force component 50 of the weighting factor to the PQ step size, both the motion and motion force components 50 better represent the associated visual properties over the luminance range. The calculation of the motion force component 50 can also be simplified when optimizing for the strongest watermark that is not visible. Furthermore, motion is more easily measured perceptually and modulation proportional to the critical modulation level is more easily generated.

Fig. 5 illustrates another exemplary method 200 of creating a digital watermark for use with an original image (video) signal comprising a sequence of original visual images. The method 200 may be performed by one or more controllers having electronic processors. Method 200 includes some steps similar to method 100, where like steps are labeled with like numerals. The method 200 includes receiving an original image (video) signal having a series of original visual images, wherein the original image (video) signal is encoded using a luminance level encoding transfer function (step 210). The original image (video) signal may be a movie or a video comprising a series of original visual images perceivable by the human eye. The original image (video) signal may be luma coded using a PQ transfer function. The method 200 further comprises receiving a watermark image signal having a watermark image (step 215). The method 200 may include modulating the watermark image by adjusting the polarity of the watermark image (step 217). The method 200 further includes adjusting the intensity of the watermark image by a weighting factor that is proportional to the predetermined number of quantized luminance level encoding steps (step 220). The strength of the watermark may be adjusted by adjusting the amplitude of the watermark signal. In particular, the method 200 may include adjusting a static force component 45 of the weighting factor (step 222) that is a predetermined first number of PQ steps and adjusting a motion force component 50 of the weighting factor that is proportional to a second number of PQ steps (step 224), wherein the second number of steps is related to image motion measured in a third number of PQ steps, and wherein the PQ steps are utilized such that the step size is proportional to the critical modulation level over the brightness range. The method 200 may further include embedding the adjusted watermark image signal in the original image (video) signal to create an embedded image signal (step 230). The method 200 may also include transmitting the embedded image signal to a playback device, such as a projector, television, computer, or smart device (step 235).

As will be appreciated, in some embodiments, one or more of these steps may not be performed, or conversely, additional steps may be performed. Further, in some embodiments, one or more steps may be performed in a different order or may be performed simultaneously. Variations of the described methods may be performed in accordance with the remainder of this disclosure. For example, the method 200 may not include the step of embedding the adjusted watermark signal in the original image signal (i.e., step 230). Rather, the method 200 may comprise the steps of transmitting the original image signal and transmitting the adjusted watermark image signal, respectively.

Exemplary System for performing digital watermarking methods

Fig. 6-9 provide schematic diagrams of various systems configured to perform the methods described herein. These figures are intended as exemplary embodiments, with various features being interchanged among the exemplary embodiments to create additional embodiments.

Fig. 6 schematically illustrates a system 300 for creating a digital watermark in accordance with the methods disclosed herein. The system includes a controller 310 in communication with an external memory source 320 and a playback device 330. The playback device may be a projector, a television, a computer, a DVD player, a smart device, or any other device capable of displaying a series of visual images. The playback device may have one or more image transmission devices 350. The controller 310 is configured to receive an original image (video) signal having a sequence of original visual images and a watermark image signal having a watermark image from an external memory source 320. The controller 310 performs a method of creating a digital watermark (e.g., methods 100 and 200) and outputs the embedded image signal to the playback device 330. For example, the controller 310 may adjust the strength of the watermark in proportion to a predetermined number of quantization steps (e.g., PQ steps). The controller 310 may then embed the adjusted watermark signal in the original image (video) signal and output the embedded image signal to the playback device 330. The embedded image signal may then be projected by the playback device 330 onto the screen 340.

In some embodiments, the controller 310 may receive the original image (video) signal and/or the watermark image signal from an internal memory within the control unit or within the same computing device. In some embodiments, the controller may be part of or integrated with the playback device. Thus, the controller may not always transmit the embedded image signal or the adjusted watermark signal to the playback device. Furthermore, in some embodiments, the controller may not embed the adjusted watermark signal in the original image signal to create an embedded image signal. Rather, in some embodiments, the controller may create an adjusted watermark signal and may transmit the original image signal and/or the adjusted watermark image signal to the playback device. Similarly, the playback device may not always transmit the embedded image signal (i.e., the combined original image signal and the watermark image signal) onto the screen. Instead, the playback device may overlay the original image signal and the adjusted watermark image signal. Furthermore, in some embodiments, the playback device may include more than one projector or image transmission device to transmit the original image signal, the adjusted watermark image signal, and/or the embedded image signal.

Fig. 7 schematically illustrates another system 400 for creating a digital watermark in accordance with the methods disclosed herein. The system includes a controller 410 in communication with an external memory source 420 and a playback device 430. The playback device may be a projector, a television, a computer, a DVD player, a smart device, or any other device capable of displaying a series of visual images. The controller 410 is configured to receive an original image (video) signal having a sequence of original visual images from an external memory source 420. The controller 410 also receives a watermark image signal having a watermark image from an internal memory source within the controller unit 410. The controller 410 performs a method of creating a digital watermark (e.g., methods 100 and 200) and outputs the embedded image signal to the playback device 430. For example, the controller 410 may adjust the strength of the watermark in proportion to a predetermined number of quantization steps (e.g., PQ steps). The controller 410 may then embed the adjusted watermark signal in the original image (video) signal and output the embedded image signal to the playback device 430. The embedded image signal may then be projected by the playback device 430 onto the screen 440.

Fig. 8 schematically illustrates a system 500 for creating a digital watermark in accordance with the methods disclosed herein. The system includes a controller 510 in communication with an external memory source 520. The controller 510 may be part of or integrated with a playback device 530 having one or more image transmission devices 550. For example, the controller 510 may be a separate physical unit that at least partially controls the playback device 530. The playback device may be a projector, a television, a computer, a smart device, or any other device capable of displaying a series of visual images. The controller 510 is configured to receive an original image (video) signal having a sequence of original visual images and a watermark image signal having a watermark image from an external memory source 520. The controller 510 performs a method of creating a digital watermark (e.g., methods 100 and 200) and outputs the embedded image signal to the image transmission device 550 of the playback device. For example, the controller 510 may adjust the strength of the watermark in proportion to a predetermined number of quantization steps (e.g., PQ steps). The controller 510 may then embed the adjusted watermark signal in the original image (video) signal and output the embedded image signal to the image transmission device 550. The embedded image signal may then be projected onto the screen 540 by the image transfer device 550.

Fig. 9 schematically illustrates a system 600 for creating a digital watermark in accordance with the methods disclosed herein. The system includes a controller 610 in communication with an external memory source 620 and a playback device 630. The playback device may be a projector, a television, a computer, a DVD player, a smart device, or any other device capable of displaying a series of visual images. The playback device may have one or more image transmission devices 650. The controller 610 is configured to receive an original image (video) signal having a sequence of original visual images and a watermark image signal having a watermark image from an external memory source 620. The controller 610 performs methods of creating digital watermarks (e.g., methods 100 and 200) to create an adjusted watermark signal. For example, the controller 610 may adjust the strength of the watermark in proportion to a predetermined number of quantization steps (e.g., PQ steps). The controller 610 may then output the original image (video) signal and the adjusted watermark image signal to the playback device 630. In the illustrated embodiment, the controller 610 may output the original image (video) signal and the adjusted watermark image signal as two separate signals. The playback device 630 may then project the original image (video) signal and the adjusted watermark image signal onto the screen 640. For example, the playback device 630 may project an original image (video) signal onto the screen 640 using a first image transmission device 650a (e.g., a first projector). The playback device 630 may project the adjusted watermark image signal onto the screen 640 using a second image transmission device 650b (e.g., a second projector). In this embodiment, the projected original image series and the projected watermark image may overlap on the screen 640.

The above-described systems and methods may be provided for creating and adjusting watermarks. Systems, methods, and devices according to the present disclosure may take any one or more of the following configurations.

(1) A digital watermarking method comprising receiving, by an electronic processor, an original image signal comprising a series of original visual images, wherein the original image signal encoded using a Perceptual Quantizer (PQ) luminance level coding transfer function produces PQ luminance steps within the original image signal, and wherein the PQ luminance steps have a magnitude that varies across a luminance range. The method further includes receiving, by the electronic processor, a watermark image signal including the watermark and adjusting the strength of the watermark by at least a first weighting factor, the first weighting factor being a predetermined first number of PQ brightness steps.

(2) The method of (1), wherein adjusting the strength of the watermark signal comprises adjusting the amplitude of the watermark signal.

(3) The method of (1 or 2), wherein the luminance range spans at least 0.001 nit to 1000 nit.

(4) The method of any one of (1) to (3), wherein the first weighting factor is a predetermined first number of PQ luminance steps over a luminance range of at least 0.01 nit to 50 nit.

(5) A method as recited in any one of (1 to 4), wherein the first weighting factor comprises a static force weighting factor, the method further comprising adjusting the strength of the watermark by a second weighting factor, the second weighting factor comprising a motion force weighting factor.

(6) The method of (5), wherein the second weighting factor is proportional to the second number of PQ brightness steps.

(7) The method of (6), wherein the second number of PQ brightness steps is equal to a change in PQ steps between a first image in the series of original visual images and a second image in the series of original visual images.

(8) The method of any one of (1 to 7), further comprising embedding the adjusted watermark signal in the original image signal to create an embedded image signal, the embedded image signal comprising a series of marked visual images.

(9) The method of (8), further comprising transmitting the embedded image signal to a playback device.

(10) The method of any one of (1 to 9), further comprising transmitting the adjusted watermark signal onto a screen.

(11) The method of any one of (1 to 10), further comprising modulating the adjusted watermark image by repeatedly changing the polarity of the adjusted watermark image.

(12) A non-transitory computer-readable storage medium having stored thereon computer-executable instructions for performing the method of (1) with one or more processors.

(13) A system for creating a digital watermark, wherein the system comprises a memory and a controller coupled to the memory and comprising a processor configured to: receiving an original image signal comprising a series of original visual images, the original image signal encoded using a Perceptual Quantizer (PQ) luminance level encoding transfer function to produce PQ luminance steps that vary across a luminance range; receiving a watermark image signal comprising a watermark; and adjusting the strength of the watermark by at least one weighting factor, the at least one weighting factor being a predetermined first number of PQ brightness steps.

(14) The system of (13), wherein the luminance range spans at least 0.001 nit to 1000 nit.

(15) The system of (13 or 14), wherein the first weighting factor comprises a static force weighting factor, and wherein the processor is further configured to adjust the strength of the watermark signal by a second weighting factor comprising a kinetic weighting factor proportional to the second number of PQ brightness steps.

(16) The system of (15), wherein the second number of PQ brightness steps is equal to a change in quantization steps between a first image in the series of original visual images and a second image in the series of original visual images.

(17) The system of any one of (13 to 16), wherein the controller is further configured to embed the adjusted watermark signal in the original image signal to create an embedded image signal comprising a series of marked visual images.

Various features and advantages of the embodiments described herein are set forth in the following claims.

Claims (19)

1. A digital watermarking method, comprising:

receiving, by an electronic processor, an original video signal comprising a series of original visual images, the original video signal encoded using a Perceptual Quantizer (PQ) luma level coding transfer function to generate PQ luma steps within the original video signal, the PQ luma steps having a size that varies across a luma range;

receiving, by the electronic processor, a watermark image signal comprising a watermark; and

the strength of the watermark is adjusted by adjusting the amplitude of the watermark image signal by at least a first weighting factor, which is a predetermined first number of PQ brightness steps, such that the watermark signal is not visible to the human eye but is recoverable by the image capturing device.

2. The method of claim 1, wherein the intensity of the watermark is adjusted by adjusting the amplitude of the watermark image signal by at least a first weighting factor that is a predetermined first number of PQ brightness steps such that the intensity of the watermark is proportional to the PQ brightness level transfer function over the brightness range.

3. The method of claim 1 or claim 2, wherein the brightness range spans at least 0.001 nit to 1000 nit.

4. A method according to any one of claims 1 to 3, wherein the first weighting factor is the predetermined first number of PQ luminance steps over a luminance range of at least 0.01 nit to 50 nit.

5. A method as defined in any one of claims 1 to 4, wherein the first weighting factor comprises a static force weighting factor, the method further comprising adjusting the strength of the watermark by a second weighting factor, the second weighting factor comprising a motion force weighting factor.

6. The method of claim 5, wherein the second weighting factor is proportional to a second number of PQ brightness steps.

7. The method of claim 6, wherein the second number of PQ brightness steps is equal to a change in PQ steps between a first image in the series of original visual images and a second image in the series of original visual images.

8. The method of any of claims 1 to 7, further comprising embedding an adjusted watermark signal in the original video signal to create an embedded video signal, the embedded video signal comprising a series of marked visual images.

9. The method of claim 8, further comprising transmitting the embedded image signal to a playback device.

10. The method of any of claims 1 to 9, further comprising transmitting the adjusted watermark signal onto a screen.

11. The method of any of claims 1 to 10, further comprising modulating the adjusted watermark image by repeatedly changing the polarity of the adjusted watermark image.

12. A non-transitory computer-readable storage medium having stored thereon computer-executable instructions for performing the method of any one of claims 1 to 11 with one or more processors.

13. A system for creating a digital watermark, the system comprising:

a memory; and

a controller coupled to the memory and comprising a processor configured to:

Receiving an original video signal comprising a series of original visual images, the original video signal being encoded using a Perceptual Quantizer (PQ) luminance level encoding transfer function to produce varying PQ luminance steps within the original video signal, the PQ luminance steps having a magnitude that varies across a luminance range,

receiving a watermark image signal comprising a watermark

The strength of the watermark is adjusted by adjusting the amplitude of the watermark image signal by at least a first weighting factor, which is a predetermined first number of PQ brightness steps, such that the watermark is invisible to the human eye but recoverable by the image capturing device.

14. The system of claim 13, wherein the controller is configured to: the intensity of the watermark is adjusted by adjusting the amplitude of the watermark image signal by at least a first weighting factor that is a predetermined first number of PQ luminance steps such that the intensity of the watermark is proportional to the PQ luminance level transfer function over the luminance range.

15. The system of claim 13 or claim 14, wherein the brightness range spans at least 0.001 nit to 1000 nit.

16. A system as defined in any one of claims 13 to 15, wherein the first weighting factor comprises a static force weighting factor, and wherein the processor is further configured to adjust the strength of the watermark signal by a second weighting factor comprising a motion force weighting factor.

17. The system of claim 16, wherein the second weighting factor is proportional to a second number of PQ brightness steps.

18. The system of claim 17, wherein the second number of PQ brightness steps is equal to a change in quantization steps between a first image in the series of original visual images and a second image in the series of original visual images.

19. The system of any of claims 13 to 18, wherein the controller is further configured to embed the adjusted watermark signal in the original video signal to create an embedded video signal comprising a series of marked visual images.