Method and apparatus for watermark signal detection based on detection area limitation

Abstract

The present invention relates to a method and apparatus for detecting a watermark embedded in digital media, and more specifically, to a method for improving detection accuracy by selectively excluding a specific area of ​​digital media when detecting a watermark, and to an apparatus for executing such a method. A method for detecting a watermark embedded in digital media according to one embodiment of the present invention may include the steps of receiving a digital media having a watermark embedded therein, designating a watermark detection target area in the input digital media, identifying a first area unsuitable for watermark detection within the watermark detection target area, and detecting a watermark from information of a second area excluding the identified first area.

Description

Technology Field

[0001] The present invention relates to a method and apparatus for detecting a watermark embedded in digital media, and more specifically, to a method for improving detection accuracy by selectively excluding a specific area of ​​digital media when detecting a watermark, and to an apparatus for executing such a method. Background Technology

[0002] Recently, almost all tasks and services in daily life, such as office work, payment services, certificate issuance, banking, e-commerce, and the provision of cultural media, are being carried out through online and computerized environments. In such online and computerized environments, digital media serves as the subject of information creation, viewing, storage, editing, and transmission. In this context, digital media can refer to various forms of digital information, including images, text, and multimedia, such as digital documents, images, audio, video, and three-dimensional data.

[0003] However, due to the nature of digital information, such digital media can be easily copied or transmitted without loss of information, making them highly vulnerable to leakage through illegal activities. Therefore, to provide a healthy digital media usage environment, protective measures—namely, security systems—that can prevent digital media from being compromised by illegal activities are essential. For example, digital media can be protected through encryption and decryption systems, watermark or fingerprinting systems, and screen capture prevention systems.

[0004] One such method is digital watermarking. Many commonly used digital watermarking methods involve a process of embedding specific additional information by applying a machine-readable manipulation at a temporally or spatially defined location on digital media. The embedded additional information can be detected by analyzing the watermarked digital media using a device or software configured to interpret the manipulation.

[0005] The above additional information may be information indicating ownership of the digital media, or information indicating that copying or viewing is restricted. In particular, such additional information may be embedded in the digital media in a form that is effectively invisible by applying operations that are imperceptible to the viewer of the digital media but readable by a machine.

[0006] Furthermore, recently, forensic watermarking methods have been used to insert traceability information as supplementary data into leaked videos using digital watermarking. Distributing digital media with recipient information embedded within it can help identify individuals who use the media in unauthorized ways or participate in the illegal leakage of digital media.

[0007] One of the leading methods for inserting digital watermarks has conventionally utilized the frequency domain. This prior art involves applying a transform function, such as the Fourier transform or cosine transform, to convert digital media in the form of general information into frequency information, inserting a watermark into the resulting frequency information, and then restoring it to its original form by performing an inverse transform.

[0008] Another method among the promising methods for inserting digital watermarks is the conventional method of using pattern overlays. This prior art includes a method of forming a digital pattern according to a predetermined rule and / or encryption method, and further overlaying the original digital media in an invisible form as described above by applying a predetermined transparency to the pattern and using various mathematical techniques.

[0009] Although the fundamental principles of the aforementioned prior art are widely known, deriving specific implementation methods remains a constant challenge, due to the requirement that sufficient robustness must be guaranteed. It cannot be asserted that digital media is distributed only in the form it was in at the time the watermark was inserted, and digital media may be altered or damaged due to technical reasons or intentional manipulation during the distribution process and / or illegal leakage. In particular, acts of damaging or distorting information to evade tracking by the aforementioned forensic watermark—so-called "attacks on watermarks"—can occur at any time.

[0010] In particular, when digital media is re-recorded through screen capturing or screen recording, various forms of distortion may occur, such as the generation of moiré patterns due to pattern overlap, the cropping of parts of the digital media, or conversely, the inclusion of background information unrelated to the digital media. Therefore, appropriate technical application methods are required to insert the aforementioned frequency-based and / or pattern overlay-based watermarks as a robust method against such attacks, while simultaneously avoiding manipulations that would affect the usability of the digital media.

[0011] Conventionally, watermark detection was performed on the entire captured data. However, this method had the problem of degrading accuracy and unnecessarily increasing detection complexity by processing the entire area, including regions where moiré patterns occurred and areas outside the digital media that did not contain watermarks. Furthermore, even when severe distortion occurred in specific areas of the digital media, processing those areas as they were resulted in a decline in overall detection performance. Moreover, processing the entire area of ​​the digital media was inefficient in terms of processing time and computing resources. Particularly when real-time watermark processing is required, this inefficiency has posed a significant constraint on practical implementation.

[0012] Therefore, a new detection method is required to improve the accuracy of watermark detection and enhance processing efficiency. The problem to be solved

[0013] The present invention aims to overcome the limitations of the aforementioned prior art and, specifically, to provide a method and apparatus that improve detection performance by selectively excluding areas that may impede the accuracy of watermark detection in digital media.

[0014] One objective of the present invention is to provide a method and apparatus for automatically identifying an area of ​​severe distortion containing a moiré pattern and excluding it from watermark detection targets. Another objective of the present invention is to provide a method and apparatus for identifying a content area where a watermark is inserted and excluding an external area that does not contain a watermark from detection targets. Through these objectives, the present invention aims to provide a method and apparatus that improves overall watermark detection performance by excluding the relevant areas from detection targets. means of solving the problem

[0015] A method for detecting a watermark inserted in digital media according to an embodiment of the present invention for solving the aforementioned technical problem may include the steps of receiving a digital media having a watermark inserted therein, specifying a watermark detection target area in the input digital media, identifying a first area unsuitable for watermark detection within the watermark detection target area, and detecting a watermark from information in a second area excluding the identified first area.

[0016] The step of specifying the watermark detection target area may be characterized by identifying the area of ​​content with a watermark inserted in the input digital media and specifying the identified area of ​​content as the watermark detection target area.

[0017] The first area may be characterized by including at least one of an area where a moiré pattern occurs, an area where distortion exceeding a predetermined threshold occurs, and an area outside the content where a watermark is inserted in the input digital media.

[0018] The above method further includes a step of identifying the region where the moiré pattern occurs, and the step of identifying the region where the moiré pattern occurs may be characterized by being performed by at least one of the detection of a regular interference pattern through frequency spectrum analysis, the detection of a directional pattern using a Gabor filter, and the detection of a repeating pattern through edge density analysis.

[0019] The above method further includes a step of identifying an area in which distortion exceeding a predetermined threshold occurs, and the step of identifying an area in which distortion exceeding a predetermined threshold occurs may be characterized by being performed by at least one method among entropy analysis, edge quality analysis, and noise characteristic analysis.

[0020] The step of detecting the watermark may include the step of obtaining frequency domain data by performing frequency conversion on the information of the second zone, and the step of detecting a watermark pattern from the frequency domain data.

[0021] The above method may further include a step of correcting geometric distortion for watermark detection, and the step of correcting geometric distortion for watermark detection may be characterized by being performed on at least one of the watermark detection target area and the second area.

[0022] The correction of the geometric distortion described above may be characterized by including at least one of rotation correction, scaling correction, and translation correction.

[0023] A device for detecting a watermark inserted in digital media according to an embodiment of the present invention for solving the aforementioned technical problem may include a processor, a memory connected to the processor, a media input unit for receiving a digital media with a watermark inserted therein, a region designation unit for designating a watermark detection target area in the input digital media, a region analysis unit for identifying a first area unsuitable for watermark detection within the watermark detection target area, and a watermark detection unit for detecting a watermark from information of a second area excluding the identified first area.

[0024] The area designation unit may be characterized by identifying an area of ​​content with a watermark inserted in the input digital media and designating the identified area of ​​content as the watermark detection target area.

[0025] The above-mentioned area analysis unit may be characterized by being configured to identify at least one of the following as the first area: an area where a moiré pattern occurs, an area where distortion exceeding a predetermined threshold occurs, and an area outside the content where a watermark is inserted in the input digital media.

[0026] The above-mentioned region analysis unit includes a moiré pattern identification unit, and the moiré pattern identification unit may be characterized by operating by at least one method among the detection of a regular interference pattern through frequency spectrum analysis, the detection of a directional pattern using a Gabor filter, and the detection of a repeating pattern through edge density analysis.

[0027] The above region analysis unit includes a distortion region identification unit, and the distortion region identification unit may be characterized by operating by at least one method among entropy analysis, edge quality analysis, and noise characteristic analysis to identify a region in which distortion exceeding a predetermined threshold has occurred.

[0028] The watermark detection unit may include a frequency conversion unit that performs frequency conversion on the information of the second zone to obtain frequency domain data, and a watermark pattern detection unit that detects a watermark pattern from the frequency domain data.

[0029] The above device may further include a geometric distortion correction unit, wherein the geometric distortion correction unit is configured to correct geometric distortion for watermark detection and may be characterized by applying distortion correction to at least one of the watermark detection target area and the second zone.

[0030] The correction of the geometric distortion described above may be characterized by including at least one of rotation correction, size change correction, and translation correction.

[0031] The above device may further include a digital media acquisition means, and the media input unit may be configured to receive digital media from the digital media acquisition means.

[0032] The above device may further include a user interface, and the user interface may be characterized by being configured to provide input or output by being connected to at least one of the watermark detection device and the functional part of the device.

[0033] A system for detecting a watermark inserted in digital media according to an embodiment of the present invention for solving the aforementioned technical problem comprises a media acquisition device for acquiring digital media having a watermark inserted therein, and a watermark detection device, wherein the watermark detection device comprises a media input unit for receiving the digital media from the media acquisition device, a region designation unit for designating a watermark detection target area in the input digital media, a region analysis unit for identifying a first area unsuitable for watermark detection within the watermark detection target area, and a watermark detection unit for detecting a watermark from information of a second area excluding the identified first area.

[0034] In the above system, the media acquisition device is configured to acquire digital media in real time, and the watermark detection device is configured to detect a watermark in real time from the digital media acquired in real time. Effects of the invention

[0035] According to the present invention, the computational efficiency of watermark detection is improved through selective region processing, thereby providing a method and apparatus that can operate effectively even in environments requiring high-resolution images or real-time processing. Brief explanation of the drawing

[0036] FIG. 1 is a conceptual diagram of a frequency domain-based watermark insertion method according to an embodiment of the present invention. FIG. 2 is a conceptual diagram of a watermark insertion method using a pattern overlay method according to an embodiment of the present invention. FIG. 3 is a conceptual diagram of a watermark detection method according to an embodiment of the present invention, FIG. 4 is a conceptual diagram of the designation of a detection target area according to an embodiment of the present invention, FIG. 5 is a flowchart illustrating a document area detection method according to an embodiment of the present invention, FIG. 6 is a conceptual diagram of a limited designation of a watermark detection area according to an embodiment of the present invention, FIG. 7 is a flowchart illustrating a moiré pattern detection method according to an embodiment of the present invention, and FIG. 8 is a block diagram of a watermark detection device using selective area processing according to one embodiment of the present invention. Specific details for implementing the invention

[0037] The present invention is capable of various modifications and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the invention to specific embodiments, and it should be understood that the invention includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention.

[0038] Terms such as "first," "second," etc., may be used to describe various components, but said components should not be limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component. The term "and / or" includes a combination of multiple related described items or any of the multiple related described items, and is non-exclusive unless otherwise indicated. Where items are listed in this specification, they are merely illustrative descriptions intended to facilitate the explanation of the inventive spirit and possible embodiments set forth in this specification, and are therefore not intended to limit the scope of the embodiments of the present invention.

[0039] In this specification, "A or B" may mean "only A," "only B," or "both A and B." Alternatively, in this specification, "A or B" may be interpreted as "A and / or B." For example, in this specification, "A, B or C" may mean "only A," "only B," "only C," or "any combination of A, B and C."

[0040] A slash ( / ) or a comma used in this specification may mean "and / or." For example, "A / B" may mean "A and / or B." Accordingly, "A / B" may mean "only A," "only B," or "both A and B." For example, "A, B, C" may mean "A, B or C."

[0041] In this specification, "at least one of A and B" may mean "only A," "only B," or "both A and B." Additionally, in this specification, the expressions "at least one of A or B" or "at least one of A and / or B" may be interpreted as synonymous with "at least one of A and B."

[0042] Additionally, in this specification, "at least one of A, B and C" may mean "only A," "only B," "only C," or "any combination of A, B and C." Also, "at least one of A, B or C" or "at least one of A, B and / or C" may mean "at least one of A, B and C."

[0043] When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. On the other hand, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between.

[0044] The terms used herein are merely for describing specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as “comprising” or “having” are intended to indicate the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0045] Unless otherwise defined, all terms used herein, including technical or scientific terms, are used with the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this specification.

[0046] In describing the invention in this specification, embodiments may be described or illustrated in terms of the described functions or unit blocks that perform the functions. The blocks may be expressed in this specification as one or more devices, units, modules, parts, etc. The blocks may be implemented in hardware by a method of implementing one or more logic gates, integrated circuits, processors, controllers, memory, electronic components, or information processing hardware, which are not limited thereto. Alternatively, the blocks may be implemented in software by a method of implementing application software, operating system software, firmware, or information processing software, which are not limited thereto. A single block may be implemented by being separated into multiple blocks that perform the same function, or conversely, a single block may be implemented to simultaneously perform the functions of multiple blocks. The blocks may also be implemented by being physically separated or combined according to any criteria. The blocks may be implemented to operate in an environment where their physical locations are not specified and they are spaced apart from each other by a communication network, the Internet, a cloud service, or a communication method not limited thereto. Since all of the above-mentioned methods of implementation fall within the scope of various embodiments that a person skilled in the art familiar with the field of information and communication technology can adopt to realize the same technical concept, any detailed methods of implementation should be interpreted as being included within the scope of the technical concept of the invention as set forth in this specification.

[0047] Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the attached drawings. In describing the present invention, to facilitate overall understanding, the same reference numerals are used for identical components in the drawings, and redundant descriptions of identical components are omitted. Furthermore, it is assumed that multiple embodiments are not mutually exclusive and that some embodiments may be combined with one or more other embodiments to form new embodiments.

[0049] Insertion of watermarks

[0050] In the embodiments of the present invention, the method of inserting a watermark into digital media is not limited. However, according to a representative embodiment, a method of inserting a watermark in the frequency domain may be applied, and according to another embodiment, a method of inserting a watermark using a pattern overlay method may be applied. Furthermore, the type of digital media is also not limited within the scope to which the technical concept of the present invention can be applied. For ease of explanation, the following description is based on a representative embodiment in which the digital media is a two-dimensional bitmap still image. However, it should be understood that the present invention can be applied without limitation within the scope of the technical concept of the present invention to other digital information that may be considered as two-dimensional vector images, two-dimensional video, panoramic or binocular field of view images and / or video, various types of three-dimensional images and / or video, digitized voice data, digitized text and / or document data, and other digital information.

[0051] FIG. 1 is a conceptual diagram of a frequency domain-based watermark insertion method according to an embodiment of the present invention. Referring to FIG. 1, the frequency domain-based watermark insertion method according to an embodiment of the present invention may be composed of two processing paths that respectively receive and process an input digital media (111) and watermark information (121).

[0052] The above input digital media (111) may refer to any digital media data to be inserted with a watermark. The above input digital media (111) may be transformed into the frequency domain in a frequency conversion unit (112) to generate frequency domain data (113). According to one embodiment of the present invention, the frequency conversion unit (112) may be implemented by at least one transformation method among Fourier transform, cosine transform, sine transform, wavelet transform, and Hadamard transform.

[0053] The above watermark information (121) may refer to content information to be watermarked on the input digital media (111). The above watermark information (121) may be converted into a frequency domain by a frequency converter (122) to generate frequency domain data (123). According to one embodiment of the present invention, the frequency converter (122) may select a conversion method within the range of a conversion method identical or similar to the frequency converter (112) for the input digital media (111) described above, and in a preferred embodiment, the same conversion method may be applied. However, according to another embodiment of the present invention, the above watermark information (121) may be configured to be generated as frequency domain data (123) from the beginning so as not to require a separate frequency converter (122).

[0054] As described above, when two frequency domain data (113, 123) are prepared, the frequency domain merging unit (131) may be configured to combine the two frequency domain data (113, 123) to generate merged frequency data (132). At this time, the method of combining the frequency data for inserting a watermark can be configured in various ways. For example, the frequency domain data may be combined by simple signal processing such as addition, difference, multiplication, or position-specific maximum / minimum value acquisition operations. As another example, the frequency domain data may be combined by modulating the amplitude of each frequency component or by modulating phase information. It should be understood that various methods of combining multiple frequency domain data may be applied without limitation within the scope that does not infringe upon the technical concept of the present invention.

[0055] Finally, the inverse conversion unit (133) may be configured to inversely convert the merged frequency data (132) to generate watermarked digital media (140). It is preferable that the inverse conversion unit (133) adopt an inverse conversion method based on the same method or having a corresponding structure as the frequency conversion unit (112) for the input digital media (111), and more preferably, it may be configured to have a structure based on the same method or having a corresponding structure as the frequency conversion unit (122) for the watermark information (121). The watermarked digital media (140) generated through this is considered to be subsequently distributed, transmitted, and / or circulated by any method.

[0056] FIG. 2 is a conceptual diagram of a watermark insertion method using a pattern overlay method according to an embodiment of the present invention. Referring to FIG. 2, the watermark insertion method using a pattern overlay method according to an embodiment of the present invention may mean a method of inserting a watermark by overlaying a watermark pattern (223) in a spatial area onto an input digital media (211).

[0057] According to one embodiment, watermark information (221) to be inserted may be converted into a watermark pattern (223) by a pattern generation unit (222). At this time, the pattern generation unit (222) may be configured to generate a regular geometric shape or a pseudo-random pattern, and the specific method thereof is not limited. A person skilled in the art will understand that various techniques for deriving the shape of a patterned watermark from any watermark information have been provided in the past or may be provided in the future, and that all of them may be applied within the scope of the technical concept of the present invention. Furthermore, according to one embodiment of the present invention, an encryption technique may be applied in the process of generating the pattern.

[0058] Next, the transparency adjustment unit (224) may be configured to adjust the visibility of the generated watermark pattern (223) to generate an adjusted watermark pattern (225). The transparency may be applied uniformly to the entire watermark pattern (223) or applied differently to each part of the pattern (223). The watermark pattern (223) with adjusted transparency may be output as an adjusted watermark pattern (225).

[0059] The pattern overlay unit (230) may be configured to generate a watermarked digital media (240) by overlaying the adjusted watermark pattern (225) onto the input digital media (211). According to one embodiment of the present invention, the overlay may be performed by an alpha blending method based on the transparency of the adjusted watermark pattern (225). According to another embodiment of the present invention, various blending modes such as multiply, screen, and overlay may be used for the overlay in addition to the alpha blending described above. According to yet another embodiment of the present invention, the transparency of the watermark pattern (223) may be determined afterward during the overlay process, particularly based on the characteristics of the input digital media (211) to be inserted and the characteristics of the insertion location, and in this case, the transparency adjustment unit (224) and the adjusted watermark pattern (225) generated by it may be omitted or replaced.

[0061] Detection of watermarks

[0062] FIG. 3 is a conceptual diagram of a watermark detection method according to an embodiment of the present invention. Referring to FIG. 3, a watermark detection method according to an embodiment of the present invention may include the step of acquiring an input image (310), the step of specifying a target area in the input image where a watermark is expected to be inserted (320), and the step of performing watermark detection on the specified target area (330).

[0063] First, an input image may be acquired (S310). According to a preferred embodiment of the present invention, the input image may be acquired by taking a picture with a camera of a detection device. In this case, the detection device may refer to a portable terminal, such as a smartphone. The portable terminal is equipped with a camera and can take a picture of a digital media with a watermark inserted through the camera. Here, the digital media with a watermark inserted may refer to a digital image displayed on the screen of an information processing device other than the portable terminal, or a printed medium on which the digital image is printed. For example, the input image may be acquired by taking a picture of an image displayed on the screen of a computer or tablet, or by taking a picture of an image printed on a printed medium such as paper or a photograph.

[0064] Next, a target area (320) in which a watermark is expected to be inserted in the input image may be designated. The designation of the target area may be performed in various ways. For example, the target area may always be considered as the entire area of ​​the acquired image, and a method of extracting the watermark included in the image through consistent computational processing of the entire image may be applied. As another example, in the image acquisition step (310), a specific area in which the image with the inserted watermark is displayed may be identified. Specifically, for instance, when photographing the screen of a computer monitor, the outline of the monitor may be identified, or when photographing a print medium, the outline of the print medium may be identified, and the location of the target area may be designated (320) based on these outlines.

[0065] According to one embodiment of the present invention, prior information regarding the area of ​​the target content into which a watermark is inserted can be utilized to designate the target area. For example, in the case of a document, if information is provided in advance that a watermark is inserted at the top of the document, and the area of ​​the document is identified by the outline, the area corresponding to the top can be designated as the target area. As another example, in the case of an image, if information is provided in advance that a watermark is inserted at the center of the image, and the area of ​​the image is identified by the outline, the area corresponding to the center can be designated as the target area.

[0066] According to another embodiment of the present invention, prior information regarding the manner in which a watermark is expressed may be utilized in designating the target area. For example, if the watermark is inserted using a pattern overlay method, the target area may be designated by identifying an area containing the pattern through image processing. As another example, if the watermark is inserted in the frequency domain, the target area may be designated by searching for an area that matches a previously known watermark pattern.

[0067] The designation method by the above image processing may include, for example, a method for identifying an area where a watermark is expected to be inserted based on image characteristics such as brightness, color, or texture. In particular, according to a preferred embodiment of the present invention, by applying image processing such as edge detection or geometric pattern detection to the acquired input image, an area having characteristics similar to the pattern used as a watermark can be identified and designated as the target area.

[0068] Finally, watermark detection (330) can be performed on the designated target area. Watermark detection can be performed in various ways depending on the insertion method. For example, in the case of a watermark inserted using a pattern overlay method, the watermark can be detected by performing pattern matching on the target area. As another example, in the case of a watermark inserted in the frequency domain, the watermark pattern in the frequency domain can be detected after performing frequency conversion on the target area.

[0069] In one embodiment of the present invention, the watermark detection (330) step may include the following detailed steps. First, the geometric distortion of the target area may be corrected (331). The geometric distortion may refer to rotation, scaling, and translation of the image that may occur during the process of acquiring the input image (the above-described type of geometric distortion may be collectively referred to as "RST distortion"). For example, when the input image is captured by the camera of the detection equipment, various geometric distortions may occur depending on the shooting angle or distance. Since such geometric distortions can be a factor that makes watermark detection difficult, the geometric distortions may be corrected prior to watermark detection.

[0070] The correction of the geometric distortion (331) can be performed in various ways. For example, the geometric distortion can be estimated and corrected using the morphological features of the watermark included in the target area. As a specific example, if the watermark includes a specific regular pattern (e.g., a grid or concentric circles), the degree of geometric distortion can be estimated and corrected from the distorted form of the pattern. As another example, if the watermark includes a plurality of markers having a specific positional relationship, the degree of geometric distortion can be estimated and corrected from the change in the relative positional relationship of the markers.

[0071] Next, the presence of a watermark pattern can be detected (332) for the target area where the geometric distortion has been corrected. The detection of the presence of the watermark pattern can be performed in various ways depending on the method of inserting the watermark. For example, in the case of a watermark inserted using a pattern overlay method, the presence of the watermark can be determined by calculating the correlation between the target area and the watermark pattern. As another example, in the case of a watermark inserted in the frequency domain, the presence of the watermark pattern can be determined in the frequency domain after performing a frequency transformation on the target area.

[0072] When a watermark pattern is detected, a watermark message can be extracted (333) from the watermark pattern. The extraction of the watermark message can be performed in various ways depending on the method of inserting the watermark. For example, if the watermark is inserted in the frequency domain, the message can be extracted by performing frequency conversion on the target domain and then analyzing the frequency component into which the watermark message is inserted. As another example, if the watermark is inserted using a pattern overlay method, the message can be extracted by analyzing features such as the shape or location of the pattern.

[0073] According to one embodiment of the present invention, the extraction (333) of the watermark message may include an error correction process. If an error correction code is inserted together during the watermark insertion process, the error correction code can be used to detect and correct errors in the extracted message. This minimizes errors in message extraction that may occur due to image distortion or noise.

[0074] According to one embodiment of the present invention, the watermark detection step (330) may be performed in parallel for a plurality of detection target areas. For example, the target area may be divided into a plurality of detection target areas, and watermark detection may be performed independently for each area. Through such parallel processing, the processing speed of watermark detection can be improved, and even if watermark detection fails in some areas, the overall accuracy of watermark detection can be improved through the detection results from other areas.

[0075] According to one embodiment of the present invention, the watermark detection step (S330) may be performed selectively depending on the characteristics of the area where the watermark is expected to be inserted. For example, areas where a moiré pattern occurs or areas where severe distortion occurs may be excluded from the detection target. Through such selective detection, the accuracy of detection can be improved and the processing time can be shortened.

[0076] According to one embodiment of the present invention, the watermark detection step (S330) may include a process for evaluating the reliability of the detected watermark. The evaluation of reliability may be performed by considering various factors. For example, reliability may be evaluated by considering the degree of matching with the watermark pattern, the error correction result of the detected message, or the quality of the detection area. If the evaluated reliability is below a specific threshold, the detection result may be invalidated or an additional verification process may be performed.

[0077] According to one embodiment of the present invention, the watermark detection can be performed in real time. For example, watermark detection can be performed continuously on a real-time video stream input through a camera of any watermark detection equipment. In this case, the accuracy of detection can be improved by utilizing the temporal correlation between consecutive frames. For example, the detection area in the next frame can be predicted using location information of a watermark detected in a specific frame, and more reliable detection results can be derived by synthesizing the results detected in consecutive frames.

[0079] Designation of watermark detection target areas and limited detection

[0080] FIG. 4 is a conceptual diagram of a detection target area designation according to an embodiment of the present invention. Referring to FIG. 4, a watermark detection device (410) may be configured to designate an area within an input image (415) where a watermark is expected to exist as a detection target area (420), and not perform a watermark detection operation on other areas. For example, when a document (430) with a watermark inserted is photographed together with a background (440), a part of the input image (415) containing the document (430) is designated as a detection target area (420), and watermark detection is not performed on the background (440), thereby preventing meaningless operations and improving overall detection efficiency.

[0081] According to an embodiment, the detection target area (420) may be configured to include at least the entire watermark extraction target, e.g., document (430), as shown in FIG. 4, and may be designated as a rectangular area, e.g., a shape that can be easily derived from the scan order and / or encoding form of the input image (415) to facilitate information processing. However, according to another embodiment, the detection target area (420) may be derived by the result of precisely detecting the document (430), etc., and may be derived in a shape that is identical to or significantly close to the outline that the document (430) actually occupies in the input image (415). In this case, along with the designation of the detection target area (420), correction of geometric distortion of the document (430), e.g., RST correction, may be performed simultaneously.

[0082] According to one embodiment of the present invention, the designation of the detection target area (420) can be performed in various ways. For example, the outline of a document can be detected through edge detection, and the document area can be specified based on this. Specifically, edges within an image can be detected by applying a Sobel filter, a Canny edge detector, or other edge detection algorithm, and among the detected edges, line segments that have high linearity and are orthogonal to each other can be identified as the outline of the document to be detected.

[0083] FIG. 5 is a flowchart illustrating a document area detection method according to an embodiment of the present invention. Referring to FIG. 5, document area detection may include an edge detection step (510), a straight line component extraction step (520), an orthogonal relationship analysis step (530), and a circular restoration step (540). In the edge detection step (510), an edge map may be generated by applying an edge detection algorithm to an input image. In the straight line component extraction step (520), straight line components may be extracted from the edge map by applying a Hough transform, etc. In the orthogonal relationship analysis step (530), the angles between the extracted straight lines may be analyzed to identify pairs of straight lines that are orthogonal to each other. Finally, in the circular restoration step (540), the original shape of the target document may be inferred using the identified orthogonal straight lines, and by restoring the circular shape, geometric distortion of the detection target, such as RST distortion, may be corrected along with the designation of the watermark detection target area, thereby completing the preparation for watermark detection.

[0084] According to another embodiment of the present invention, the designation of the detection target area (420) can be performed using a machine learning model. For example, a Convolutional Neural Network (CNN) can be trained to automatically identify document areas within an input image. This machine learning-based approach has the advantage of reliably detecting document areas even under complex backgrounds or lighting conditions.

[0085] FIG. 6 is a conceptual diagram of a limited designation of a watermark detection area according to an embodiment of the present invention. Referring to FIG. 6, among various zones (621, 622, 623) within a detection target area (620), watermark detection may be selectively performed only on a zone (622) that satisfies specific conditions. For example, a moiré pattern (621) or a zone (623) in which severe distortion occurs may be excluded from the detection target.

[0086] According to one embodiment of the present invention, the detection of the region where the moiré pattern occurs can be performed through frequency domain analysis. FIG. 7 is a flowchart illustrating a moiré pattern detection method according to one embodiment of the present invention. Referring to FIG. 7, moiré pattern detection may include a frequency conversion step (710), a frequency spectrum analysis step (720), a regular interference pattern identification step (730), and a moiré region determination step (740).

[0087] According to one embodiment of the present invention, in the frequency conversion step (710), frequency conversion can be performed on the inspection target area using a conversion method including a Fast Fourier Transform (FFT). In the frequency spectrum analysis step (720), the energy distribution of a specific frequency band in the converted frequency area can be analyzed. In the regular interference pattern identification step (730), a regular pattern appearing on the frequency spectrum can be identified. For example, if a moiré pattern exists, a phenomenon in which peaks are repeated at specific intervals on the frequency spectrum can be observed. Finally, in the moiré area determination step (740), the characteristics of the identified pattern can be analyzed to determine whether the area contains a moiré pattern.

[0088] According to another embodiment of the present invention, the detection of the moiré pattern may also be performed through analysis in a spatial domain. For example, a repeating pattern having a specific directionality can be detected by applying a directional filter such as a Gabor filter. In addition, the presence of a moiré pattern may be determined by analyzing spatial features such as edge density or texture characteristics.

[0089] According to one embodiment of the present invention, the degree of distortion in each region can be evaluated through multidimensional analysis. For example, individual disorder can be measured for at least one region included in an image through entropy analysis. A high entropy value may suggest the presence of severe noise or distortion in the corresponding region. As another example, edge quality can be analyzed. For instance, the intensity and direction of edges can be analyzed by applying a Sobel filter, and the continuity or sharpness of the edges can be evaluated. As yet another example, noise characteristics can be evaluated through multi-scale analysis using wavelet transform. Through this, the distribution and intensity of noise can be measured by analyzing the transform coefficients at each level.

[0090] According to one embodiment of the present invention, each zone may be assigned a different weight depending on the measured degree of distortion. For example, a zone with severe distortion may be assigned a low weight or completely excluded during the watermark detection process. Conversely, a zone with low distortion may be assigned a high weight to have a greater influence on the watermark detection results.

[0091] FIG. 8 is a block diagram of a watermark detection device using selective area processing according to an embodiment of the present invention. Referring to FIG. 8, the watermark detection device (800) may include a media input unit (810), an area analysis unit (820), a watermark detection unit (830), and a result output unit (840). The media input unit (810) may perform the function of capturing or receiving digital media with a watermark inserted therein. The area analysis unit (820) may perform the function of designating a detection target area in the input image and identifying areas with moiré patterns or severe distortion. The watermark detection unit (830) may perform the function of detecting watermarks only for the selected areas. Finally, the result output unit (840) may perform the function of displaying the detected watermark information to a user or transmitting it to another system.

[0092] According to one embodiment of the present invention, the watermark detection device (800) may include a processor (850) and a memory (860) linked to the processor (850). For example, the processor (850) may be utilized as a basis for operating at least one functional unit belonging to the device (800), and may be configured to process an acquired image to execute each step of the watermark detection method described above, and may support functions such as parallel processing or GPU acceleration as needed. For example, the memory (860) may be configured to temporarily or permanently store data provided from at least one functional unit belonging to the device (800) and the processor (850).

[0093] According to another embodiment of the present invention, the watermark detection device (800) may be configured to include a digital media acquisition means (815) as a functional part or to be connected to such a functional part / device, etc. The image acquisition means may include, for example, a camera, a sensor, and other information acquisition means having various input bands. The digital media acquisition means (815) may be connected to the media input part (810). According to an embodiment, the digital media acquisition means (815) may be included in the watermark detection device (800) or configured as an independent external device. If the digital media acquisition means (815) is configured as a device independent of the watermark detection device (800), the digital media acquisition means (815) may be considered as a component of a watermark detection system including the watermark detection device (800).

[0094] According to another embodiment of the present invention, the watermark detection device (800) may display detection results or receive input from a user through a user interface (845). For example, it may display the location of the detected watermark on a screen or decode and display the detected message. Additionally, it may be configured to receive input from the user, such as specifying a detection area or adjusting detection parameters. The user interface (845) may be connected to the result output unit (840) and may also be connected to provide input to any functional unit belonging to the watermark detection device (800). According to an embodiment, the user interface (845) may be included in the watermark detection device (800) or configured as an independent external device. If the user interface (845) is configured as a device independent of the watermark detection device (800), the user interface (845) may be considered as a component of a watermark detection system including the watermark detection device (800).

[0095] According to one embodiment of the present invention, the watermark detection device (800) may be configured to support real-time processing. For example, it may be configured to specify a detection target area and detect a watermark in real time for consecutive video frames input as digital media to the media input unit (810). In this case, detection performance can be improved by utilizing the temporal correlation between consecutive frames. For example, document area detection in the current frame can be performed more efficiently by utilizing location information of a document detected in the previous frame.

[0096] According to another embodiment of the present invention, the device (800), the processor (850), and the memory (860) may be configured to support parallel processing. For example, the processor (850) may be parallelized to simultaneously perform watermark detection for a plurality of detection target areas, thereby improving the overall processing speed. Additionally, the processor (850) may utilize parallel processing hardware, such as a Graphics Processing Unit (GPU), to accelerate computationally intensive tasks such as frequency conversion or pattern matching. For example, referring again to FIG. 6, the detection target area (620) may be divided into a plurality of processing areas (621, 622, 623). In this case, watermark detection for each area may be performed in parallel by independent processing threads. Each processing thread may independently perform tasks such as frequency conversion, pattern matching, and message extraction for the corresponding area. The results of the completed processing may be subsequently aggregated to derive a final watermark detection result.

[0097] According to one embodiment of the present invention, the parallel processing structure can be adaptedly configured. For example, the size or number of processing zones can be dynamically adjusted according to available computing resources or the characteristics of the data to be processed. Additionally, the processing priority of each zone can be set differentially according to the characteristics of the zone. For example, a high processing priority may be assigned to zones with low distortion or zones with a high probability of watermarks existing.

[0098] According to another embodiment of the present invention, the watermark detection system may be linked with an external server via a network. For example, detected watermark information may be transmitted to a server for verification, or watermark pattern information or detection parameters may be received from a server. Through such server linkage, the functionality of the detection system can be expanded and its performance improved.

[0099] The processor (850) described in this specification may mean one or more general-purpose computers or special-purpose computers, such as a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. Even if the processor (850) is expressed in the singular for convenience of understanding, a person of ordinary knowledge in the art will know that the processor (850) may include a plurality of processing elements and / or a plurality of types of processing elements. For example, a device according to one embodiment of the present invention may include a plurality of processors or one processor and one controller as the processor (850). In addition, the processor (850) may be implemented by various processing configurations, such as a parallel processor or a multi-core processor.

[0100] The processor (850) may be configured to execute an operating system (OS) and one or more software programs executed on the operating system. Additionally, the processor (850) may access, store, manipulate, process, and generate data in response to the execution of the software.

[0101] The software may include a computer program, code, instructions, or a combination of one or more of these, and may be configured to control the processor (850) to operate as desired and to issue instructions to the processor (850) independently or collectively. The software may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave in order to be interpreted by the processor (850) or to provide instructions or data to the processor (850). The software may be distributed over a networked computer system and may be stored or executed in a distributed manner.

[0102] The software may also be implemented in the form of program instructions that can be executed through various computer means and may be recorded or stored in the memory (860). The memory (860) may be a computer-readable recording medium, and program instructions, data files, data structures, etc. may be recorded in the computer-readable recording medium alone or in combination. The program instructions stored in the memory (860) may be based on a command system specifically designed and configured for an embodiment of the present invention, or may follow a command system known and available to those skilled in the art of computer software, such as Assembly, C, C++, Java, Python, etc. It should be understood that the command system and the program instructions thereunder include not only machine code such as that produced by a compiler, but also high-level language code that can be executed by the device and / or the processor (850) according to an embodiment of the present invention using an interpreter, etc.

[0103] A computer-readable recording medium constituting an apparatus according to an embodiment of the present invention, including the memory (860) described in this specification, may include a temporary or volatile recording medium that is maintained only while the processor (850) is operating, such as a processor cache, RAM, or flash memory; or may include a relatively non-volatile or long-term recording medium such as a magnetic media such as a hard disk, floppy disk, and magnetic tape; an optical recording medium such as a CD-ROM or DVD; a magneto-optical media such as a floptical disk; or a solid state memory; or may include a read-only recording medium such as a ROM placed on hardware; furthermore, the hardware itself configured to perform a series of program instructions and equivalent operations by means of a hard-wired structure by circuit wiring, and since each step for performing the operation implementing the embodiment of the present invention can be considered to be recorded by the connection and arrangement of the hardware components, the connection and It is obvious to a person skilled in the art that the placement method can be seen as equivalent to the above memory (860).

[0104] The embodiments described above with respect to the processor (850) and the memory (860) are not mutually exclusive and may be selected or combined as needed. For example, a hardware device may be configured to operate as a module composed of one or more of the software to perform the operation of the embodiment of the present invention, and vice versa. As another example, in this specification, all or part of the operation assigned to any function may be implemented by one or more of the software stored in a device according to one embodiment of the present invention (preferably in a recording medium within the category of the memory (860)) and configured to be executed by the processor (850), in which case such function may be implemented as a function "included" in the processor (850).

[0106] Although the present invention has been described above with reference to the drawings and embodiments, as previously stated, the scope of protection of the present invention is not limited by the drawings or embodiments presented above, and those skilled in the art will understand that various modifications and changes can be made to the present invention without departing from the spirit and scope of the invention as described in the claims of the present invention.

Claims

Claim 1 A method for detecting a watermark embedded in digital media, comprising: receiving a digital media having a watermark embedded therein; specifying a watermark detection target area in the input digital media; identifying a first area unsuitable for watermark detection within the watermark detection target area; and detecting a watermark from information in a second area excluding the identified first area. Claim 2 A watermark detection method according to claim 1, wherein the step of designating the watermark detection target area comprises identifying the area of ​​content in which a watermark is inserted in the input digital media and designating the identified area of ​​content as the watermark detection target area. Claim 3 A watermark detection method according to claim 1, wherein the first zone comprises at least one of an area where a moiré pattern occurs, an area where distortion exceeding a predetermined threshold occurs, and an area corresponding to the outside of the content in which the watermark is inserted in the input digital media. Claim 4 A watermark detection method according to claim 3, further comprising the step of identifying the region where the moiré pattern occurs; wherein the step of identifying the region where the moiré pattern occurs is performed by at least one method among the detection of a regular interference pattern through frequency spectrum analysis, the detection of a directional pattern using a Gabor filter, and the detection of a repeating pattern through edge density analysis. Claim 5 A watermark detection method according to claim 3, further comprising the step of identifying an area in which distortion exceeding a predetermined threshold occurs; wherein the step of identifying an area in which distortion exceeding a predetermined threshold occurs is performed by at least one method among entropy analysis, edge quality analysis, and noise characteristic analysis. Claim 6 A watermark detection method according to claim 1, wherein the step of detecting the watermark comprises: a step of obtaining frequency domain data by performing frequency conversion on the information of the second zone; and a step of detecting a watermark pattern from the frequency domain data. Claim 7 A watermark detection method according to claim 1, further comprising a step of correcting geometric distortion for watermark detection; wherein the step of correcting geometric distortion for watermark detection is performed on at least one of the watermark detection target area and the second zone. Claim 8 A watermark detection method according to claim 7, wherein the correction of the geometric distortion comprises at least one of rotation correction, scaling correction, and translation correction. Claim 9 A watermark detection device for detecting a watermark embedded in digital media, comprising: a processor; a memory connected to the processor; a media input unit for receiving a digital media having a watermark embedded therein; a region designation unit for designating a watermark detection target area in the input digital media; a region analysis unit for identifying a first region unsuitable for watermark detection within the watermark detection target area; and a watermark detection unit for detecting a watermark from information of a second region excluding the identified first region. Claim 10 A watermark detection device according to claim 9, wherein the area designation unit identifies an area of ​​content in which a watermark is inserted in the input digital media and designates the identified area of ​​content as the watermark detection target area. Claim 11 A watermark detection device according to claim 9, wherein the area analysis unit is configured to identify at least one of the following as the first area: an area where a moiré pattern occurs, an area where distortion exceeding a predetermined threshold occurs, and an area outside the content in which the watermark is inserted in the input digital media. Claim 12 A watermark detection device according to claim 11, wherein the region analysis unit includes a moiré pattern identification unit, and the moiré pattern identification unit operates by at least one method among detection of a regular interference pattern through frequency spectrum analysis, detection of a directional pattern using a Gabor filter, and detection of a repeating pattern through edge density analysis. Claim 13 A watermark detection device according to claim 11, wherein the region analysis unit includes a distortion region identification unit, and the distortion region identification unit operates by at least one method among entropy analysis, edge quality analysis, and noise characteristic analysis to identify a region in which distortion exceeding a predetermined threshold has occurred. Claim 14 A watermark detection device according to claim 9, wherein the watermark detection unit comprises: a frequency conversion unit that performs frequency conversion on the information of the second zone to obtain frequency domain data; and a watermark pattern detection unit that detects a watermark pattern from the frequency domain data. Claim 15 A watermark detection device according to claim 9, further comprising a geometric distortion correction unit; wherein the geometric distortion correction unit is configured to correct geometric distortion for watermark detection and applies distortion correction to at least one of the watermark detection target area and the second zone. Claim 16 A watermark detection device according to claim 15, wherein the correction of the geometric distortion comprises at least one of rotation correction, size change correction, and translation correction. Claim 17 A watermark detection device according to claim 9, further comprising a digital media acquisition means; wherein the media input unit is configured to receive digital media from the digital media acquisition means. Claim 18 A watermark detection device according to claim 9, further comprising a user interface; wherein the user interface is configured to be connected to at least one of the watermark detection device and a functional part of the device to provide an input or output. Claim 19 A watermark detection system for detecting a watermark embedded in digital media, comprising: a media acquisition device for acquiring digital media having a watermark embedded therein; and a watermark detection device; wherein the watermark detection device comprises: a media input unit for receiving the digital media from the media acquisition device; a region designation unit for designating a watermark detection target area in the input digital media; a region analysis unit for identifying a first area unsuitable for watermark detection within the watermark detection target area; and a watermark detection unit for detecting a watermark from information of a second area excluding the identified first area. Claim 20 A watermark detection system according to claim 19, wherein the media acquisition device is configured to acquire digital media in real time, and the watermark detection device is configured to detect a watermark in real time from the digital media acquired in real time.

Images