[go: up one dir, main page]

WO2010076952A2 - Système de test automatique pour évaluer une déformation d'image dans un dispositif d'image numérique - Google Patents

Système de test automatique pour évaluer une déformation d'image dans un dispositif d'image numérique Download PDF

Info

Publication number
WO2010076952A2
WO2010076952A2 PCT/KR2009/005991 KR2009005991W WO2010076952A2 WO 2010076952 A2 WO2010076952 A2 WO 2010076952A2 KR 2009005991 W KR2009005991 W KR 2009005991W WO 2010076952 A2 WO2010076952 A2 WO 2010076952A2
Authority
WO
WIPO (PCT)
Prior art keywords
watermark
image
signal
reference image
image distortion
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/KR2009/005991
Other languages
English (en)
Korean (ko)
Other versions
WO2010076952A3 (fr
Inventor
배기혁
김현태
조인제
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CK&B Co Ltd
Original Assignee
CK&B Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by CK&B Co Ltd filed Critical CK&B Co Ltd
Publication of WO2010076952A2 publication Critical patent/WO2010076952A2/fr
Publication of WO2010076952A3 publication Critical patent/WO2010076952A3/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N17/00—Diagnosis, testing or measuring for television systems or their details
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N17/00—Diagnosis, testing or measuring for television systems or their details
    • H04N17/004—Diagnosis, testing or measuring for television systems or their details for digital television systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N17/00—Diagnosis, testing or measuring for television systems or their details
    • H04N17/002—Diagnosis, testing or measuring for television systems or their details for television cameras
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/76—Television signal recording
    • H04N5/91—Television signal processing therefor
    • H04N5/913—Television signal processing therefor for scrambling ; for copy protection

Definitions

  • the present invention relates to an image distortion determination test automation system. More particularly, a signal capable of identifying and analyzing a distortion occurrence point of an image is inserted into a reference image in the form of a watermark, and then extracted from the reference image. An image distortion determination test automation system for automatically determining a distortion occurrence point of a digital imaging apparatus.
  • digital imaging devices such as video, DVD players, digital television receivers, digital set-top boxes, digital multimedia broadcasting (DMB) phones, portable multimedia players (PMPs), etc.
  • DMB digital multimedia broadcasting
  • PMPs portable multimedia players
  • the inspector In such a test, the inspector (monitoring agent) performs a subjective evaluation on several evaluation items by the inspector's supervision while watching a test image that is familiar to him.
  • the inspector evaluates evaluation items such as image quality, contrast, brightness, color reproduction, noise, and the like. Since the test method is manually performed by a person, evaluation criteria vary depending on the inspector. The disadvantage is that you can't do a level test.
  • An object of the present invention is to provide an image distortion determination test automation system in which image distortion due to a problem of a digital imaging apparatus is determined using digital watermarking technology, thereby eliminating subjectivity of determination and ensuring objectivity thoroughly.
  • Another object of the present invention is to provide an image distortion determination test automation system that can reduce the time and effort required to test the image quality.
  • a preferred embodiment of the image distortion determination test automation system of the digital image output apparatus of the present invention for solving the above problems is a watermark generator for generating an image distortion determination signal and using the same; A watermark inserter for embedding the reference image, an image output device for displaying the reference image into which the watermark is inserted, a frame storage server for capturing and storing the reference image output from the output terminal of the image output device in real time; And extracting the watermark from the captured reference image, and using the extracted image distortion determination signal, the image distortion determination signal test module determines and reports the image distortion of the corresponding image.
  • the image distortion determination signal may include a Temporal Drop-Frame Detection (TDFD) signal capable of determining a frame drop occurring on a time axis of a video and an SDBD (Spatial) capable of determining image distortion within an image frame. And a Distort-Block Detection) signal.
  • TDFD Temporal Drop-Frame Detection
  • SDBD Spatial Deformation Deformation
  • the watermark generator may further include randomizing a TDFD signal generator for generating the TDFD signal, an SDBD signal generator for generating the SDBD signal, and a position at which the TDFD signal and the SDBD signal are inserted into the reference image. And a signal randomizer for generating a watermark.
  • the watermark inserter may further include: a watermark embedding strength determiner configured to determine a watermark embedding strength for embedding the watermark into a reference image; and tiling the watermark into the full size of the reference image; And a watermark embedding unit for inserting the tiled watermark into the reference image by applying the determined watermark embedding strength.
  • a watermark embedding strength determiner configured to determine a watermark embedding strength for embedding the watermark into a reference image; and tiling the watermark into the full size of the reference image
  • a watermark embedding unit for inserting the tiled watermark into the reference image by applying the determined watermark embedding strength.
  • the frame storage server may include a frame capture unit for capturing a reference image displayed on the image output apparatus in real time, and a frame buffer for storing the captured image frame.
  • the image distortion determination signal test module may further include: a watermark detection unit configured to detect a watermark by applying the predicted watermark embedding intensity after estimating the embedding intensity of the watermark embedded in the captured reference image; An image distortion determination unit that determines whether the image is distorted in the corresponding video frame using the SDBD signal among the detected watermark signals, and a frame drop determination that determines whether a frame drop is generated using the TDFD signal among the detected watermark signals. And a reporting unit for generating a report on a result of whether the image is distorted and whether a frame drop occurs.
  • a preferred embodiment of the image distortion determination test automation system of the digital image acquisition device of the present invention is a watermark generator for generating an image distortion determination signal and generating a watermark using the same, and embedding the watermark in a reference image.
  • An image acquisition apparatus for acquiring a reference image into which the watermark is inserted, and extracting the watermark from the obtained reference image, and automatically using the extracted image distortion determination signal.
  • an image distortion determination signal test module for determining whether the focus and exposure functions are normally operated.
  • the image distortion determination signal is an AFED (Auto Focus Exposure Detection) signal that can determine whether the normal operation of the auto focus and exposure function of the image acquisition device and a pilot for corresponding to the two-dimensional movement of the reference image Characterized in that it comprises a signal.
  • AFED Automatic Focus Exposure Detection
  • the watermark generator may randomize an AFED signal generator for generating the AFED signal, a synchronization signal generator for generating the pilot signal, and a position at which the AFED signal and the pilot signal are to be inserted into the reference image. And a signal randomizer for generating a watermark.
  • the watermark inserter may further include: a watermark embedding strength determiner configured to determine a watermark embedding strength for embedding the watermark into a reference image; and tiling the watermark into the full size of the reference image; And a watermark embedding unit for inserting the tiled watermark into the reference image by applying the determined watermark embedding strength.
  • a watermark embedding strength determiner configured to determine a watermark embedding strength for embedding the watermark into a reference image; and tiling the watermark into the full size of the reference image
  • a watermark embedding unit for inserting the tiled watermark into the reference image by applying the determined watermark embedding strength.
  • the image distortion determination signal test module may further include predicting a watermark embedding intensity after predicting a watermark embedding intensity inserted into the obtained reference image and a preprocessing unit performing a preprocessing process on the obtained reference image. Applying a watermark detection unit to detect the watermark, a geometric deformation prediction unit predicting the degree of geometric deformation of the reference image, and a geometric deformation recovery using the predicted geometric deformation degree to restore the geometric deformation of the reference image And a comparison synchronization signal generator for generating a comparison pilot signal for restoring two-dimensional movement of the reference image, and measuring the similarity between the generated pilot signal and a pilot signal among the detected watermark signals.
  • a two-dimensional movement recovery unit for restoring two-dimensional movement of the image, and the detected watermark signal Including using a signal AFED parts auto focus / exposure function is determined for determining whether the auto-focus and normal operation of the exposure function of the image capture device characterized by comprising.
  • the image quality automation test can be automatically performed without an inspector, thereby ensuring preciseness and objectivity.
  • the time and effort required for the image distortion test may be reduced, and the cost for the image distortion determination test may be saved.
  • FIG. 1 is a diagram showing the configuration of an image distortion determination test automation system of a digital image output device according to the present invention
  • FIG. 2 is a diagram illustrating a configuration of a watermark generator using an image distortion determination signal in an image distortion determination test automation system of a digital image output apparatus of the present invention.
  • FIG. 3 is a view showing the configuration of the watermark inserter in the image distortion determination test automation system of the digital image output device of the present invention.
  • FIG. 4 is a diagram illustrating a configuration of an image distortion determination signal test module in an image distortion determination test automation system of a digital image output apparatus according to the present invention.
  • FIG. 5 is a flowchart illustrating a method for automating an image distortion determination test of a digital image output apparatus of the present invention.
  • FIG. 6 is a view showing the configuration of the image distortion determination test automation system of the digital image acquisition device of the present invention.
  • FIG. 7 is a view showing the configuration of a watermark generator using the image distortion determination signal in the image distortion determination test automation system of the digital image acquisition device of the present invention.
  • FIG. 8 is a view showing the configuration of the watermark inserter in the image distortion determination test automation system of the digital image acquisition device of the present invention.
  • FIG. 9 is a view showing the configuration of the image distortion determination signal test module in the image distortion determination test automation system of the digital image acquisition device of the present invention.
  • FIG. 10 is a graph showing an autocorrelation pattern of a watermark signal.
  • FIG. 11 is a flowchart illustrating a method for automating an image distortion determination test of a digital image acquisition device of the present invention.
  • FIGS. 1 to 11 a specific embodiment of an image distortion determination test automation system of a digital imaging apparatus of the present invention will be described with reference to FIGS. 1 to 11.
  • this is only an example and the present invention is not limited thereto.
  • digital watermarking is concealed by inserting predetermined information in a signal form that cannot be visually or audibly recognized in multimedia content such as text, video, still images, and audio, and hiding hidden information.
  • This technology is extracted and used as additional information for copyright, video certification and video monitoring.
  • the present invention is characterized by automatically testing whether the digital imaging device is operating normally by using such digital watermarking technology.
  • the digital imaging apparatus is classified into two types.
  • One is a digital image output apparatus, for example, a digital television receiver, a digital set-top box, a PMP, a DMB phone, etc.
  • the other is a digital image acquisition apparatus.
  • a digital image output apparatus for example, a digital television receiver, a digital set-top box, a PMP, a DMB phone, etc.
  • a digital image acquisition apparatus for example, it refers to a digital camera, a digital camcorder and the like.
  • the digital image output apparatus automatically tests and monitors a case in which an input image is not normally output due to a hardware or software problem.
  • an image distortion determination signal is inserted into a reference image using a watermarking technique, and a frame of the image is captured in real time while the image into which the image distortion determination signal is inserted is displayed, and then the image distortion determination signal is captured from the captured image. It detects whether the input image is broken, distorted or damaged, and stores the result in a database and reports it.
  • the digital image capturing apparatus tests and monitors whether auto focus and exposure detection are normally performed on an image acquired through a digital camera.
  • an autofocus / exposure determination signal is inserted into a reference image using a watermarking technique, an image in which the autofocus / exposure determination signal is inserted is acquired through a digital image acquisition device, and then autofocus / exposure is obtained from the acquired image.
  • the determination signal is extracted to determine whether the auto focus and exposure function of the digital image acquisition device is normally operating.
  • FIG. 1 is a diagram illustrating a configuration of an image distortion determination test automation system of a digital image output apparatus according to the present invention.
  • the image distortion determination test automation system 100 of the present invention includes a watermark generator 110, a watermark inserter 120, an image output device 130, a frame storage server 140, and image distortion.
  • the determination signal test module 150 is included.
  • the watermark generator 110 may determine a frame drop occurring on the time axis of the video and an image distortion within one frame of the video. A distortion determination signal is generated and a unit watermark is generated using the distortion determination signal.
  • the frame drop means that the n-th image is displayed and the n-th image is not displayed after the n-th image is displayed due to a hardware or software problem when the n-th image is to be displayed in the currently input video. Say the phenomenon.
  • the watermark generator 110 receives a Temporal Drop-Frame Detection (TDFD) signal capable of determining the frame drop and a Spatial Distort-Block Detection (SDBD) signal capable of determining image distortion within a frame. After each generation, a unit watermark is generated by randomizing positions where the TDFD signal and the SDBD signal are to be inserted into the reference image.
  • TDFD Temporal Drop-Frame Detection
  • SDBD Spatial Distort-Block Detection
  • the watermark inserter 120 projects a watermark to an area to insert a watermark with respect to a reference image input from the outside, and inserts the watermark generated by the watermark generator 110 into the reference image. After the mark insertion intensity is determined, the watermark is inserted by applying the determined watermark embedding intensity.
  • the watermark inserter 120 tiles the unit watermark generated by the watermark generator 110 to the entire image size and inserts the unit watermark into the reference image.
  • the image output device 130 receives a reference image into which the image distortion determination signal is inserted in the form of a watermark and displays the reference image.
  • the frame storage server 140 includes a frame capture unit 141 and a frame buffer 144.
  • the frame capture unit 141 captures a reference image displayed by the image output device 130 in real time.
  • the frame buffer 144 stores image frames captured by the frame capture unit 141.
  • the frame storage server 140 is connected to the output terminal of the image output device 130 to capture an image signal output from the output terminal of the image output device 130 through the frame capture unit 141, It is stored in the frame buffer 144.
  • the image distortion determination signal test module 150 receives image frames stored in the frame buffer 144 in order, projects the input image frames into a region for detecting a watermark, and then inserts a watermark.
  • a watermark signal is extracted from the image frame by predicting the intensity.
  • the image distortion determination signal test module 150 determines whether frame drop and image distortion in one frame of the video are generated by using the extracted watermark signal, that is, the TDFD signal and the SDBD signal. Report the results.
  • the image distortion determination signal test module 150 stores the image frame in which the image distortion has occurred in a separate database, and reports the position where the image distortion occurs in each image frame.
  • FIG. 2 is a diagram illustrating a configuration of a watermark generator using an image distortion determination signal in an image distortion determination test automation system of a digital image output apparatus of the present invention.
  • the watermark generator 110 using the image distortion determination signal of the present invention includes a TDFD signal generator 111, an SDBD signal generator 114, and a signal randomizer 117.
  • the TDFD signal generator 111 In the watermark generator 110 configured as described above, the TDFD signal generator 111 generates a TDFD (Temporal Drop-Frame Detection) signal for detecting whether the frame is dropped on the time axis of the video. To this end, the TDFD signal generator 111 receives user information, for example, frame number data, inserted into each image frame of the reference image.
  • TDFD Temporal Drop-Frame Detection
  • the TDFD signal generator 111 encodes the input frame number data using an error correction code such as a Reed-Solomon (RS) code or a Low Density Parity Check (LDPC) code.
  • error correction code such as a Reed-Solomon (RS) code or a Low Density Parity Check (LDPC) code.
  • the reason for encoding the frame number data using an error correction code is to recover the original data even if the frame number data is distorted or deformed.
  • the TDFD signal generator 111 divides the encoded message into predetermined units and generates a TDFD signal by generating a pseudo noise code having an autocorrelation characteristic.
  • the SDBD signal generator 114 generates a spatial distortion-block detection (SDBD) signal for detecting image distortion in one frame of a video.
  • SDBD spatial distortion-block detection
  • the SDBD signal generator 114 receives a seed value for detecting image distortion in a frame, and then generates a pseudo-irregularity having an autocorrelation property. Pseudo Noise code is generated to generate SDBD signal.
  • the sum of the TDFD signal and the SDBD signal is referred to as an image distortion determination signal.
  • the magnitude of the image distortion determination signal is N ⁇ M (where N and M are natural numbers), and N and M may be the same size.
  • the size of the image distortion determination signal is preferably 16 ⁇ 16, which is a macroblock unit of a moving picture expert group (MPEG). In this case, the total length of the image distortion determination signal is 256.
  • the TDFD signal satisfies Equation 1 below. And adjust the length of the SDBD signal.
  • the signal randomizer 117 generates a unit watermark by randomizing a position at which the TDFD signal and the SDBD signal are inserted into the reference image.
  • the reason for randomizing the positions where the TDFD signal and the SDBD signal are inserted into the reference video is that the video occurs when the video compression is performed in units of macro blocks or less, for example, 8 ⁇ 8 units of DCT blocks. This is to counter the loss.
  • the position of the TDFD signal and the SDBD signal to be inserted into the reference video is limited to a specific area in the 16 ⁇ 16 image distortion determination signal, the image loss and the frame drop and the one generated when the 8 ⁇ 8 unit is compressed It is not possible to distinguish the image distortion in the image frame.
  • FIG. 3 is a diagram illustrating a configuration of a watermark inserter in the image distortion determination test automation system of the digital image output apparatus according to the present invention.
  • the watermark inserter 120 of the present invention includes a watermark embedding strength determiner 121, a tiling unit 124, and a watermark inserter 127.
  • the watermark inserter 120 may be provided with a separate device for projecting the input reference image into a region into which the watermark is to be inserted. That is, the apparatus may further include a device for converting the reference image in the frequency domain into the spatial domain or converting the reference image in the RGB color space into the YUV color space.
  • the watermark embedding strength determiner 121 determines the image adaptive watermark embedding strength to non-visually insert the watermark, ie, the image distortion determination signal, into the reference image. do.
  • the watermark embedding strength determiner 121 sequentially obtains the primary moment and the secondary moment of the reference image, and then applies the non-linear quantization function to the secondary moment to apply the watermark embedding strength. Determine.
  • the first moment of the reference image may be obtained through Equation 2 below.
  • PI (i, j) represents a reference image
  • 1stMoment (i, j) represents a primary moment of the reference image
  • the second moment of the reference image may be obtained through Equation 3 below.
  • 2ndMoment (i, j) represents the second moment of the reference image.
  • Equation 4 determining the watermark embedding strength by applying a nonlinear quantization function to the second moment is given by Equation 4 below.
  • Q () represents a nonlinear quantization function and q represents the watermark embedding strength.
  • the watermark embedding strength is set to 1 when the value is 0 to 4, and the watermark embedding strength is set to 2 when the value is 5 to 16, and the watermark embedding strength is set to 17 to 81. Decide on 3.
  • the tiling unit 124 receives the unit watermark generated by the watermark generator 110 and tiles the entire watermark of the reference image to be transmitted to the watermark insertion unit 127. That is, since the size of the unit watermark is 16 ⁇ 16, the unit watermark is extended by the size of the actual reference image to be inserted into the reference image.
  • the watermark inserting unit 127 inserts a watermark into the input reference image and applies the watermark embedding strength determined by the watermark embedding strength determining unit 121.
  • the watermark inserting unit 127 inserts a watermark into the reference image using Equation 5 below.
  • PI (i, j) represents a reference image with a watermark embedded therein
  • PI (i, j) represents an input reference image
  • R () represents a random position function
  • r Denotes an initial value for randomization.
  • FIG. 4 is a diagram illustrating a configuration of an image distortion determination signal test module in an image distortion determination test automation system of a digital image output apparatus according to the present invention.
  • the image distortion determination signal test module 150 of the present invention includes a watermark detection unit 151, an image distortion determination unit 152, a frame drop determination unit 153, a database 154, and a reporting unit ( 155, and a display unit 156.
  • the image distortion determination signal test module 150 may further include a separate device for projecting the reference image into which the watermark is inserted into the area to detect the watermark. That is, the apparatus may further include a device for converting the reference image in the frequency domain into the spatial domain or converting the reference image in the RGB color space into the YUV color space.
  • the watermark detection unit 151 predicts the embedding intensity of the watermark embedded in the reference image, and applies the predicted watermark embedding intensity to apply a watermark.
  • the watermark is detected from the inserted reference image.
  • the watermark detection unit 151 predicts the embedding strength of the watermark using Equations 2 to 4 and detects the watermark from the reference image using the modified Wiener Filter.
  • the filter coefficient value used in the Wiener Filter is adaptively modified for each pixel, the watermark signal can be accurately predicted even if there is distortion caused by various image deformations.
  • the watermark detection unit 151 detects the watermark using the modified Wiener Filter equation, as shown in Equation (6).
  • h (i, j) represents the modified Wiener Filter
  • ⁇ 2 f represents the local variance of the reference image
  • ⁇ 2 v represents the local variance of the predicted watermark embedding intensity
  • M represents the filter size
  • the image distortion determiner 152 determines whether the image is distorted in the image frame by using the SDBD signal among the watermark signals detected by the watermark detector 151.
  • the image distortion determination unit 152 detects the image frame from the SDBD signal inserted into the reference image and the reference image by using the following equation (7) for each unit watermark size (for example, 16 ⁇ 16). Measure the similarity between the signals.
  • C nc represents the degree of similarity
  • SDBD SDBD represents a signal inserted in the reference image
  • E (SDBD) is an illustration of one SDBD signal detected in the reference image.
  • the image distortion determination unit 152 determines that image distortion has occurred when the measured similarity value C nc does not exceed a threshold value.
  • the image distortion determination unit 152 may determine a threshold value in consideration of the characteristics and the length of the SDBD signal, and may adjust the precision of the image distortion determination according to the threshold value.
  • an error rate of image distortion determination is about 10 -15 when the threshold value is set to 0.8, and an error rate of image distortion determination is 10 -6 when the threshold value is set to 0.45. That's enough.
  • the frame drop determination unit 153 determines whether a frame drop is generated in the corresponding video by using a TDFD signal among the watermark signals detected by the watermark detection unit 151.
  • the frame drop determination unit 153 superimposes the TDFD signal detected by the watermark detection unit 151 in N ⁇ M units (for example, 16 ⁇ 16), and then autocorrelates the pseudo irregular signal.
  • a frame number inserted as user information is extracted by using a similarity measurement method using a decoder and a decoder for decoding an error correction code (for example, an RS code or an LDPC code).
  • the reason for using the autocorrelation property of the pseudo-random signal is that the auto-correlation property of the pseudo-random signal shows a very low similarity with respect to the pseudo-random signal group except for itself. This is because it shows very high similarity.
  • the frame drop determination unit 153 extracts a frame number from each image frame of the corresponding video, so that the frame drop determination unit 153 may determine which frame is the currently acquired image frame. You can judge.
  • the database 154 may determine whether the image distortion occurs in the image distortion determination unit 152 and the frame drop determination unit 153, that is, whether or not image distortion occurs or whether a frame drop occurs. It stores the image frame in which the image distortion occurs.
  • the reporting unit 155 generates a report for automated image quality testing by using the determination result of the image distortion determination unit 152 and the frame drop determination unit 153.
  • the reporting unit 155 creates a list by dividing whether or not the image distortion occurs in the image frame and the location of the image distortion by image frame, and whether the frame drop occurs in the video and the frame number where the frame drop has occurred. Create a report showing the information.
  • the display unit 156 displays an image quality test report generated by the reporting unit 155 on the screen so that the user can check it.
  • FIG. 5 is a flowchart illustrating a method for automating an image distortion determination test of a digital image output apparatus of the present invention.
  • the watermark generator 110 generates an image distortion determination signal capable of determining a frame drop occurring on a time axis of a video and an image distortion in one frame of the video. (S 100).
  • the TDFD signal generator 111 of the watermark generator 110 generates a Temporal Drop-Frame Detection (TDFD) signal capable of determining the frame drop
  • the SDBD signal generator 114 Spatial Distort-Block Detection (SDBD) signals for determining image distortion in one frame are respectively generated.
  • the signal randomization unit 117 of the watermark generator 110 generates a unit watermark from the image distortion determination signal by randomizing the position where the TDFD signal and the SDBD signal are inserted into the reference image (S 110). .
  • the watermark embedding strength determiner 121 of the watermark inserter 120 determines the watermark embedding strength for embedding the watermark generated by the watermark generator 110 into the reference image (S120). .
  • the watermark inserter 120 inserts the watermark into the reference image by applying the determined watermark embedding intensity (S130).
  • the watermark inserter 120 tiles the unit watermark generated by the watermark generator 110 to the entire image size and inserts the unit watermark into the reference image.
  • the image output device 130 receives and displays the reference image in which the image distortion signal is inserted as a watermark (S 140).
  • the frame capture unit 141 of the frame storage server 140 captures the reference image displayed by the image output device 130 in real time, and stores the captured image frames in the frame buffer 144. (S 150).
  • the watermark detection unit 151 of the image distortion determination signal test module 150 predicts the embedding strength of the watermark inserted into the reference image, and applies the watermark embedding strength predicted to the reference image to which the watermark is inserted.
  • the watermark is detected in step S160.
  • the image distortion determination unit 152 determines whether the image is distorted in the corresponding image frame by using the SDBD signal among the watermark signals detected by the watermark detection unit 151 (S 170).
  • the frame drop determination unit 153 determines whether a frame drop is generated in the corresponding video by using the TDFD signal among the watermark signals detected by the watermark detection unit 151 (S180).
  • the reporting unit 155 stores the determination results of the image distortion determination unit 152 and the frame drop determination unit 153 in the database 154 (S 190).
  • the reporting unit 155 generates an image distortion determination test report by using the determination results of the image distortion determination unit 152 and the frame drop determination unit 153 (S200).
  • the display unit 156 displays the image distortion determination test report generated by the reporting unit 155 on the screen so that the user can check it (S210).
  • FIG. 6 is a diagram showing the configuration of the image distortion determination test automation system of the digital image acquisition device of the present invention.
  • the image distortion determination test automation system 200 includes a watermark generator 210, a watermark inserter 220, an image acquisition device 230, and an image distortion determination signal test module 240. It is made, including.
  • the watermark generator 210 generates a signal and a synchronization signal for determining whether the autofocus and exposure functions of the image capturing apparatus 230 are normally operated. A unit watermark is generated using the same.
  • the watermark generator 210 corresponds to the AFF (Auto Focus / Exposure Detection) signal and the two-dimensional movement of the original image, which can determine whether the auto focus and exposure function of the image capturing apparatus 230 is normally operated.
  • the unit watermark is generated by randomizing the position where the AFED signal and the synchronization signal are to be inserted into the reference image.
  • the watermark inserter 220 projects the watermark to the area where the watermark is to be inserted from the external reference image, and inserts the watermark generated by the watermark generator 210 into the reference image. After the mark insertion intensity is determined, the watermark is inserted by applying the determined watermark embedding intensity.
  • the watermark inserter 220 tiles the unit watermark generated by the watermark generator 210 to the entire image size and inserts the unit watermark into the reference image.
  • the image capturing apparatus 230 is an image apparatus including a camera, and acquires an image including the watermark through a camera.
  • the watermark-embedded image is distributed as an analog image (for example, a printed output to a printer) or as a digital image (for example, a digital TV or PMP).
  • the image is acquired by the camera.
  • the image distortion determination signal test module 240 receives a reference image having a watermark embedded therein from the image acquisition apparatus 230, and projects the input reference image into an area for detecting a watermark. Perform pretreatment.
  • the image distortion determination signal test module 240 extracts the watermark signal from the reference image by predicting the watermark embedding intensity, restores the geometric deformation and the two-dimensional movement of the reference image, and then extracts the extracted AFED signal. It is determined whether the auto focus and exposure function of the image capturing apparatus 230 operates normally.
  • FIG. 7 is a diagram illustrating a configuration of a watermark generator using an image distortion determination signal in an image distortion determination test automation system of a digital image acquisition device of the present invention.
  • the watermark generator 210 using the image distortion determination signal of the present invention includes an AFED signal generator 211, a synchronization signal generator 214, and a signal randomizer 217.
  • the AFED signal generator 211 generates an AFED (Auto Focus / Exposure Detection) signal for determining whether the auto focus and exposure function of the image capturing apparatus 230 is normally operated. do.
  • AFED Auto Focus / Exposure Detection
  • the AFED signal generator 211 receives user information and encodes the received data using an error correction code such as a Reed-Solomon (RS) code or a Low Density Parity Check (LDPC) code. Encoding.
  • RS Reed-Solomon
  • LDPC Low Density Parity Check
  • the user information may be a line number of an image quality automated tesis system, and may be product-specific information of an image capturing apparatus.
  • the AFED signal generator 211 divides the encoded message into predetermined units and generates an AFED signal by generating a pseudo noise code having an autocorrelation characteristic.
  • the AFED signal has a length longer than that of the image distortion determination signal in the image distortion determination test automation system of the digital image output apparatus described with reference to FIGS. 1 to 5, for example, a signal such as 64 ⁇ 64 or 128 ⁇ 128. Should be used.
  • the length of the AFED signal must be sufficiently long.
  • an error rate increases in determining whether the auto focus and exposure function is normally operated.
  • the length of the AFED signal may be formed to be the same as the length of the image distortion determination signal and the threshold may be increased, the length of the AFED signal may be increased and the threshold may be maintained. have.
  • the synchronization signal generator 214 generates a synchronization signal corresponding to a two-dimensional movement (eg, cropping, shift, etc.) of the reference image.
  • the synchronization signal generator 214 receives a seed value for generating a synchronization signal, and then generates a pseudo noise code having an autocorrelation property to generate a pilot signal as a synchronization signal. Occurs.
  • the degree of geometric deformation of the image can be determined using the AFED signal. have.
  • the two-dimensional movement of the image cannot be identified by the AFED signal alone, and a synchronization signal is generated for this purpose.
  • the signal randomizer 217 generates a unit watermark by randomizing a position at which the AFED signal and the pilot signal are to be inserted into the reference image.
  • FIG. 8 is a diagram showing the configuration of a watermark inserter in the image distortion determination test automation system of the digital image acquisition device of the present invention.
  • the watermark inserter 220 of the present invention includes a watermark embedding strength determiner 221, a tiling unit 224, and a watermark inserter 227.
  • the watermark inserter 220 may further include a separate device for projecting the input reference image into an area to insert the watermark. That is, the apparatus may further include a device for converting the reference image in the frequency domain into the spatial domain or converting the reference image in the RGB color space into the YUV color space.
  • the watermark embedding strength determiner 221 determines an image adaptive watermark embedding intensity in order to invisibly embed the watermark in the reference image.
  • the watermark embedding strength determiner 221 sequentially obtains the primary moment and the secondary moment of the reference image, and then applies the non-linear quantization function to the secondary moment to apply the watermark embedding strength. , Which is the same as the description for Equations 2 to 4 described above.
  • the tiling unit 224 receives the unit watermark generated by the watermark generator 210, tiling the entire size of the reference image, and transmits the tile to the watermark inserting unit 227.
  • the watermark inserting unit 227 inserts a watermark into the input reference image, and applies the watermark embedding strength determined by the watermark embedding strength determining unit 221.
  • the watermark inserting unit 227 embeds the watermark in the reference image using Equation 8 below.
  • PI (i, j) represents a reference image with a watermark embedded therein
  • PI (i, j) represents an input reference image
  • R () represents a random position function
  • r Denotes an initial value for randomization.
  • FIG. 9 is a view showing the configuration of the image distortion determination signal test module in the image distortion determination test automation system of the digital image acquisition device of the present invention.
  • the image distortion determination signal test module 240 of the present invention includes a preprocessor 241, a watermark detector 242, a geometric deformation predictor 243, a geometric distortion recovery unit 244, and a synchronization signal. And a generator 245, a two-dimensional movement recovery unit 246, and an auto focus / exposure function determination unit 247.
  • the image distortion determination signal test module 240 may further include a separate device for projecting the reference image into which the watermark is inserted into the area to detect the watermark.
  • the preprocessor 241 performs a preprocessing process on the reference image into which the watermark is inserted.
  • the gain of the modified Wiener Filter is increased by lowering the image signal ratio.
  • the watermark detection unit 242 predicts the insertion strength of the watermark inserted into the reference image, and detects the watermark from the reference image into which the watermark is inserted by applying the predicted watermark embedding intensity.
  • the watermark detection unit 242 predicts the embedding strength of the watermark using Equations 2 to 4, and detects the watermark from the reference image using the modified Wiener Filter.
  • the watermark detection unit 242 detects a watermark, that is, an AFED signal and a pilot signal, using the modified Wiener Filter equation, as shown in Equation (6).
  • the geometric deformation prediction unit 243 predicts the degree of geometric deformation (eg, Scale Up, Scale Down, Rotation, etc.) of the image.
  • the geometric deformation predicting unit 243 changes the autocorrelation value in a two-dimensional space at a very high frequency (N ⁇ M) of the unit watermark when the watermark is inserted, and the watermark is changed in accordance with the geometric deformation.
  • the geometric deformation can be predicted using the fact that the original size and position can be analyzed.
  • the geometric deformation prediction unit 243 calculates a convolutional form calculation in a spatial domain in a frequency domain in order to calculate an autocorrelation of a watermark signal. Reduce the amount of computation by substituting
  • the geometric deformation predicting unit 243 measures an autocorrelation pattern using periodic characteristics of the watermark signal, and then extracts a coordinate value having a high autocorrelation.
  • FIG. 10 is a graph illustrating an autocorrelation pattern of a watermark signal. Referring to FIG. 10, it can be seen that autocorrelation values are very high at regular intervals. In this case, the geometric deformation predictor 243 detects positions of peaks having a very high autocorrelation value.
  • the geometric deformation recovery unit 244 restores the original image by performing an inverse transformation on the predicted geometric deformation.
  • the geometric deformation recovery unit 244 generates a reverse affinity parameter for the geometric deformation by using the detected pixel positions, and then reconstructs the original image using the generated inverse transform coefficient.
  • Equation 9 shows a general Affine Transform.
  • (x, y) represents the position of the original image
  • (x ', y') represents the position of the deformed image
  • a, b, c, d are affine parameters indicating rotation, zooming, etc. (Affine Parameter)
  • e and f are affine parameters indicating a linear movement distance.
  • Equation 9 two pairs of coordinates, that is, (x1, y1), (x'1, y'1) and (x2, y2), (x'2, y'2) are applied to Equation 9
  • an inverse transform parameter for reconstructing an image can be obtained. This is shown in Equation 10.
  • the geometric deformation recovery unit 244 restores the original image by using the inverse transform coefficients obtained by Equation 10.
  • the image restored by the geometric deformation recovery unit 244 is input to the watermark detection unit 242 to re-execute the watermark embedding strength prediction.
  • the comparison synchronization signal generator 245 generates a comparison pilot signal used to recover the two-dimensional movement of the image.
  • the two-dimensional movement recovery unit 246 restores the two-dimensional movement of the image by synchronizing the two-dimensional movement of the image.
  • the two-dimensional movement recovery unit 246 restores the two-dimensional movement of the image based on the similarity measurement between the comparison pilot signal generated by the comparison synchronization signal generator 245 and the pilot signal detected by the watermark.
  • the 2D movement recovery unit 246 uses a folding method of a watermark signal as a preprocessing process to perform synchronization for 2D movement, and overlaps the detected watermark signal in units of watermarks. , The correlation between the pilot signal of the watermark predicted by Equation 6 and the comparison pilot signal (Cross Correlation) is measured to recover the two-dimensional movement of the image.
  • the unit watermark is tiled and inserted into the entire image size.
  • the watermark in the predicted watermark signal A signal other than the signal (ie, an error signal) is stabilized, and the overlap of the predicted watermark signal can maximize the signal strength.
  • the two-dimensional movement recovery unit 246 measures a correlation between the predicted pilot signal and the comparison pilot signal among the watermark signals predicted by the folding technique, and then extracts a message code based on the position with the highest correlation. do.
  • a Fourier Transform is used to reduce the amount of computation by substituting a multiplication calculation in the frequency domain.
  • the autofocus / exposure determination unit 247 determines whether the autofocus and exposure function of the image capturing apparatus 230 operates normally by using the AFED signal detected as the watermark.
  • the auto focus / exposure determination unit 247 measures the similarity between the AFED signal detected in the reference image and the AFED signal inserted in the reference image in units of N ⁇ M blocks.
  • the auto focus and the image acquisition apparatus 230 It is determined that the exposure function operates normally.
  • the total number of blocks is 48.
  • the ratio of the number of blocks exceeding the threshold value is 0.8 or more (that is, 37 or more)
  • the bending phenomenon occurs. This bending phenomenon may vary depending on the camera lens and the camera function. The ratio of the number of blocks above the threshold must be adjusted.
  • the autofocus / exposure determination unit 247 determines whether the autofocus and exposure function of the image capturing apparatus 230 operates normally based on whether the user information is properly extracted from the AFED signal detected as the watermark. You may.
  • the auto focus / exposure determination unit 247 extracts a message code from the AFED signal in which the geometric deformation and the two-dimensional movement are recovered, and extracts user information from the extracted message code.
  • FIG. 11 is a flowchart illustrating a method for automating an image distortion determination test of a digital image acquisition device of the present invention.
  • the watermark generator 210 first, the watermark generator 210 generates a signal for determining whether the auto focus and exposure function of the image capturing apparatus 230 is normally operated (S300).
  • the AFED signal generator 211 of the watermark generator 210 generates an AFED (Auto Focus / Exposure Detection) signal that can determine whether the auto focus and exposure function of the image capturing apparatus 230 is normally operated.
  • the synchronization signal generator 214 generates a pilot signal corresponding to the two-dimensional movement of the image.
  • the signal randomization unit 217 of the watermark generator 210 generates a unit watermark by randomizing the position where the AFED signal and the pilot signal are to be inserted into the reference image (S 310).
  • the watermark embedding strength determiner 221 of the watermark inserter 220 determines the watermark embedding strength for embedding the watermark generated by the watermark generator 210 into the reference image (S320). .
  • the watermark inserter 220 inserts the watermark into the reference image by applying the determined watermark embedding intensity (S330).
  • the watermark inserter 220 tiles the unit watermark generated by the watermark generator 210 to the entire image size and inserts the unit watermark into the reference image.
  • the image capturing apparatus 230 acquires the reference image in which the AFED signal and the pilot signal are inserted into the watermark through the camera (S340).
  • the preprocessing unit 241 of the image distortion determination signal test module 240 performs a preprocessing process on the obtained image (S 350).
  • the watermark detection unit 242 predicts the embedding intensity of the watermark inserted into the reference image, and detects the watermark from the reference image into which the watermark is inserted by applying the predicted watermark embedding intensity (S360). .
  • the geometric deformation prediction unit 243 predicts the degree of geometric deformation (eg, Scale Up, Scale Down, Rotation, etc.) of the image (S370).
  • the geometric deformation recovery unit 244 reconstructs the original image by performing inverse transformation on the predicted geometric deformation (S380).
  • comparison synchronization signal generator 245 generates a comparison pilot signal to be used when restoring the two-dimensional movement of the image (S390).
  • the two-dimensional movement recovery unit 246 restores the two-dimensional movement of the image by synchronizing the two-dimensional movement of the image (S400).
  • the two-dimensional movement recovery unit 246 restores the two-dimensional movement of the image based on the similarity measurement between the comparison pilot signal generated by the comparison synchronization signal generator 245 and the pilot signal detected by the watermark.
  • the autofocus / exposure determination unit 247 determines whether the autofocus and exposure function of the image capturing apparatus 230 operates normally by using the AFED signal detected as the watermark (S410).

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Editing Of Facsimile Originals (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)

Abstract

L'invention concerne un système de test automatique pour évaluer une déformation d'image dans un dispositif d'image numérique. Afin d'évaluer si une image est déformée ou non, l'invention consiste à : générer un filigrane par génération d'un signal de détection temporelle d'abandon de trame (TDFD) permettant d'évaluer un abandon de trame qui se produit sur un axe temporel de vidéo et d'un signal de détection spatiale de bloc de déformation (SDB) permettant d'évaluer une déformation d'image dans une trame d'image ; introduire le filigrane généré dans une image de référence ; et extraire un signal d'évaluation de déformation d'image à partir de l'image de référence lorsque le signal de référence dans lequel le filigrane a été introduit est émis via un dispositif de sortie d'image. Afin d'évaluer si des fonctions de mise au point automatique et d'exposition fonctionnent normalement ou non dans un dispositif d'acquisition d'image, l'invention consiste également à : générer le filigrane par génération non seulement d'un signal de détection d'exposition et de mise au point automatique (AFED) permettant d'évaluer si des fonctions de mise au point automatique et d'exposition fonctionnent normalement ou non dans un dispositif d'acquisition d'image, mais également d'un signal pilote pour répondre au mouvement 2D de l'image de référence ; introduire le filigrane généré dans l'image de référence ; introduire le filigrane généré dans l'image de référence ; obtenir l'image de référence comprenant le filigrane introduit via le dispositif d'acquisition d'image ; et extraire le signal d'évaluation de déformation d'image à partir de l'image de référence.
PCT/KR2009/005991 2008-10-17 2009-10-16 Système de test automatique pour évaluer une déformation d'image dans un dispositif d'image numérique Ceased WO2010076952A2 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2008-0102028 2008-10-17
KR1020080102028A KR100902887B1 (ko) 2008-10-17 2008-10-17 디지털 영상 장치의 영상 왜곡 판단 테스트 자동화 시스템

Publications (2)

Publication Number Publication Date
WO2010076952A2 true WO2010076952A2 (fr) 2010-07-08
WO2010076952A3 WO2010076952A3 (fr) 2010-08-19

Family

ID=40982772

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2009/005991 Ceased WO2010076952A2 (fr) 2008-10-17 2009-10-16 Système de test automatique pour évaluer une déformation d'image dans un dispositif d'image numérique

Country Status (2)

Country Link
KR (1) KR100902887B1 (fr)
WO (1) WO2010076952A2 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105323579A (zh) * 2014-07-10 2016-02-10 宁波舜宇光电信息有限公司 一种手势识别模组测试机器及其测试方法
WO2016171496A1 (fr) * 2015-04-23 2016-10-27 엘지전자 주식회사 Dispositif d'émission de signal de radiodiffusion, dispositif de réception de signal de radiodiffusion, procédé d'émission de signal de radiodiffusion, et procédé de réception de signal de radiodiffusion
CN115396661A (zh) * 2022-07-29 2022-11-25 北京奇艺世纪科技有限公司 设备解码性能确定方法、装置、电子设备及存储介质

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101496854B1 (ko) * 2013-03-04 2015-03-02 황선철 디지털 핑거프린팅 방법

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR940001754B1 (ko) * 1991-05-30 1994-03-05 주식회사 금성사 음극선관내의 유해물 포집구조체
KR950004133B1 (ko) * 1992-06-16 1995-04-25 현대전자산업주식회사 모니터 화면 검사/조정장치 및 방법
KR100484675B1 (ko) * 2002-09-04 2005-04-20 삼성에스디아이 주식회사 플라즈마 디스플레이 패널 검사 시스템 및 이를 이용한검사 방법
CN100461864C (zh) * 2005-06-25 2009-02-11 华为技术有限公司 基于数字水印的多媒体视频通信客观质量评价方法
KR101273002B1 (ko) * 2006-08-24 2013-06-10 엘지전자 주식회사 데이터 방송 수신기를 테스트하는 장치 및 방법

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105323579A (zh) * 2014-07-10 2016-02-10 宁波舜宇光电信息有限公司 一种手势识别模组测试机器及其测试方法
CN105323579B (zh) * 2014-07-10 2019-01-29 宁波舜宇光电信息有限公司 一种手势识别模组测试机器及其测试方法
WO2016171496A1 (fr) * 2015-04-23 2016-10-27 엘지전자 주식회사 Dispositif d'émission de signal de radiodiffusion, dispositif de réception de signal de radiodiffusion, procédé d'émission de signal de radiodiffusion, et procédé de réception de signal de radiodiffusion
KR101805538B1 (ko) 2015-04-23 2017-12-07 엘지전자 주식회사 방송 신호 송신 장치, 방송 신호 수신 장치, 방송 신호 송신 방법, 및 방송 신호 수신 방법
KR101838084B1 (ko) 2015-04-23 2018-03-13 엘지전자 주식회사 방송 신호 송신 장치, 방송 신호 수신 장치, 방송 신호 송신 방법, 및 방송 신호 수신 방법
CN115396661A (zh) * 2022-07-29 2022-11-25 北京奇艺世纪科技有限公司 设备解码性能确定方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
WO2010076952A3 (fr) 2010-08-19
KR100902887B1 (ko) 2009-06-16

Similar Documents

Publication Publication Date Title
WO2022071647A1 (fr) Procédé et appareil d'évaluation de qualité vidéo
WO2010151049A2 (fr) Procédé et appareil de transformation automatique de vidéo tridimensionnelle
WO2021261830A1 (fr) Procédé et appareil d'évaluation de qualité de vidéo
WO2010011035A2 (fr) Appareil et procédé permettant de générer un filigrane à structure à motifs multiples, appareil d'insertion de filigrane et procédé utilisant un tel appareil, détecteur de filigrane et procédé utilisant un tel détecteur
WO2020080665A1 (fr) Procédés et appareils permettant d'effectuer un codage par intelligence artificielle et un décodage par intelligence artificielle sur une image
WO2015141925A1 (fr) Appareil photographique, son procédé de commande, et support d'enregistrement lisible par ordinateur
WO2019098778A1 (fr) Appareil d'affichage, procédé permettant de commander cet appareil d'affichage, et appareil de fourniture d'images
WO2012141392A1 (fr) Système de mesure d'un pore d'échantillon à l'aide d'un scanner et d'un échantillon de référence, et procédés associés
WO2022025423A1 (fr) Procédé et appareil d'évaluation de qualité de vidéo
EP3172894A1 (fr) Appareil de prise de vues d'images et procede de prise de vues d'images
WO2015009107A1 (fr) Procédé et appareil permettant de générer une image à afficher ayant une résolution 3k pour l'écran d'un terminal mobile
WO2017034155A1 (fr) Procédé, appareil et système pour insérer des données de tatouage numérique
WO2017005066A1 (fr) Procédé et appareil d'enregistrement d'estampille temporelle de synchronisation audio et vidéo
WO2018045682A1 (fr) Procédé et dispositif de test de synchronisation d'audio et d'image
WO2015126044A1 (fr) Procédé de traitement d'image et appareil électronique associé
WO2022005060A1 (fr) Dispositif et procédé d'élimination par filtrage d'un fichier vidéo nuisible
WO2010076952A2 (fr) Système de test automatique pour évaluer une déformation d'image dans un dispositif d'image numérique
WO2022085929A1 (fr) Appareil électronique et son procédé de commande
WO2015111833A1 (fr) Appareil et procédé pour fournir des annonces publicitaires virtuelles
WO2017082607A1 (fr) Dispositif électronique et son procédé de fonctionnement
EP3326359A1 (fr) Appareil de photographie numérique, et procédé de photographie numérique
WO2019160191A1 (fr) Dispositif et procédé de traitement d'images
EP3335155A1 (fr) Dispositif électronique et son procédé de fonctionnement
WO2017138761A1 (fr) Procédé et dispositif de codage vidéo et procédé et dispositif de décodage vidéo
WO2019093674A1 (fr) Dispositif d'affichage, dispositif de détermination de reproduction et procédé de vérification de reproduction de contenu

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09836273

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09836273

Country of ref document: EP

Kind code of ref document: A2