KR102137229B1 - Electronic device for extracting fingerprinting signal from image and operating method thereof - Google Patents

Abstract

The present invention relates to an electronic device for extracting a fingerprint signal from an image and a method for operating the same. An electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document includes: an optical unit that acquires image data for an external object through an image sensor in response to a signal for photographing being received, the image In the data, if the shape and size of the first rectangle approximating the border of the external object does not correspond to the shape and size of the preset second rectangle, the perspective of the image data is such that the first rectangle corresponds to the second rectangle Perspective correction unit that adjusts, a watermark signal prediction unit that extracts noise data from the image data using at least one filter, and performs auto-correlation on the image data, and In scanning a plurality of pixels in the image data using a mask window, the pixels having the largest result value of the autocorrelation operation for each region corresponding to the mask window are peak pixels for each of the regions. A peak detector that detects as, linearly connects points corresponding to each of the detected peak pixels to obtain one or more figures, and based on the gradient information, size information, and location information identified from the obtained figures, the external object Compute the gradient difference, size difference, and position difference between the original image data corresponding to and the image data, and correct the slope, size, and position of the image data so that the calculated gradient difference, size difference, and position difference are below a threshold A geometric deformation correction unit to obtain a pre-inserted fingerprint signal from the image data whose tilt, size and position are corrected, and an original message extraction unit to extract the original message, and to transmit the extracted original message to a predetermined server And, it may include a communication unit for receiving the data corresponding to the extracted original message from the server.

Description

ELECTRONIC DEVICE FOR EXTRACTING FINGERPRINTING SIGNAL FROM IMAGE AND OPERATING METHOD THEREOF}

The present invention relates to an electronic device for extracting a fingerprint signal from an image and a method for operating the same.

Digital content, such as images, audio and video, has the advantage of being easy to copy and distribute, so it is convenient for multiple users to share and use, but it has the disadvantage that it is difficult to protect intellectual property rights related to them, such as copyright.

Accordingly, various studies have been conducted to protect intellectual property rights related to digital content, and recently, it has become common to protect intellectual property rights of digital content by applying watermark technology to digital content. Watermark technology refers to a technique of modifying at least a part of bits in digital content so as to identify copyright information of digital content and inserting an intended bit pattern into the digital content. Watermark data such as bit patterns embedded in digital content may be inserted blurring enough to be visually discernible to the user without compromising the original digital content, but in some cases may be inserted so that the user cannot visually discriminate. The former case is classified into perceptible watermarking, and the latter case is classified into imperceptible watermarking.

Meanwhile, the watermark technology may be divided into a rigid watermarking technology, a soft watermarking technology, and a fingerprinting technology depending on the application. Rigid watermarking technology is a technology used to prevent attempts by a specific person to intentionally damage or modify a watermark for the purpose of illegally using the content. To this end, the quality of data is severe when there is an external attempt to tamper with the original watermarking. It is a watermarking technology designed to not break the watermark before it is damaged. Soft watermarking technology is used in cases where the external leakage of files should be prevented, such as the original photographic text of a hospital's clinical photographic images, government documents, or military classified documents. That is, it is a watermarking designed to damage the original content at the same time as the watermark is easily broken when the data is modified. It is used as a method to provide authentication and integrity by inspecting whether the data is modified. Fingerprinting technology is a concept similar to the currently used'bar code', and inserts a unique number or identifier into the content. By applying fingerprinting technology, it is possible to know to whom it is distributed, so it can be used to classify products in the logistics business or to check distribution channels.

Various embodiments disclosed in this document are related to a technique of adding data for the purpose of activation or distribution/logistics tracking on a photo or label using a fingerprinting technique that the user cannot recognize. Such a technique can be regarded as developed in a manner of providing a user with intended information by planting a QR code or barcode on at least a part of a photo or label. For example, QR codes or barcodes, which have been widely used in the past, are easy to transmit information, but have a disadvantage of being vulnerable to security due to easy forgery or copying. The present invention proposes a method for safely and easily transmitting information by integrating a security-invisible fingerprinting technique and a QR code technique capable of transmitting information using an image recognized through a camera.

Various embodiments disclosed in this document propose a method of effectively extracting a fingerprint signal embedded in an image. In particular, in various embodiments disclosed in this document, when an image in which a fingerprint signal is inserted is output through a printer and transferred onto a physical object such as paper or label paper, the fingerprint signal is inserted from the image transferred onto the physical object/ It describes how to extract.

For example, the electronic device according to an embodiment of the present invention may acquire image data corresponding to an image transferred on a physical object through an equipped optical unit, and receive a fingerprint signal embedded in the image from the acquired image data. Can be extracted. Meanwhile, in acquiring image data corresponding to an image transferred on a physical object, the electronic device according to an embodiment of the present invention may perform correction on a geometrically deformed image due to various variables. Through this, the fingerprint signal embedded in the image can be accurately and quickly extracted.

An electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document includes: an optical unit that acquires image data for an external object through an image sensor in response to a signal for photographing being received, the image In the data, if the shape and size of the first rectangle approximating the border of the external object does not correspond to the shape and size of the preset second rectangle, the perspective of the image data is such that the first rectangle corresponds to the second rectangle Perspective correction unit that adjusts, a watermark signal prediction unit that extracts noise data from the image data using at least one filter, and performs auto-correlation on the image data, and In scanning a plurality of pixels in the image data using a mask window, the pixels having the largest result value of the autocorrelation operation for each region corresponding to the mask window are peak pixels for each of the regions. A peak detector that detects as, linearly connects points corresponding to each of the detected peak pixels to obtain one or more figures, and based on the gradient information, size information, and location information identified from the obtained figures, the external object Compute the gradient difference, size difference, and position difference between the original image data corresponding to and the image data, and correct the slope, size, and position of the image data so that the calculated gradient difference, size difference, and position difference are below a threshold A geometric deformation correction unit to obtain a pre-inserted fingerprint signal from the image data whose tilt, size and position are corrected, and an original message extraction unit to extract the original message, and to transmit the extracted original message to a predetermined server And, it may include a communication unit for receiving the data corresponding to the extracted original message from the server.

According to various embodiments disclosed in the present disclosure, an operation method of an electronic device for extracting watermark data from an image may include acquiring image data for an external object through an image sensor in response to a signal for photographing being received, In the image data, if the shape and size of the first rectangle approximating the border of the external object does not correspond to the shape and size of the preset second rectangle, the image data such that the first rectangle corresponds to the second rectangle Adjusting the perspective of the image, extracting noise data from the image data using at least one filter, and performing an auto-correlation operation on the image data, and then using a mask window of a preset size In the scanning of a plurality of pixels in the image data, detecting a pixel having the largest result value of the autocorrelation operation for each region corresponding to the mask window as peak pixels for each of the regions. , Acquiring one or more figures by linearly connecting points corresponding to each of the detected peak pixels, and an original image corresponding to the external object based on the gradient information, size information, and location information identified from the obtained figures Calculating gradient difference, size difference, and position difference between data and the image data, and correcting the slope, size, and position of the image data so that the calculated gradient difference, size difference, and position difference is less than or equal to a threshold, and the gradient , Acquiring a pre-inserted fingerprint signal from the image data whose size and position are corrected, extracting an original message, and transmitting the extracted original message to a predetermined server, and data corresponding to the extracted original message It may include the step of receiving from the server.

According to various embodiments disclosed in this document, unlike conventional QR codes that are directly inserted into an image and affect the design of the image, watermarking that increases security without compromising the design of the image according to invisible features As the technique is proposed, it is possible to prevent forgery and tampering through random extraction or illegal copying of data without compromising the original design of the image.

As the electronic device according to various embodiments disclosed in the present document is capable of accurately and quickly extracting a fingerprint signal having superior robustness and security and invisible characteristics, it can include only fragmentary information such as URLs and visible characteristics. As a result, it is possible to overcome the limitations of two-dimensional codes (eg, QR codes, barcodes) that have been limited in application, and through this, various user experiences and conveniences can be provided to users.

Various embodiments disclosed in this document consider the combination of convenience and security according to invisible features that information can be transmitted using an image acquired through a camera, and through the present invention, a user inserts a finger inserted into an image by an information provider. Print signals can be easily extracted and related information can be safely provided.

1 is a diagram illustrating the configuration of a server and an electronic device according to an embodiment of the present invention.
2 is a diagram illustrating a process of inserting a fingerprint signal into an original image and extracting the inserted fingerprint signal again from a server and an electronic device according to an embodiment of the present invention.
3 is a flowchart illustrating operations for correcting a slope, size, and position of image data and extracting an original message in an electronic device according to an embodiment of the present invention.
4 is a flowchart illustrating a process of performing perspective correction on image data in an electronic device according to an embodiment of the present invention.
5 is a flowchart illustrating a process of extracting a fingerprint signal and an original message from an electronic device according to an embodiment of the present invention.

The various embodiments disclosed in this document are not intended to limit the present invention to specific embodiments, and components introduced through various embodiments may include all changeable equivalents or substitutes included in the spirit and scope of the present invention. Those skilled in the art will readily understand that the present invention is provided as an inclusive meaning. In addition, in describing each drawing, unless defined otherwise, all terms used in this specification, including technical or scientific terms, are those that are generally understood by those skilled in the art to which the present invention pertains. It can be interpreted as having the same meaning. In addition, the objects and effects of the present invention, and technical configurations for achieving them will be clarified through embodiments described in detail with reference to the accompanying drawings. In the description of the present invention, when it is determined that a detailed description of known functions or configurations may unnecessarily obscure the subject matter of the present invention, detailed descriptions related thereto may be omitted, and the terms described later are structures in the present invention. , Terms defined in consideration of roles and functions, etc., which may be interpreted differently from the meanings previously used according to the intention or customs of users and operators.

It is noted that the present invention is not limited to the embodiments disclosed below, but can be implemented in various different forms. Various embodiments disclosed in this document are provided to make the disclosure of the present invention complete, and to fully inform the person of ordinary skill in the art to which the present invention pertains, the scope of the present invention, and the present invention is solely claimed. It is only defined by the scope of the claims recited in the scope.

In this document, when it is said that a part "includes" a component, this means that other components may be further included rather than excluding other components, unless specifically stated to the contrary. In addition, in various embodiments of the present invention, each component, functional blocks, or means may be composed of one or more sub-components, and the electrical, electronic, and mechanical functions performed by each component are electronic circuits. , May be implemented with various known devices or mechanical elements, such as an integrated circuit, an application specific integrated circuit (ASIC), or may be implemented separately or in combination of two or more.

On the other hand, the steps of the block diagrams or flowcharts of the accompanying block diagrams refer to computer program instructions mounted on a processor or memory of data processing equipment such as a general purpose computer, a special purpose computer, a portable notebook computer, and a network computer to perform designated functions. Can be interpreted. Since these computer program instructions can be stored in a memory provided in a computer device or in a computer readable memory, the functions described in the steps of the blocks or flowcharts of the block diagram are produced as an article containing instruction means for performing this. It may be. In addition, each block or step may represent a module, segment, or portion of code that includes one or more executable instructions for executing the specified logical function(s). It should also be noted that in some alternative embodiments, the functions mentioned in blocks or steps may be executed in a different order. For example, two blocks or steps shown in succession may be performed substantially simultaneously, or may be performed in reverse order, and in some cases, some blocks or steps may be omitted.

Hereinafter, a method of changing a state of a user interface in an electronic device according to various embodiments of the present invention will be described with reference to the accompanying drawings.

1 is a diagram illustrating the configuration of a server and an electronic device according to an embodiment of the present invention. According to various embodiments, the server 10 includes an original message generation unit 11, a key generation unit 12, a fingerprint signal generation unit 13, a communication unit 14, a fingerprint signal management unit 15, and a database ( 16). In addition, the electronic device 100 includes an optical unit 110, a perspective correction unit 120, a watermark signal prediction unit 130, a peak detection unit 140, a geometric deformation correction unit 150, and an original message extraction unit 160 ) And the communication unit 170. Here, the original message generation unit 11, the key generation unit 12, the fingerprint signal generation unit 13 and the fingerprint signal management unit 15 of the server 10 and the perspective correction unit 120 of the electronic device 100 ), the watermark signal prediction unit 130, the peak detection unit 140, the geometric deformation correction unit 150 and the original message extraction unit 160 were used to mean a module that performs a specific instruction, through the modules It is noted that the instructions executed may be understood to be performed by a processor (not shown) provided in the server 10 and the electronic device 100.

According to one embodiment, the original message generator 11 of the server 10 may generate information to be delivered to arbitrary consumers, such as a user of the electronic device 100, in the form of digital data. For example, the original message generation unit 11 may generate user information or product information according to a combination of uppercase and lowercase letters and numbers based on an American standard code for information interchange (ASCII) code as an original message.

The key generating unit 12 of the server 10 may generate a secret key capable of encrypting the original message generated by the original message generating unit 11. According to an embodiment, when the original message generated by the original message generating unit 11 is provided as a fingerprint signal or the like, a key is generated in the original message in order to be protected from an attempt to extract randomly by an unauthorized user. The secret key generated by the unit 12 may be added.

The fingerprint signal generation unit 13 of the server 10 may generate a fingerprint signal using an original message and a secret key. According to an embodiment, the fingerprint signal generator 13 generates 496 random random permutations (eg, (26*2+10)*8=496) to modulate the original message, and then generates the random permutations. 496 subkeys can be additionally generated using the field and the secret key. Subsequently, the fingerprint signal generation unit 13 may generate a fingerprint signal having a length corresponding to a predetermined unit block from the additionally generated subkey.

According to an embodiment, the fingerprint signal generator 13 may generate a watermark signal in order to add the generated fingerprint signal in various types of patterns or noise shapes on a designated original image. The watermark signal may be inserted in an invisible form into the designated original image. Meanwhile, in inserting a watermark signal on the original image, HVS (human visual system) technology may be used to insert the watermark signal as robustly as possible while maintaining invisibility.

The communication unit 14 of the server 10 establishes and establishes a direct (eg, wired) communication channel 30 or a wireless communication channel 20 between the server 10 and an external electronic device, for example, the electronic device 100. It can support the execution of communication through the established communication channel. According to an embodiment, the communication unit 14 may include a wireless communication module (eg, a cellular communication module, a short-range wireless communication module, or a global navigation satellite system (GNSS) communication module) or a wired communication module (eg, a local area network (LAN)). Communication module, or power line communication module).

The fingerprint signal management unit 15 of the server 10 responds to a request corresponding to the original message generated by the original message generation unit 11 received from an external electronic device, and the information corresponding to the received request server It can be searched in the database 16 of (10). In addition, the fingerprint signal management unit 15 may be set to transmit the searched information to the external electronic device that has transmitted the request through the communication unit 14.

The database 16 of the server 10 may store data or information generated from at least one component in the server 10. For example, the database 16 may store the original message generated by the original message generator 11 and information associated with the original message.

According to an embodiment, the optical unit 110 of the electronic device 100 may acquire image data for an external object through an image sensor in response to receiving a signal for photographing. The optical unit 110 may photograph still images and videos, and may include one or more lenses, image sensors, image signal processors, or flashes.

The perspective correction unit 120 of the electronic device 100, in the image data obtained through the optical unit 110, the shape and size of the second rectangle in which the shape and size of the first rectangle approximating the border of the external object is preset It can be determined whether or not. If, at least one of the shape and size of the first rectangle does not correspond to at least one of the shape and size of the second rectangle, the perspective correction unit 120 has the shape and size of the first rectangle as the shape of the second rectangle. And the perspective of the image data can be adjusted to correspond to the size. For example, the perspective correction unit 120 may adjust the perspective of the acquired image data so that the first square approximating the border of the external object corresponds to a preset square.

The watermark signal prediction unit 130 of the electronic device 100 may extract noise data from the acquired image data using at least one filter. Here, the at least one filter may include a Wiener filter and a morphology filter. For reference, the Winner filter is a filter that converts the input to the desired output as closely as possible, meaning'approximately as much as possible' means that the sum of the squares of the difference between the filter output and the desired result is minimal. Can. In addition, the morphology filter is a filter used for transforming the shape of a specific object existing in an image. Through erosion, a small chunk of objects disappear depending on the size and number of uses of the filter, or the size of the filter through expansion. And small noises inside the object may disappear according to the number of uses. According to an embodiment, the watermark signal prediction unit 130 may search for an area in which noise data is concentrated in image data through erosion and expansion operations. At this time, the searched region may be classified as a region of interest for extracting the fingerprint signal. On the other hand, it can be understood that the more the noise data is concentrated, the higher the probability that the fingerprint signal is inserted.

According to an embodiment, the watermark signal prediction unit 130 may have the same shape as noise data from image data obtained through the optical unit 110 or image data whose perspective is adjusted through the perspective correction unit 120. The mark signal can be separated. At this time, the watermark signal prediction unit 130 may use a noise data extraction filter or a high-pass filter to detect a watermark signal without original image data. Meanwhile, the noise data filtered by the noise data extraction filter may include a pre-inserted watermark signal.

According to one embodiment, the watermark signal prediction unit 130 may obtain prediction data of an original image corresponding to an external object and a watermark signal identified from the prediction data from the acquired image data. At this time, the watermark signal prediction unit 130 may obtain prediction data and a watermark signal using a Winner filter set so that an error between an output value and a preset output setting value is equal to or less than a preset threshold.

After the noise data is extracted from the image data through the filtering process, the peak detector 140 of the electronic device 100 may scan a plurality of pixels in the image data using a mask window having a preset size. For example, the peak detector 140 may start a scan operation from left to right. In this case, the peak detector 140 may detect a pixel having the highest auto-correlation value for each region corresponding to the mask window as peak pixels for each of the regions. According to an embodiment, the peak detector 140 maintains the values of the detected peak pixels, and the values of pixels other than the peak pixels may be changed to 0 (zero) or -1. On the other hand, as the mask window is moved in the area unit, not in pixel units, when the total size of the image data is not a multiple of the size of the mask window, the scan operation is performed from right to left, from top to bottom and from bottom to top. Can also be repeated.

The geometrical distortion correction unit 150 of the electronic device 100 may acquire one or more figures by connecting points corresponding to each of the peak pixels detected through the peak detection unit 140. Subsequently, the geometric deformation correcting unit 150 may acquire tilt information, size information, and location information from the obtained figures, and use the obtained information to determine the original image data and the optical unit 110 corresponding to the external object. The gradient difference, size difference, and position difference between image data obtained through the calculation may be calculated. The geometric deformation correcting unit 150 may adjust the slope, size, and position of the acquired image data so that the calculated slope difference, size difference, and position difference are less than or equal to a threshold. For example, the threshold may be set to a value close to 0, in which case the slope, size, and position of the acquired image data may be adjusted so that there is little difference in tilt, size, and position. According to an embodiment, the geometrical distortion correction unit 150 performs cross-correlation of a synchronization signal inserted into an image and an original synchronization pattern, and based on the result value, the The operation of translating the position, that is, adjusting the position may be performed.

According to an embodiment, the geometric deformation correcting unit 150 may divide the image data whose slope and size are corrected into at least two or more blocks, and then merge the divided blocks into one block. In addition, the geometric deformation correction unit 150 may correct the position of the image data through a cross-correlation operation between a synchronization signal inserted in the image data and an original synchronization pattern. Through this, RST (rotation, scaling, translation) correction may be performed on the image data acquired through the optical unit 110.

The original message extracting unit 160 of the electronic device 100 may obtain a pre-inserted fingerprint signal from the image data in which at least one of the tilt, size, and position is corrected, and receive the original message from the obtained fingerprint signal. Can be extracted.

The communication unit 170 of the electronic device 100 may transmit the original message extracted through the original message extraction unit 160 to a predetermined server (eg, the server 10). Also, the communication unit 170 may receive data corresponding to the extracted original message from the server (eg, the server 10). For example, the electronic device 100 may request information corresponding to the original message from the server 10 that generated the original message through the communication unit 170, and communicate the information transmitted from the server 10 to the communication unit 170. Can be received through. According to various embodiments, the data transmitted to the server 10 may be the fingerprint signal itself or an original message extracted from the fingerprint signal.

According to one embodiment, the original message generated by the server 10 may include text or numeric data. Although the fingerprint signal can be viewed as a concept similar to the currently used'two-dimensional barcode', it is different from the'two-dimensional barcode', which is just a set of figures in that information can be transmitted through an invisible image or image. There is a clear difference. A fingerprint signal configured to include a unique number or an identifier can be inserted into digital content without compromising the original shape or color of the digital content. As such, it is possible to know who distributed and to whom through the fingerprint signal embedded in the digital content, so it can be used to classify products in the logistics business or to check distribution channels. On the other hand, the original message or serial number included in the fingerprint signal can only be obtained by querying the database provided on the server 10 or the electronic device 100, and is not visually confirmed, so it is in the form of a picture such as a QR code or barcode. The security is excellent because it is not speculated as

2 is a diagram illustrating a process of inserting a fingerprint signal into an original image and extracting the inserted fingerprint signal again from a server and an electronic device according to an embodiment of the present invention.

According to an embodiment, steps 210 to 215 may be performed by the server 10, and steps 220 to 224 may be performed by the electronic device 100.

First, in step 210, the server 10 may generate an original message. The original message generated at this time may be converted into a fingerprint signal, and then inserted into the original image data 200 in the form of a watermark signal including the fingerprint signal and transmitted to arbitrary users.

In step 211, the server 10 may generate a fingerprint signal including the original message. The server 10 may generate a watermark signal in order to add the generated fingerprint signal in a pattern or noise form of various types on a designated original image.

In step 212, the server 10 may search for and determine an area into which the watermark signal is to be inserted in the original image data 200. For example, the server 10 includes a high frequency band and a diagonal area known to be relatively less sensitive to human vision to changes in image data, a texture area in which high frequency is concentrated, and a dark area and a bright area that is above a threshold compared to the surroundings. An area to which the watermark signal is to be inserted may be searched by assigning a weight to the area. On the other hand, when an area in which the watermark signal is to be inserted in the original image data 200 is determined, the server 10 may repeatedly insert the watermark signal into the original image several times.

In step 213, the server 10 may use the discrete wavelet transform of step 2 in inserting the watermark signal into the original image data 200. Here, the discrete wavelet transform of the second stage may refer to a technique of separating/recovering the original image into a spatial domain and a frequency domain over two stages. According to an embodiment, the server 10 may convert the original image data 200 into a two-step wavelet, and then repeatedly insert a watermark signal into at least a portion of the original image data 200. In addition, a synchronization signal generated by a fixed key and composed of a unit block size may be repeatedly inserted into an area in which the watermark signal is not inserted in the original image data 200. On the other hand, the server 10 may repeatedly insert a watermark signal or a synchronization signal modulated with the size of a unit block into the original image in order to correct if the original image or image data corresponding to the original image is geometrically deformed. have. Through this, the server 10 may generate image data 201 in which a watermark signal is inserted.

According to various embodiments, the server 10 may output the image data 201 on a physical object. For example, the server 10 may transfer the image data 201 on paper or label paper using a printing device or the like. The electronic device 100 may use the optical unit 110 provided to photograph an image transferred on paper or label paper, and acquire image data corresponding to the photographed image. According to an embodiment, the image data obtained through the optical unit 110 may be essentially the same image data as the image data 201 into which the watermark signal is inserted. However, since the inclination of the electronic device 100 or the distance between the paper or label sheet and the electronic device 100 may vary each time an image transferred on the paper or label sheet is taken, the image data acquired through the optical unit 110 may be changed. The slope, size, and position may be different from the slope, size, and position of the image data 201 into which the watermark signal is inserted. Accordingly, the electronic device 100 needs to perform an operation of correcting the tilt, size, and position of the image data obtained through the optical unit 110.

According to an embodiment, in operation 220, the electronic device 100 may predict a watermark signal from image data acquired through the optical unit 110. For example, the electronic device 100 may extract noise data and a region of interest from the image data obtained through the optical unit 110 using at least one filter, and correct the image data. Also, the electronic device 100 may identify data 202 predicted to correspond to a watermark signal including a fingerprint signal (and an original message) from the extracted image data. Meanwhile, a WINNER filter may be used to identify data 202 predicted to correspond to a watermark signal.

In operation 221, the electronic device 100 may detect one or more peak pixels from the data 202. According to an embodiment, the data 202 may include watermark signals repeatedly inserted by the server 10 according to the discrete wavelet transform. The electronic device 100 may detect watermark signals repeatedly inserted into the original image according to the discrete wavelet transform as a peak pixel in the spatial domain. At this time, the peak pixels may have characteristics that are periodically repeated in the data 202. For example, peak pixels detected in data 202 can be identified by nine points as identified in data 203.

In operation 222, the electronic device 100 may correct geometrical deformation, such as a slope, size, and location, identified in the data 203 using periodicity of basic blocks identified through the identified peak pixels. Through this, the electronic device 100 may acquire data 204 whose geometric deformation has been corrected.

In step 223, the electronic device 100 may extract the fingerprint signal from the data 204. Also, the electronic device 100 may extract the original message using the extracted fingerprint signal.

3 is a flowchart illustrating operations for correcting a slope, size, and position of image data and extracting an original message in an electronic device according to an embodiment of the present invention.

In step 300, image data for an external object may be acquired. According to an embodiment, image data for an external object may be obtained through an optical unit such as an image sensor provided in the electronic device.

In step 310, perspective correction on image data may be performed. Here, perspective correction means identifying at least one object (or figure) included in the acquired image data, and modifying the identified at least one object in a form corresponding to a preset rectangle or square, and then including it in the image data It may mean correcting the size and position of the pixels.

In step 320, it may be determined whether at least one object included in the image data has a rectangular shape. If at least one object is modified to a shape corresponding to a preset rectangle or square through step 310 or if at least one object having a square shape is included in the image data, step 330 may be performed. On the other hand, if an object having a rectangular shape is not identified in the image data, it may be induced to newly acquire image data by performing step 300 again.

In step 330, the slope and size of the image data may be corrected. At this time, an auto-correlation operation is performed on the image data, and a plurality of pixels in the image data may be scanned using a mask window having a preset size. In addition, a pixel having the largest result value of the autocorrelation operation for each region corresponding to the mask window may be detected as peak pixels for each of the regions.

Subsequently, in step 340, a cross-correlation operation between the synchronization signal and the original synchronization signal of the image data whose slope and size are corrected may be performed, and through this, peak pixels in the image data may be identified.

In step 350, it may be determined whether the peak values of each of the identified peak pixels are appropriate. If it is determined that each of the peak values is equal to or greater than a preset threshold and is determined to be appropriate, step 360 may be performed. On the other hand, if it is determined that each of the peak values is less than a preset threshold and it is determined that it is not appropriate, step 300 may be performed again to induce image data to be newly acquired.

In step 360, the position of the image data can be corrected. Subsequently, in step 370, the image data whose position is corrected may be divided into at least two or more blocks, and the divided blocks may be merged into one block. In addition, it is possible to identify the accuracy of the moved position of the image data by calculating a cross-correlation between the synchronization signal inserted in the image data and the original synchronization signal.

In step 380, it may be determined whether the identified accuracy is greater than or equal to a threshold. If it is determined that the identified accuracy is greater than or equal to the threshold, step 390 may be performed. On the other hand, if it is determined that the identified accuracy is below a threshold, step 300 may be performed again to induce image data to be newly acquired.

In step 390, the original message may be extracted from the image data. That is, the electronic device according to an embodiment may extract a fingerprint signal similar to that inserted into the original image data from the image data whose geometric deformation is corrected, such as tilt, size, and position, and extract the original message from the extracted fingerprint signal. Can be extracted.

4 is a flowchart illustrating a process of performing perspective correction on image data in an electronic device according to an embodiment of the present invention. The electronic device according to an embodiment may perform different correction operations according to preset conditions in performing perspective correction on image data in operation 310 of FIG. 3.

In step 400, when the perspective correction request for the image data is received, the electronic device may preferentially perform the determination in step 405. That is, in step 405, the electronic device may determine whether at least one rectangle is detected from the image data. Here, the at least one rectangle is a rectangle composed of angles within a predetermined threshold angle, and may preferably mean a rectangle or a square.

If at least one rectangle is detected, the electronic device may perform step 410. On the other hand, if no rectangle is detected, the electronic device may perform the determination according to step 425.

In operation 410, the electronic device may identify four sides constituting the detected at least one rectangle. Subsequently, in step 415, the electronic device may adjust the lengths of the four identified sides to correspond to the lengths of the predetermined square sides. Through this, in step 420, the electronic device may correct the detected at least one square to correspond to a preset square.

In step 425, the electronic device may determine whether original image data can be obtained. If it is determined that the original image data can be obtained, the electronic device may perform step 430, and vice versa.

In operation 430, the electronic device may search for previously acquired image data based on the original image data. Based on this, in step 435, the electronic device may adjust the size and position of the detected image data to correspond to the size and position of the original image data. According to various embodiments, when the operations according to steps 430 to 435 are performed, in step 440, the electronic device performs slope and size correction on the image data regardless of whether at least one rectangle is detected in the image data. It can be done.

In operation 445, the electronic device may extract noise data from the image data through a Wiener filter. In addition, in step 450, the electronic device may perform segmentation on the extracted noise data using a morphology filter. In a subsequent step 455, the electronic device may extract a border for one or more objects in the acquired image data, and in step 460, at least one rectangle may be identified through shape approximation of the extracted border. Finally, in step 465, the electronic device may correct at least one square to correspond to a preset square.

5 is a flowchart illustrating a process of extracting a fingerprint signal and an original message from an electronic device according to an embodiment of the present invention.

In step 500, in response to receiving a signal for photographing, it is possible to acquire image data for an external object through an image sensor.

In step 510, if the shape and size of the first rectangle that approximates the border of the external object in the image data does not correspond to the shape and size of the second preset rectangle, the image data of the first rectangle is matched to the second rectangle. Perspective can be adjusted.

In operation 520, noise data may be extracted from the acquired image data using at least one filter.

In step 530, after performing an auto-correlation operation on the image data, scanning a plurality of pixels in the image data using a mask window having a preset size, autocorrelation for each region corresponding to the mask window The pixel having the largest result of the operation can be detected as peak pixels for each of the regions.

In operation 540, one or more figures may be obtained by linearly connecting points corresponding to each of the detected peak pixels.

In operation 550, a gradient difference, a size difference, and a location difference between the original image data and the image data corresponding to the external object may be calculated based on the gradient information, size information, and location information identified from the obtained figures.

In step 560, the slope, size, and position of the image data may be corrected such that the calculated slope difference, size difference, and position difference are less than or equal to a threshold.

In operation 570, a pre-inserted fingerprint signal may be obtained from the image data whose tilt, size and position are corrected.

In step 580, the original message may be extracted.

In step 590, the extracted original message may be transmitted to a predetermined server, and data corresponding to the extracted original message may be received from the server.

According to various embodiments, some of the operations disclosed in FIGS. 3, 4, and 5 may be omitted or repeated multiple times. In addition, each of the operations disclosed in FIGS. 3, 4, and 5 is appropriately viewed as an embodiment, and cannot be interpreted as being limited to one operation being dependent on the other.

An electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document includes: an optical unit that acquires image data for an external object through an image sensor in response to a signal for photographing being received, the image In the data, if the shape and size of the first rectangle approximating the border of the external object does not correspond to the shape and size of the preset second rectangle, the perspective of the image data is such that the first rectangle corresponds to the second rectangle Perspective correction unit that adjusts, a watermark signal prediction unit that extracts noise data from the image data using at least one filter, and performs auto-correlation on the image data, and In scanning a plurality of pixels in the image data using a mask window, the pixels having the largest result value of the autocorrelation operation for each region corresponding to the mask window are peak pixels for each of the regions. A peak detector that detects as, linearly connects points corresponding to each of the detected peak pixels to obtain one or more figures, and based on the gradient information, size information, and location information identified from the obtained figures, the external object Compute the gradient difference, size difference, and position difference between the original image data corresponding to and the image data, and correct the slope, size, and position of the image data so that the calculated gradient difference, size difference, and position difference are below a threshold A geometric deformation correction unit, an original message extraction unit for obtaining a pre-inserted fingerprint signal from the image data whose slope, size and position are corrected, and extracting an original message from the fingerprint signal using a decryption key, and And a communication unit that transmits the extracted original message to a designated server and receives data corresponding to the extracted original message from the server.

In the electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document, the watermark signal prediction unit identifies from the prediction data and the prediction data of the original image corresponding to the external object from the image data It may be characterized in that the acquired watermark signal.

In an electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document, the watermark signal prediction unit may include an error between an output value for the at least one filter and a preset output setting value below a preset threshold. It may be characterized in that it comprises at least one filter set to be.

In an electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document, the at least one filter may include a Wiener filter and a morphology filter. .

In the electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document, the geometrical distortion correction unit divides the image data whose slope and size are corrected into at least two or more blocks, and then the It is characterized in that the divided blocks are merged into one block, and the position of the image data is corrected through a cross-correlation operation between a synchronization signal inserted in the image data and an original synchronization pattern. Can.

In the electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document, the perspective correction unit, according to a first mode, performs edge detection for one or more objects in the image data. In response to detecting at least one rectangle consisting of angles within a threshold angle, identifying four sides constituting the detected at least one rectangle, and determining the lengths of the identified four sides of a predetermined size. In response to acquiring the original image data according to the second mode, the at least one square detected is corrected to a square having the preset size by adjusting to correspond to the length of the side constituting the square, and according to the second mode, The size and position are corrected to correspond to the size and position of the original image data, and according to the third mode, in response to noise data being extracted through a wiener filter, a morphology filter is used. The segmentation of the extracted noise data is performed to extract a border for one or more objects in the image data, identify at least one rectangle through shape approximation to the extracted border, and identify the identified at least one It may be characterized in that the square is corrected to a square of the preset size.

According to various embodiments disclosed in the present disclosure, an operation method of an electronic device for extracting watermark data from an image may include acquiring image data for an external object through an image sensor in response to a signal for photographing being received, In the image data, if the shape and size of the first rectangle approximating the border of the external object does not correspond to the shape and size of the preset second rectangle, the image data such that the first rectangle corresponds to the second rectangle Adjusting the perspective of the image, extracting noise data from the image data using at least one filter, and performing an auto-correlation operation on the image data, and then using a mask window of a preset size In the scanning of a plurality of pixels in the image data, detecting a pixel having the largest result value of the autocorrelation operation for each region corresponding to the mask window as peak pixels for each of the regions. , Acquiring one or more figures by linearly connecting points corresponding to each of the detected peak pixels, and an original image corresponding to the external object based on the gradient information, size information, and location information identified from the obtained figures Calculating gradient difference, size difference, and position difference between data and the image data, and correcting the slope, size, and position of the image data so that the calculated gradient difference, size difference, and position difference is less than or equal to a threshold, and the gradient , Acquiring a pre-inserted fingerprint signal from the image data whose size and position are corrected, extracting an original message, and transmitting the extracted original message to a predetermined server, and data corresponding to the extracted original message It may include the step of receiving from the server.

In an operation method of an electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document, the step of extracting noise data from the image data includes: extracting noise data from the image data of the original image corresponding to the external object. The method may further include obtaining prediction data and a watermark signal identified from the prediction data.

In an operation method of an electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document, extracting noise data from the image data includes setting an output value and a preset output value for the at least one filter. It may be characterized in that it further comprises the step of removing the noise data using the at least one filter set so that the error between values is below a preset threshold.

In an operation method of an electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document, the at least one filter includes a Wiener filter and a morphology filter can do.

In an operation method of an electronic device for extracting watermark data from an image according to various embodiments disclosed in the present document, the step of correcting the slope, size, and position of the image data may include correcting the slope and size corrected image data. After dividing into at least two or more blocks, merging the divided blocks into one block, and correcting the position of the image data through a cross-correlation operation between a synchronization signal inserted in the image data and an original synchronization signal It may be characterized in that it further comprises a step.

In an operation method of an electronic device for extracting watermark data from an image according to various embodiments disclosed in the present disclosure, adjusting the perspective of the image data may include one or more objects in the image data according to a first mode. In response to detection of at least one rectangle composed of angles within a threshold angle by performing edge detection for, four sides constituting the detected at least one rectangle are identified, and the identified four Compensating the detected at least one rectangle to a square of the preset size by adjusting the length of the sides to correspond to the length of the sides constituting the square of the preset size, according to the second mode, the original image data In response to being acquired, correcting the size and position of the image data to correspond to the size and position of the original image data, in response to noise data being extracted through a Wiener filter according to a third mode Then, a morphology filter is used to partition the extracted noise data to extract a border for one or more objects in the image data, and at least one of shapes is approximated to the extracted border. It may be characterized in that it further comprises the step of identifying a rectangle, and correcting the identified at least one rectangle to a square of the preset size.

A method of controlling an electronic device that extracts a fingerprint signal from an image according to an embodiment of the present invention may be implemented as a computer program stored in a storage medium for execution through combination with a computer.

In addition, a method of controlling an electronic device that extracts a fingerprint signal from an image according to an embodiment of the present invention may be implemented in a form of program instructions that can be executed through various computer means and recorded on a computer readable medium. The computer-readable medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention or may be known and usable by those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, and magnetic media such as floptical disks. -Hardware devices specifically configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language code that can be executed by a computer using an interpreter, etc., as well as machine language codes produced by a compiler.

As described above, in the present invention, specific matters such as specific components and the like have been described by limited embodiments and drawings, but they are provided only to help a more comprehensive understanding of the present invention, and the present invention is not limited to the above embodiments , Anyone having ordinary knowledge in the field to which the present invention pertains can make various modifications and variations from these descriptions.

Therefore, the spirit of the present invention should not be limited to the described embodiments, and should not be determined, and all claims that are equivalent or equivalent to the scope of the claims as well as the claims to be described later belong to the scope of the spirit of the invention. .

Claims (14)

An optical unit that acquires image data for an external object through an image sensor in response to receiving a signal for shooting;
In the image data, if the shape and size of the first rectangle approximating the border of the external object does not correspond to the shape and size of the preset second rectangle, the image data such that the first rectangle corresponds to the second rectangle Perspective correction unit to adjust the perspective of the:
A watermark signal prediction unit extracting noise data from the image data using at least one filter;
In performing a self-correlation operation on the image data, and scanning a plurality of pixels in the image data using a mask window having a preset size, the autocorrelation for each region corresponding to the mask window A peak detector configured to detect pixels having the largest result of the calculation as peak pixels for each of the regions;
Original image data corresponding to the external object is obtained by linearly connecting points corresponding to each of the detected peak pixels to obtain one or more figures, and based on the gradient information, size information, and location information identified from the obtained figures. And a geometric deformation correcting unit that calculates a gradient difference, a size difference, and a position difference between the image data, and corrects the slope, size, and position of the image data so that the calculated gradient difference, a size difference, and a position difference fall below a threshold value;
An original message extracting unit that acquires a pre-inserted fingerprint signal from the tilted, sized, and corrected image data, and extracts the original message; And
And a communication unit that transmits the extracted original message to a predetermined server, and receives data corresponding to the extracted original message from the server,
The watermark signal prediction unit,
An electronic device for extracting a fingerprint signal from an image, characterized in that the prediction data of the original image corresponding to the external object and the watermark signal identified in the prediction data are obtained from the image data.
delete According to claim 1,
The watermark signal prediction unit,
And the at least one filter set so that an error between an output value for the at least one filter and a preset output setting value is equal to or less than a preset threshold, an electronic device for extracting a fingerprint signal from an image.
The method of claim 3,
The at least one filter comprises a Wiener filter (wiener filter) and a morphology filter, characterized in that the electronic device for extracting a fingerprint signal from the image.
According to claim 1,
The geometrical distortion correction unit,
After dividing the image data whose slope and size are corrected into at least two or more blocks, the divided blocks are merged into one block, and between the synchronization signal inserted in the image data and the original sync pattern. An electronic device for extracting a fingerprint signal from an image, characterized in that the position of the image data is corrected through a cross-correlation operation.
According to claim 1,
The perspective correction unit,
According to the first mode, in response to detecting at least one rectangle consisting of angles within a threshold angle by performing edge detection on one or more objects in the image data, the detected at least one rectangle Identifying the four sides constituting the, and adjusting the length of the identified four sides to correspond to the length of the sides constituting a square of a predetermined size to correct the detected at least one rectangle to a square of the preset size and,
According to the second mode, in response to the original image data being obtained, the size and position of the image data is corrected to correspond to the size and position of the original image data,
According to the third mode, in response to the noise data being extracted through the Wiener filter, the morphology filter is used to partition the extracted noise data to perform one or more objects in the image data In the image, characterized by extracting a border for a field, identifying at least one rectangle through shape approximation to the extracted border, and correcting the identified at least one rectangle to a square of the preset size. An electronic device that extracts a fingerprint signal.
In response to receiving a signal for shooting, obtaining image data for an external object through an image sensor;
In the image data, if the shape and size of the first rectangle approximating the border of the external object does not correspond to the shape and size of the preset second rectangle, the image data such that the first rectangle corresponds to the second rectangle Steps to adjust the perspective of:
Extracting noise data from the image data using at least one filter;
In performing a self-correlation operation on the image data, and scanning a plurality of pixels in the image data using a mask window having a preset size, the autocorrelation for each region corresponding to the mask window Detecting a pixel having the largest result of the calculation as peak pixels for each of the regions;
Original image data corresponding to the external object is obtained by linearly connecting points corresponding to each of the detected peak pixels to obtain one or more figures, and based on the gradient information, size information, and location information identified from the obtained figures. And calculating a gradient difference, a size difference, and a position difference between the image data, and correcting the slope, size, and position of the image data so that the calculated gradient difference, a size difference, and a position difference are below a threshold value;
Obtaining a pre-inserted fingerprint signal from the tilt, size and position corrected image data and extracting an original message; And
And sending the extracted original message to a pre-specified server, and receiving data corresponding to the extracted original message from the server.
The step of extracting noise data from the image data,
And obtaining the prediction data of the original image corresponding to the external object from the image data and the watermark signal identified in the prediction data, the operation of the electronic device extracting the fingerprint signal from the image. Way.
delete The method of claim 7,
The step of extracting noise data from the image data,
In the image, characterized in that it further comprises the step of removing the noise data using the at least one filter set so that the error between the output value for the at least one filter and a preset output setting value is below a preset threshold. Method of operation of an electronic device for extracting a fingerprint signal.
The method of claim 9,
The at least one filter comprises a Wiener filter (wiener filter) and a morphology filter (morphology filter), the operation method of the electronic device for extracting a fingerprint signal from the image.
The method of claim 7,
Compensating the slope, size and position of the image data,
Dividing the image data whose slope and size are corrected into at least two or more blocks, and then merging the divided blocks into one block;
And correcting the position of the image data through a cross-correlation operation between a synchronization signal inserted in the image data and an original synchronization signal, wherein the electronic device extracts a fingerprint signal from the image.
The method of claim 7,
Adjusting the perspective of the image data,
According to the first mode, in response to detecting at least one rectangle consisting of angles within a threshold angle by performing edge detection on one or more objects in the image data, the detected at least one rectangle Identifying the four sides constituting the, and adjusting the length of the identified four sides to correspond to the length of the sides constituting a square of a predetermined size to correct the detected at least one square to a square of the preset size To do;
According to a second mode, in response to obtaining the original image data, correcting the size and position of the image data to correspond to the size and position of the original image data;
According to the third mode, in response to the noise data being extracted through the Wiener filter, the morphology filter is used to partition the extracted noise data to perform one or more objects in the image data And extracting a border for the fields, identifying at least one square through shape approximation to the extracted border, and correcting the identified at least one square to a square having the preset size. A method of operating an electronic device that extracts a fingerprint signal from an image.
A computer-readable recording medium recording a program that causes a computer to perform the method of any one of claims 7, 9, 10, 11 or 12.
A computer program stored in a storage medium for executing the method of any one of claims 7, 9, 10, 11 or 12 through a combination with a computer.

Images