WO2017016294A1 - 一种视频去水印方法及装置 - Google Patents

一种视频去水印方法及装置 Download PDF

Info

Publication number
WO2017016294A1
WO2017016294A1 PCT/CN2016/083055 CN2016083055W WO2017016294A1 WO 2017016294 A1 WO2017016294 A1 WO 2017016294A1 CN 2016083055 W CN2016083055 W CN 2016083055W WO 2017016294 A1 WO2017016294 A1 WO 2017016294A1
Authority
WO
WIPO (PCT)
Prior art keywords
watermark
area
region
pixel
mask
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/CN2016/083055
Other languages
English (en)
French (fr)
Inventor
刘阳
魏伟
白茂生
蔡砚刚
祁海
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.)
Le Holdings Beijing Co Ltd
LeCloud Computing Co Ltd
Original Assignee
Le Holdings Beijing Co Ltd
LeCloud Computing 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 Le Holdings Beijing Co Ltd, LeCloud Computing Co Ltd filed Critical Le Holdings Beijing Co Ltd
Priority to US15/242,339 priority Critical patent/US20170024843A1/en
Publication of WO2017016294A1 publication Critical patent/WO2017016294A1/zh
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
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/238—Interfacing the downstream path of the transmission network, e.g. adapting the transmission rate of a video stream to network bandwidth; Processing of multiplex streams
    • H04N21/2389—Multiplex stream processing, e.g. multiplex stream encrypting
    • H04N21/23892—Multiplex stream processing, e.g. multiplex stream encrypting involving embedding information at multiplex stream level, e.g. embedding a watermark at packet level
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—Two-dimensional [2D] image generation
    • G06T11/60—Creating or editing images; Combining images with text
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
    • H04N21/4402—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving reformatting operations of video signals for household redistribution, storage or real-time display
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
    • H04N21/835—Generation of protective data, e.g. certificates
    • H04N21/8358—Generation of protective data, e.g. certificates involving watermark

Definitions

  • the embodiments of the present invention relate to the field of video technologies, and in particular, to a video removal method and apparatus.
  • the current de-watermarking technology is roughly divided into the following types.
  • the first one simply uses pixels outside the watermark area instead of all pixels inside the watermark area. Although this method can remove the watermark, the generated occlusion area will be overly obvious and the color change is large, and finally the occluded watermark area is very abrupt in the video picture, which has a great influence on the viewing effect. .
  • the second is to directly blur the watermark area, so that the watermark is blurred and cannot be seen clearly.
  • this method can ensure that the occlusion area is excessively smooth with the surrounding picture to some extent, but because it is only fuzzy processing, when the degree of blur is small, the style of the watermark will still be faintly visible, and there is a problem of incomplete removal. When the degree of blur is large, it is impossible to guarantee smoothness.
  • the embodiment of the present application proposes a new method for completely removing the watermark and ensuring a smooth transition between the watermark removal region and the surrounding region.
  • the embodiment of the present invention provides a video de-watermarking method and device, which are used to solve the problem that the watermark removal in the prior art is not complete or the image is not smoothed.
  • a video de-watermarking method includes:
  • a computer readable recording medium having recorded thereon a program for executing the above method.
  • a video watermarking apparatus includes:
  • a region selection module configured to select, in the video frame, a rectangular region including a watermark as a watermark region, configured to select a watermark extension region on the video frame and copy all pixels inside the watermark extension region, and copy the image a region formed by the pixel as a mask region, wherein the watermark extension region is selected by expanding four sides of the watermark region outward by a first pixel range;
  • a watermark filling module configured to expand the four sides of the watermark region outward by the second pixel range to form a filling region on the video frame, and fill the pixel values on the four sides of the rectangle of the filling region Describe all areas inside the watermark area to form a filled video frame;
  • a blur processing module configured to perform blur processing on the mask area
  • a superimposing module according to the position of the watermark extension area in the video frame, superimposing the masked area after blurring to a corresponding position of the filled video frame.
  • Embodiment 1 is a technical flowchart of Embodiment 1 of the present application.
  • Embodiment 2 is a technical flowchart of Embodiment 2 of the present application.
  • FIG. 3 is a schematic diagram of a watermark filling area according to an embodiment of the present application.
  • Embodiment 4 is a technical flowchart of Embodiment 4 of the present application.
  • FIG. 5 is a schematic view of a mask area according to an embodiment of the present application.
  • FIG. 6 is a schematic diagram of calculation of superposition parameters in a superposition process according to an embodiment of the present application.
  • Figure 7 is a structural diagram of a device according to Embodiment 5 of the present application.
  • FIG. 8 is a schematic diagram of padding of an application scenario according to an embodiment of the present application.
  • FIG. 9 is a schematic diagram of superposition of an application scenario in the embodiment of the present application.
  • the video watermarking method in the embodiment of the present application includes the following steps:
  • Step 101 Select a rectangular area including a watermark in the video frame as a watermark area
  • the selection of the watermark region can be done by manually selecting the rectangular region by using the mouse frame according to the position of the watermark in the video frame.
  • Step 102 Select a watermark extension area on the video frame, copy all pixels inside the watermark extension area, and use a region formed by the copied pixel as a mask area, where the watermark extension area is a The four sides of the watermark region are outwardly extended by the first pixel range;
  • Step 103 On the video frame, extend the four sides of the watermark area outward to form a filling area;
  • Step 104 Fill all areas inside the watermark area with pixel values on four sides of the padding area to form a filled video frame.
  • Step 105 Perform blurring on the mask area
  • the blurring process is a smoothing technique in image processing.
  • the Gaussian Blur also called Gaussian smoothing, used in the embodiments of the present application is a widely used processing effect in image processing software, and is generally used to reduce image noise and reduce the level of detail.
  • Step 106 Superimpose the masked area after the blurring process to a corresponding position of the filled video frame according to the position of the watermark extended area in the video frame.
  • the second pixel range is smaller than the first pixel range.
  • FIG. 2 in the video watermarking method in the embodiment of the present application, four sides of the watermark area are outwardly extended by the second pixel range to form a filling area on the video frame, and four areas of the filling area are used.
  • the pixel values on the side fill all the areas inside the watermark area to form a filled video frame.
  • the expansion and division of the watermark area is as shown in FIG. 3, and further includes the following steps:
  • Step 201 Extend the four sides of the watermark region outward to form a padding region, and further include the following three methods:
  • Step 201a extending a T side of the watermark region upward by a second pixel range while expanding a B side of the watermark region downward by a second pixel range, and using the expanded new height as a basis
  • the L side of the watermark region is extended to the left by the second pixel range, and the R side of the watermark region is extended to the right by the second pixel range based on the expanded new height to obtain the filled region of the rectangle;
  • the L side of the watermark area may be extended to the left by the second pixel range, and the R side of the watermark area may be extended to the right by the second pixel range.
  • the second width of the watermark region is extended upward by the second pixel range, and the B side of the watermark region is extended downward by the second pixel range to obtain the filled region of the rectangle;
  • the center of the rectangle of the watermark area may further change the horizontal and vertical coordinates of the four end points of the rectangle of the watermark area to the second pixel along the diagonal of the rectangle.
  • the range is obtained by the filling area, that is, the center point of the rectangle of the watermark area is assumed to be the coordinate origin, and the four end points of the rectangle of the watermark area are respectively P(x 1 , y 1 ), P(x 2 , y 2 ), P(x 3 , y 3 ), P(x 4 , y 4 ), the end point of the filling region obtained by changing the first number along the diagonal of the rectangle is P(x 1 - 1, y 1 +1), P (x 2 -1, y 2 -1), P (x 3 +1, y 3 +1), P (x 4 +1, y 4 -1).
  • the steps in the embodiment of the present application enumerate three methods for expanding the four sides of the watermark region outward to form the padding region, but the present application is not limited to the above three methods.
  • Step 202 divide the watermark region into two new rectangles in the width direction by dividing lines into a first rectangle and a second rectangle, respectively, and the widths of the first rectangle and the second rectangle are equal to the watermark region. One-half of the width;
  • Step 203 respectively dividing the first rectangle and the second rectangle into two identical edge regions and one intermediate region along a length direction, wherein each of the edge regions has one edge respectively different from a wide side of the watermark region
  • the middle region has an edge that coincides with the long side of the watermark region.
  • the side corresponding to the length of the watermark area is defined as a T side and a B side
  • a side corresponding to the width of the watermark area is an L side and an R side.
  • the three regions included in the first rectangle are a first edge region, a second edge region, and a first intermediate region, and the first edge region is a first equilateral right triangle region in the first rectangle.
  • the first equilateral right triangle region is based on one-half of the width of the watermark region And a height, a height of the first equilateral right triangle region coincides with an L side of the watermark region, and a bottom of the first equilateral right triangle region coincides with the boundary line;
  • the second edge region is a second equilateral right triangle region in the first rectangle, and the second equilateral right triangle region is bottom and high in a width of the watermark region. a height of the second equilateral right triangle region coincides with an R side of the watermark region, and a bottom of the second equilateral right triangle region coincides with the boundary line;
  • the first intermediate area is an area other than the first equilateral right triangle area and the second equilateral right triangle area in the first rectangle, and the shape of the first intermediate area is not limited to an isosceles trapezoid Or an isosceles triangle;
  • the second rectangle is divided into three regions, which are a third edge region, a fourth edge region, and a second intermediate region:
  • the third edge region is a third equilateral right triangle region in the second rectangle, and the third equilateral right triangle region is bottom and high in a width of the watermark region. a height of the third equilateral right triangle region coincides with an L side of the watermark region, and a bottom of the third equilateral right triangle region coincides with the boundary line;
  • the fourth edge region is a fourth equilateral right triangle region in the second rectangle, and the fourth equilateral right triangle region is bottom and high with a width of the watermark region. a height of the fourth equilateral right triangle region coincides with an R side of the watermark region, and a bottom of the fourth equilateral right triangle region coincides with the boundary line;
  • the second intermediate region is a region of the second rectangle except the third equilateral right triangle region and the fourth equilateral right triangle region, and the shape of the second intermediate region is not limited to an isosceles trapezoid Or an isosceles triangle;
  • 1 is the second pixel range, and for the watermark region R1, the width of the R1 region is w and the height is h. Extending one pixel outward based on the watermark region R1 to form a new rectangular region, that is, the filling region R1_1, and setting the R1_1 region
  • the four sides of the domain are L, R, T, and B, respectively, wherein the sides of L and R are h+2, and the sides of T and B are w+2;
  • the R1 region is equally divided into two portions in the height direction, and the height of each portion is h/2, as shown in FIG.
  • the two sides of the first rectangle are respectively L and R as one side, and the other side coincides with the horizontal boundary line to form two isosceles right triangles M1 with a right side and a length of h/2 And N1. That is, the first equilateral right triangle region M1 and the second equilateral right triangle region N1 are bottom and high with a width of the watermark region, and the first equilateral right triangle region is high.
  • a bottom of the first equilateral right triangle region coincides with the boundary line
  • a height of the second equilateral right triangle region and an R side of the watermark region The sides coincide
  • the bottom of the second equilateral right triangle region coincides with the boundary line.
  • Step 204 The pixel value of each pixel corresponding to the coordinate on the T side of the filling area in the first rectangle is the pixel of the pixel corresponding to the T side coordinate of the filling area rectangle. Value instead
  • the upper half of the R1 region is filled with the pixel value on the side T of the filled region R1_1.
  • the specific operation is to use the R1_1 edge T on the horizontal coordinate of each column of the upper half of the R1 region.
  • the pixel value at the same position is replaced.
  • the edge region obtained by dividing the first rectangle and the second rectangle is not limited to a triangle, and may be a trapezoidal or fan-shaped geometric shape, and the present application is not limited thereto.
  • Step 205 Fill the edge region located in the first rectangle with pixels on the side of the padding region corresponding to the broad side and the long side of the watermark region.
  • the pixel corresponding to the coordinate point of the pixel on the T side of the filling region is acquired. Taking the value as the first pixel value, acquiring the pixel point on the L side of the filling area The pixel value of the corresponding point is used as the second pixel value, and the average value of the first pixel value and the second pixel value is used as the pixel value of the pixel point;
  • the first equilateral right triangle area is the pixel value on the T side of the filling area. Therefore, the first equilateral right triangle area may be read out first when filling the first equilateral right triangle.
  • the pixel value of each pixel in the pixel is further averaged with the pixel value of the pixel corresponding to the coordinate on the L side of the filling area to update the pixel value of the pixel;
  • the pixel corresponding to the coordinate point of the pixel on the T side of the filling region is acquired. And as a third pixel value, acquiring a pixel value of the pixel corresponding point on the R side of the filling area as a fourth pixel value, and using the average of the third pixel value and the fourth pixel value as the The pixel value of the pixel;
  • the second equilateral right triangle area is the pixel value on the T side of the filling area. Therefore, the second equilateral right triangle area may be read out first when filling the second equilateral right triangle.
  • the pixel value of each pixel in the pixel is further averaged with the pixel value of the pixel corresponding to the coordinate on the R side of the filling region to update the pixel value of the pixel;
  • a plane rectangular coordinate system is established with the geometric center of the watermark region as a center.
  • the value of the same pixel on the left L of R1_1 and the ordinate of the current pixel is the same as the horizontal coordinate on the upper edge T and the current pixel.
  • the mean value of the pixel value is substituted for the current pixel value; for the region N1, the value of the pixel point on the right R of R1_1 and the ordinate of the current pixel point is the same as the mean value of the pixel value on the upper side T and the abscissa of the current pixel point. To replace the current pixel value.
  • Equation 1 Ly is the pixel value on the L side corresponding to the ordinate y, and Tx is the pixel value on the T side corresponding to the abscissa x, where -w/2 ⁇ x ⁇ -w/2+h /2&&0 ⁇ y ⁇ h/2.
  • Equation 2 The calculation process of filling in the N1 area is as shown in Equation 2.
  • Equation 2 Ry is the pixel value on the R side corresponding to the ordinate y, and Tx is the pixel value on the corresponding T side when the abscissa is x, where w/2-h/2 ⁇ x ⁇ w/2&&0 ⁇ y ⁇ h/2.
  • Step 206 The pixel value of each pixel corresponding to the coordinate on the B side of the filling area in the second rectangle is the pixel of the pixel corresponding to the B side coordinate of the filled area rectangle. Value instead
  • the lower half of the R1 region is filled with the pixel value on the side B of the filled region R1_1.
  • the specific operation is to divide the lower half of the R1 region from the triangular shaded region, and the pixel values of each column are Replace with the pixel value at the same position on the abscissa of R1_1 side B.
  • Step 207 Fill the edge region located in the second rectangle with pixels on the side of the padding region corresponding to the broad side and the long side of the watermark region.
  • the pixel corresponding to the coordinate point of the pixel on the B side of the filling region is acquired.
  • a pixel value of a coordinate corresponding point of the pixel on the L side of the filling area is obtained as a sixth pixel value, and an average value of the fifth pixel value and the sixth pixel value is used as a The pixel value of the pixel;
  • the third equilateral side is the pixel value on the B side of the fill area, therefore, this When filling the third equilateral right triangle, the pixel value of each pixel in the third equilateral triangle region may be read first, and then corresponding to the coordinates on the L side of the filling region. The pixel values of the pixels are averaged to update the pixel values of the pixels;
  • the pixel corresponding to the coordinate point of the pixel on the B side of the filling region is acquired.
  • a pixel value of a point corresponding to the coordinate of the pixel on the R side of the padding area is obtained as an eighth pixel value, and an average value of the seventh pixel value and the eighth pixel value is used as a The pixel value of the pixel.
  • the fourth equilateral right triangle area is the pixel value on the B side of the filling area. Therefore, the fourth equilateral right triangle area may be read out first when filling the fourth equilateral right triangle.
  • the pixel value of each pixel in the pixel is further averaged with the pixel value of the pixel corresponding to the coordinate on the R side of the filling region to update the pixel value of the pixel;
  • a plane rectangular coordinate system is established with the geometric center of the watermark region as a center.
  • the value of the pixel on the left L of R1_1 is the same as the vertical coordinate of the current pixel, and the value on the lower B is the same as the current pixel.
  • the mean value of the pixel value is used instead of the current pixel value; for the region N2, the value of the pixel point on the right R of R1_1 is the same as the ordinate of the current pixel point and the mean value of the pixel value on the lower side B is the same as the abscissa of the current pixel point. To replace the current pixel value.
  • Equation 3 The calculation process of filling in the M2 area is as shown in Equation 3.
  • Equation 3 Ly is the pixel value on the L side corresponding to the ordinate y, and Bx is the pixel value on the B side corresponding to the abscissa x, where -w/2 ⁇ x ⁇ -w/2+h /2&&-h/2 ⁇ y ⁇ 0.
  • Equation 4 The calculation process of filling in the N2 area is as shown in Equation 4.
  • Equation 4 Ry is the pixel value on the R side corresponding to the ordinate y, and Bx is the pixel value on the B side corresponding to the abscissa x, where w/2-h/2 ⁇ x ⁇ w/2 && -h/2 ⁇ y ⁇ 0.
  • Step 203, step 204, step 205, and step 206 of the second embodiment of the present application may also first fill the first equilateral right triangle region M1, the second equilateral right triangle region N1, the third equilateral right triangle region M2, and the first
  • the method of the watermark region other than the above four equilateral right triangle regions is still the same as the above steps, but the filling order of the embodiment of the present application is not limited thereto.
  • a watermark extension area is selected on the video frame, all pixels inside the watermark extension area are copied, and a region formed by the copied pixel is used as a mask area, wherein the watermark extension area
  • the four sides of the watermark region are outwardly extended by the first pixel range, and further includes the following steps:
  • Step 401 The first pixel range is the Nth of the smaller of the width and the height of the watermark area, and the four sides of the watermark area are expanded into the watermark by the first pixel range.
  • the extended range is that the selected first pixel range is 1/10 of the smaller value of the width and height of the watermark region.
  • Step 402 Perform blur processing on the mask region by using Gaussian blur, wherein the larger the Gaussian blur radius, the smoother the blur effect.
  • Gaussian blur uses a normal distribution (ie, "Gaussian distribution") for image processing.
  • the value of each pixel of the Gaussian template "center point” is replaced by the average of the surrounding pixel points, that is, the “intermediate point” takes “week” The average of the "point”.
  • this is a kind of “smoothing”.
  • it is equivalent to producing a “fuzzy” effect, and the "intermediate point” loses detail.
  • the calculated average refers to the calculation of the weighted average.
  • the normal distribution is a bell curve. The closer to the center, the larger the value, the farther away from the center, the smaller the value, that is, the closer to the center point. The greater the weight of the point, the smaller the weight of the point is.
  • Equation 5 The Gaussian formula for calculating the weighted average weight is as shown in Equation 5:
  • the weight of the pixel points around the current pixel in the Gaussian template cover in the weighted average can be calculated by Equation 5.
  • Step 403 The distance from the current pixel point in the mask area to the center of the rectangle of the mask area is a first distance, and the distance from the intersection point to the center of the rectangle is a second distance, and the first distance is The ratio of the second distance is a superposition parameter, wherein the intersection point is an intersection of a straight line passing through the center of the rectangle and passing the current pixel point and a rectangular boundary of the mask area.
  • the position of the current pixel in the mask region is P(x, y), and a line passing through the center O and the P point of the mask region, the line and the mask region An intersection is created on the four sides of the rectangle.
  • the distance from the point P to the center of the rectangle is d xy
  • the distance from the intersection to the center of the rectangle is d.
  • Step 404 Superimpose the padding video frame and the masked area after the blurring process, further comprising:
  • the integer power of the superposition parameter is used as the first weight, and the first weight is multiplied by the pixel value of the filled video frame as the first superposition;
  • the positional coordinates are calculated according to the distance of the pixel points to be superimposed at the current position of the mask area from the center point of the mask area.
  • Equation 7 q is a superposition parameter, q m is the first weight, 1-q m is the second weight, and y is the new pixel value obtained after the superposition.
  • the video watermarking apparatus of the embodiment of the present application includes the following modules: an area selection module 701, a watermark filling module 702, a blur processing module 703, and a superposition module 704, where:
  • the area selection module 7501 is configured to select a rectangular area including a watermark as a watermark area in the video frame, and to select a watermark extension area on the video frame and copy all pixels inside the watermark extension area. And the area formed by the copied pixels is used as a mask area, wherein the watermark extension area is selected by expanding four sides of the watermark area outward by the first pixel range;
  • the area selection module 701 is further configured to use the height of the watermark area as the height of the watermark area, extend the T side of the watermark area upward by a first number, and expand the B side of the watermark area downward.
  • the first quantity is obtained as the padding area;
  • the width of the watermark area is taken as the width of the watermark area, the L side of the watermark area is extended to the left by a first number, and the R side of the watermark area is extended to the right.
  • the first quantity is obtained by the filling area;
  • the watermark filling module 702 is configured to expand the four sides of the watermark area outwardly to form a filling area on the video frame, and use pixels on the four sides of the rectangular area of the filling area Values fill all areas within the watermark region to form a filled video frame;
  • the watermark filling module 702 is further configured to replace pixel values of each column of the first rectangle with pixel values of columns corresponding to the T-side abscissa of the filled area rectangle, where:
  • the first equilateral right triangle region and the second equilateral right triangle region in the first rectangle are bottom and high with a height of the watermark region, the first, etc.
  • the height of the right-angled triangle region coincides with the left side of the watermark region
  • the bottom of the first equilateral right-angled triangle region coincides with the boundary line
  • the height of the second equilateral right-angled triangle region is the same as
  • the sides of the right side of the watermark area coincide
  • the bottom of the second equilateral right triangle area coincides with the boundary line;
  • the watermark filling module 702 is further configured to replace the pixel value of each column of the second rectangle with the pixel value of the column corresponding to the B-side abscissa of the filled area rectangle, and further includes:
  • the pixel value of the ordinate on the corresponding point of the R side of the padding area is taken as the eighth pixel value, and the average of the seventh pixel value and the eighth pixel value is used as the pixel value of the pixel point.
  • the third equilateral right triangle region and the fourth equilateral right triangle region in the second rectangle are bottom and high with a height of the watermark region, wherein the third The height of the equilateral right triangle region coincides with the left side of the watermark region, the bottom of the third equilateral right triangle region coincides with the boundary line, and the height of the fourth equilateral right triangle region The sides of the right side of the watermark region coincide, and the bottom of the fourth equilateral right triangle region coincides with the boundary line.
  • the blur processing module 703 is configured to perform blur processing on the mask area
  • the mask region is blurred by Gaussian blur, wherein the larger the blur radius, the smoother the blur effect.
  • the superimposing module 704 is configured to superimpose the masked area after the blurring process to a corresponding position of the filled video frame according to the position of the watermark extended area in the video frame.
  • the superimposing module 704 is further configured to superimpose the watermark region and the mask region, and the superimposed size is based on the size of the mask region, and the integer power of the first superimposed parameter is used as the first weight, The first weight is multiplied by a pixel value of the filled video frame as a first superposition item;
  • the superimposed size is based on the size of the mask region after the blurring processing, and the first power value is multiplied by the pixel value of the watermark filling region by using the integer power of the first superposition parameter as the first weight value.
  • the first superposition item As the first superposition item;
  • the positional coordinates are calculated, and the first superimposed parameter and the second superimposed parameter are calculated according to a distance of a pixel point to be superimposed at a current position of the mask area from a center point of the mask area.
  • the embodiment of the present application is further described in conjunction with the actual application scenario.
  • the watermark area pixel size is 400*200.
  • the rectangle B1D1D2B2 is the watermarked area in the selected video frame, ie the watermark area.
  • the minimum value of the length and the width is 200, so that a tenth of the minimum value of 200, that is, 20, is used as the expanded first pixel range.
  • the four sides of D1D2, B1D1, B1B2, and B2D2 are respectively extended by 20 pixels to form a new region C1E1E2C2, that is, the watermark extension region has a size of 440*240.
  • the four sides of D1D2, B1D1, B1B2, and B2D2 are respectively extended by one pixel to form a new region C1C2E1E2, that is, the padding region, and the size of the padding region is 402*202.
  • the A2D1D2 A3 region is filled with the pixel values on the edge E1E2, that is, the pixel values of each column in A2D1D2 A3 are replaced by the pixel values on E1E2 whose abscissa is equal.
  • the pixel value of the Q point in FIG. 8 is replaced by the pixel value of the F2 point which is equal to the abscissa.
  • the triangular hatched portion A1A2D1 of the first rectangle is replaced by the mean value of the pixel values of the points corresponding to the horizontal and vertical coordinates of the pixel value currently required to be filled on the side C1E1 and the side E1E2.
  • the pixel value of the point S in FIG. 8 is replaced by the mean value of the pixel values of the same point S1 on the upper side E1E2 of the filling area and the same S2 on the left side C1E1 of the filling area. ;
  • the mean value of the pixel values of the points corresponding to the horizontal and vertical coordinates of the current pixel to be filled is replaced by the edges C2E2 and E1E2.
  • the pixel value of the point P in FIG. 8 is replaced by the mean value of the pixel values of the same point P1 on the upper side E1E2 of the filling area and the same point P2 on the right side C2E2 in the filling area. ;
  • the filling method is similar to the first rectangle, except that the lower side C1C2 of the filling area is used instead of the upper side E1E2 in the calculation process of the upper half area, which is not described here.
  • the fuzzy radius of the Gaussian blurring process is set to 10, and an optimal blurring effect can be achieved at this time.
  • the coordinate of the G point is (100, 50)
  • the two-dimensional Gaussian template is formed with a radius of 10 centered on G (100, 50)
  • the pixel value of G is the two-dimensional Gaussian template. The average of all pixels.
  • the second weight is 1 minus the first weight.
  • the padded video frame and the masked region after the blurring are superimposed on the basis of the processed mask region size.
  • the coordinates of the Q point in Fig. 8 are Q (100, 50), and the coordinates of the G point in Fig. 9 are G (100, 50). Therefore, when the two regions are superimposed, the pixels with the same coordinates are superimposed, that is, The first weight is multiplied by the pixel value of the Q point plus the second weight multiplied by the pixel value of the G point as a new pixel value instead of the pixel value of the coordinate position in the watermark region.
  • the video watermarking method and apparatus provided by the embodiments of the present application can completely remove the watermark and ensure a smooth transition between the watermarked area and the surrounding area.
  • Ben The new watermark removal method of the application embodiment can easily erase the extra watermark in the video for the video communicator, and can ensure the comfort of the original video viewing while removing the watermark, and has an important improvement on the viewer's viewing experience. effect.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • Editing Of Facsimile Originals (AREA)
  • Image Processing (AREA)

Abstract

本申请实施例提供一种视频去水印方法，在视频中找出水印区域并选定适当的水印扩展区域，拷贝出所述水印扩展区域的像素形成掩膜区域；在原视频帧上将水印区域进行扩展形成填充区域，利用填充区域边上的像素对所述水印区域进行填充得到填充后的视频帧；对所述掩膜区域进行模糊处理，最后叠加所述填充后的视频帧与模糊处理后的所述掩膜区域即可实现完全去除水印，同时又能保证去除水印区域与周围区域平滑过渡。

Description

一种视频去水印方法及装置
交叉引用
本申请引用于2015年7月24日递交的名称为“一种视频去水印方法及装置”的第201510441679.2号中国专利申请，其通过引用被全部并入本申请。
技术领域
本申请实施例涉及视频技术领域，尤其涉及一种视频去除水印的方法及装置。
背景技术
在多媒体技术快速发展的时代中，大量的视频文件被生产出来。在一些视频文件中，制作者为了进行广告宣传或是出于对视频的版权进行保护以及侵权行为跟踪等，往往会以水印标记这些视频。但这些被标记了水印的视频往往会降低到观众的观看体验，此外对于视频传播者来说，也不希望传播视频的同时，将其他人的水印作为视频内容的一部分进行传播。因此，在视频处理中产生了一项去除视频水印的技术。
目前的去水印技术大致分为以下几种，第一种是简单利用水印区域外侧的像素代替水印区域内部的所有像素。这种方法虽然可以去除水印，但产生的遮挡区域在与周围画面会存在过度明显，色彩变化较大等特征，最终使得遮挡后的水印区域在视频画面内十分突兀，对观看效果产生很大影响。第二种是直接对水印区域进行模糊处理，使水印模糊无法看清。这种方法虽然可以在一定程度上保证遮挡区域在与周围画面平滑过度，但由于只是模糊处理，当模糊程度小的时候，水印的样式仍会依稀可见，存在去除不彻底的问题。而模糊程度大的时候，又无法保证平滑过度。
本申请实施例提出了一种既可以完全去除水印，又能保证去除水印的区域与周围区域平滑过渡的新方法。
发明内容
本申请实施例提供一种视频去水印方法及装置，用以解决现有技术中存在的水印去除不彻底或者无法保证画面平滑过度的问题。
为达到上述目的，本申请实施例采用如下技术方案：
第一方面，一种视频去水印方法，其包括：
在视频帧中选定包含水印的矩形区域为水印区域并选定水印扩展区域，其中，所述水印扩展区域是将所述水印区域的四条侧边向外扩展第一像素范围选定的；
拷贝出所述水印扩展区域内部的所有像素，将拷贝出的像素形成的区域作为掩膜区域；
在所述视频帧上，将所述水印区域的四条侧边向外扩展第二像素范围形成填充区域，用所述填充区域的矩形的四条侧边上的像素值填充所述水印区域内部的所有区域，形成填充视频帧，其中，所述第二像素范围小于所述第一像素范围；
对所述掩膜区域进行模糊处理；
根据所述视频帧中所述水印扩展区域的位置，将模糊处理后的所述掩膜区域叠加到所述填充视频帧的相应位置。
第二方面，一种在其上记录有用于执行上述方法的程序的计算机可读记录介质。
第三方面，一种视频去水印装置，其包括：
区域选择模块，用于在视频帧中选定包含水印的矩形区域为水印区域，用于在所述视频帧上选定水印扩展区域并拷贝出所述水印扩展区域内部的所有像素，将拷贝出的像素形成的区域作为掩膜区域，其中，所述水印扩展区域是将所述水印区域的四条侧边向外扩展第一像素范围选定的；
水印填充模块，用于在所述视频帧上，将所述水印区域的四条侧边向外扩展第二像素范围形成填充区域，用所述填充区域的矩形的四条侧边上的像素值填充所述水印区域内部的所有区域，形成填充视频帧；
模糊处理模块，用于对所述掩膜区域进行模糊处理；
叠加模块，根据所述视频帧中所述水印扩展区域的位置，将模糊处理后的所述掩膜区域叠加到所述填充视频帧的相应位置。
附图概述
为了更清楚地说明本申请或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本申请实施例一的技术流程图；
图2是本申请实施例二的技术流程图；
图3是本申请实施例的填充水印区域示意图；
图4是本申请实施例四的技术流程图；
图5是本申请实施例的掩膜区域示意图；
图6是本申请实施例的叠加过程中叠加参数的计算示意图；
图7是本申请实施例五的装置结构图；
图8是本申请实施例的应用场景的填充示意图；
图9是本申请实施例的应用场景的叠加示意图。
本申请的较佳实施方式
为使本申请的目的、技术方案和优点更加清楚，下面将结合本申请中的附图，对本申请中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
实施例一
如图1所示，本申请实施例中的视频去水印方法，包括如下步骤：
步骤101：在视频帧中选定包含水印的矩形区域为水印区域；
由于在一个视频中，水印位置一般是固定的，所以对于水印区域的选定，可以通过人工根据水印在视频帧中的位置，利用鼠标框选矩形区域完成。
步骤102：在所述视频帧上选定水印扩展区域，拷贝出所述水印扩展区域内部的所有像素，将拷贝出的像素形成的区域作为掩膜区域，其中，所述水印扩展区域是将所述水印区域的四条侧边向外扩展第一像素范围选定的；
步骤103：在所述视频帧上，将所述水印区域的四条侧边向外扩展第二像素范围形成填充区域；
步骤104：用所述填充区域的四条侧边上的像素值填充所述水印区域内部的所有区域，形成填充视频帧；
步骤105：对所述掩膜区域进行模糊处理；
所述模糊处理，是图像处理中的一种平滑技术。本申请实施例中采用的高斯模糊(Gaussian Blur)，也叫高斯平滑，是在图像处理软件中广泛使用的处理效果，通常用它来减少图像噪声以及降低细节层次。
步骤106：根据所述视频帧中所述水印扩展区域的位置，将模糊处理后的所述掩膜区域叠加到所述填充视频帧的相应位置。
其中，所述第二像素范围小于所述第一像素范围。
实施例二
根据图2，本申请实施例中的视频去水印方法中，在所述视频帧上，将所述水印区域的四条侧边向外扩展第二像素范围形成填充区域，用所述填充区域的四条侧边上的像素值填充所述水印区域内部的所有区域，形成填充视频帧，对所述水印区域的扩充及划分示例如图3，进一步包括如下步骤：
步骤201：将所述水印区域的四条侧边向外扩展第二像素范围形成填充区域，进一步包括如下三种方法：
步骤201a:将所述水印区域的T侧边向上扩展第二像素范围，同时将所述水印区域的B侧边向下扩展第二像素范围，以所述扩展后的新高度为基础将所述水印区域的L侧边向左扩展第二像素范围，同时以所述扩展后的新高度为基础将所述水印区域的R侧边向右扩展第二像素范围得到矩形的所述填充区域；
步骤201b:本申请实施例中，可以先将所述水印区域的L侧边向左扩展第二像素范围，同时将所述水印区域的R侧边向右扩展第二像素范围，以所述扩展后的新宽度为基础将所述水印区域的T侧边向上扩展第二像素范围，同时将所述水印区域的B侧边向下扩展第二像素范围得到矩形的所述填充区域；
步骤201c:本申请实施例中，还可以所述水印区域的矩形的中心，将所述水印区域的矩形的四个端点的横纵坐标同时沿所述矩形的对角线向外变化第二像素范围得到所述填充区域，即假设以所述水印区域的矩形的中心点为坐标原点，所述水印区域的矩形的四个端点分别为P(x1，y1)，P(x2，y2)，P(x3，y3)，P(x4，y4)，沿所述矩形的对角线向外变化第一数量后得到的所述填充区域的端点为P(x1-1，y1+1)，P(x2-1，y2-1)，P(x3+1，y3+1)，P(x4+1，y4-1)。
本申请实施例的步骤中列举将所述水印区域的四条侧边向外扩展第二像素范围形成填充区域的三种方法，但本申请并不仅限于上述三种方法。
步骤202：用分界线将所述水印区域在宽度方向上平均分成两个新的矩形分别为第一矩形和第二矩形，所述第一矩形和所述第二矩形的宽度等于所述水印区域的宽度的二分之一；
步骤203：分别将所述第一矩形和第二矩形沿长度方向划分为两个相同的边缘区域和一个中间区域，其中，每个所述边缘区域有一条边分别于与水印区域的不同宽边重合，所述中间区域有一条边与水印区域的长边重合。本申请实施例中，规定所述水印区域的长度对应的边为T侧边和B侧边，所述水印区域的宽度对应的边为L侧边和R侧边。
所述第一矩形所包含的三个区域，分别为第一边缘区域、第二边缘区域、第一中间区域，所述第一边缘区域为所述第一矩形中的第一等边直角三角形区域，所述第一等边直角三角形区域以所述水印区域的宽度的二分之一为底 和高，所述第一等边直角三角形区域的高与所述水印区域的L侧边重合，所述第一等边直角三角形区域的底与所述分界线重合；
所述第二边缘区域为所述第一矩形中的第二等边直角三角形区域，所述第二等边直角三角形区域以所述水印区域的宽度的二分之一为底和高，所述第二等边直角三角形区域的高与所述水印区域的R侧边重合，所述第二等边直角三角形区域的底与所述分界线重合；
所述第一中间区域为所述第一矩形中除所述第一等边直角三角形区域和第二等边直角三角形区域之外的区域，所述第一中间区域的形状不限于是等腰梯形或者等腰三角形；
将所述第二矩形分为三个区域，分别为第三边缘区域、第四边缘区域、第二中间区域：
所述第三边缘区域为所述第二矩形中的第三等边直角三角形区域，所述第三等边直角三角形区域以所述水印区域的宽度的二分之一为底和高，所述第三等边直角三角形区域的高与所述水印区域的L侧边重合，所述第三等边直角三角形区域的底与所述分界线重合；
所述第四边缘区域为所述第二矩形中的第四等边直角三角形区域，所述第四等边直角三角形区域以所述水印区域的宽度的二分之一为底和高，所述第四等边直角三角形区域的高与所述水印区域的R侧边重合，所述第四等边直角三角形区域的底与所述分界线重合；
所述第二中间区域为所述第二矩形中除所述第三等边直角三角形区域和第四等边直角三角形区域之外的区域，所述第二中间区域的形状不限于是等腰梯形或者等腰三角形；
如图3示，本申请实施例中以1为所述第二像素范围，对于所述水印区域R1，设所述R1区域的宽度为w，高度为h。以所述水印区域R1为基础向外扩展一个像素，形成新的矩形区域即所述填充区域R1_1，设所述R1_1区 域的四个边分别为L、R、T、B，其中L、R的边长为h+2，T、B的边长为w+2；
将所述R1区域在高度方向平均分成两个部分，每部分高度是h/2，如图3所示。
进一步地，在所述第一矩形中，分别以第一矩形的L、R两边为一条边，另一条边与水平分界线重合，形成两个直角边长为h/2的等腰直角三角形M1和N1。即所述第一等边直角三角形区域M1和所述第二等边直角三角形区域N1以所述水印区域的宽度的二分之一为底和高，所述第一等边直角三角形区域的高与所述水印区域的L侧的边重合，所述第一等边直角三角形区域的底与所述分界线重合，所述第二等边直角三角形区域的高与所述水印区域的R侧的边重合，所述第二等边直角三角形区域的底与所述分界线重合。
步骤204：所述第一矩形中的与所述填充区域的T侧边上坐标相对应的每个像素点的像素值都用所述填充区域矩形的T侧边坐标相对应的像素点的像素值代替；
如图3所示，用所述填充区域R1_1边T上的像素值填充R1区域的上半个区域，具体操作为，将R1上半个区域每一列的像素值都用R1_1边T上横坐标相同位置处的像素值代替。
本步骤中，对所述第一矩形和所述第二矩形进行划分得到的边缘区域并不限于是三角形，也可以是梯形、扇形等几何形状，本申请并不仅限于此。
步骤205：用与水印区域的宽边和长边对应的填充区域的侧边上的像素点填充位于所述第一矩形中的所述边缘区域。
本申请实施例中，对于所述第一边缘区域，即第一等边直角三角形区域M1内的每一像素点，获取所述像素点在所述填充区域的T侧边上坐标对应点的像素值作为第一像素值，获取所述像素点在所述填充区域的L侧边上坐 标对应点的像素值作为第二像素值，用所述第一像素值和第二像素值的均值作为所述像素点的像素值；
本步骤中，由于所述第一等边直角三角形位于所述第一矩形内，而所述第一矩形已经由所述填充区域的T侧边上的像素值进行填充即所述第一等边直角三角形区域内的像素值为所述填充区域的T侧边上的像素值，因此，此处在填充所述第一等边直角三角形时也可以先读出所述第一等边直角三角形区域内的每一像素点的像素值，再与所述填充区域的L侧边上坐标相对应的像素点的像素值求均值来更新所述像素点的像素值；
本申请实施例中，对于所述第二边缘区域，即第二等边直角三角形区域N1内的每一像素点，获取所述像素点在所述填充区域的T侧边上坐标对应点的像素值作为第三像素值，获取所述像素点在所述填充区域的R侧边上坐标对应点的像素值作为第四像素值，用所述第三像素值和第四像素值的均值作为所述像素点的像素值；
本步骤中，由于所述第二等边直角三角形位于所述第一矩形内，而所述第一矩形已经由所述填充区域的T侧边上的像素值进行填充即所述第一等边直角三角形区域内的像素值为所述填充区域的T侧边上的像素值，因此，此处在填充所述第二等边直角三角形时也可以先读出所述第二等边直角三角形区域内的每一像素点的像素值，再与所述填充区域的R侧边上坐标相对应的像素点的像素值求均值来更新所述像素点的像素值；
如图3所示，以所述水印区域的几何中心为圆心建立平面直角坐标系。上半个区域两侧宽度为h/2的两个三角形阴影区域，对于区域M1，利用R1_1左边L上与当前像素点纵坐标相同的像素点的值与上边T上与当前像素点横坐标相同的像素点值的均值来替代当前像素值；对于区域N1，利用R1_1右边R上与当前像素点纵坐标相同的像素点的值与上边T上与当前像素点横坐标相同的像素点值的均值来替代当前像素值。
M1区域中填充的计算过程如公式1所示
P(x，y)＝(Ly+Tx)/2          公式1
公式1中，Ly是纵坐标为y时对应的L边上的像素值，Tx是横坐标为x时对应的T边上的像素值，其中-w/2≤x≤-w/2+h/2&&0≤y≤h/2。
N1区域中填充的计算过程如公式2所示
P(x，y)＝(Ry+Tx)/2          公式2
公式2中，Ry是纵坐标为y时对应的R边上的像素值，Tx是横坐标为x时对应的T边上的像素值，其中w/2-h/2≤x≤w/2&&0≤y≤h/2。
步骤206：所述第二矩形中的与所述填充区域的B侧边上坐标相对应的每个像素点的像素值都用所述填充区域矩形的B侧边坐标相对应的像素点的像素值代替；
如图3所示，用所述填充区域R1_1边B上的像素值填充R1区域的下半个区域，具体操作为，将R1下半个区域除三角形阴影区域之外，每一列的像素值都用R1_1边B上横坐标相同位置处的像素值代替。
步骤207：用与水印区域的宽边和长边对应的填充区域的侧边上的像素点填充位于所述第二矩形中的所述边缘区域。
本申请实施例中，对于所述第三边缘区域，即第三等边直角三角形区域M2内的每一像素点，获取所述像素点在所述填充区域的B侧边上坐标对应点的像素值作为第五像素值，获取所述像素点在所述填充区域的L侧边上坐标对应点的像素值作为第六像素值，用所述第五像素值和第六像素值的均值作为所述像素点的像素值；
本步骤中，由于所述第三等边直角三角形位于所述第二矩形内，而所述第二矩形已经由所述填充区域的B侧边上的像素值进行填充即所述第三等边直角三角形区域内的像素值为所述填充区域的B侧边上的像素值，因此，此 处在填充所述第三等边直角三角形时也可以先读出所述第三等边直角三角形区域内的每一像素点的像素值，再与所述填充区域的L侧边上坐标相对应的像素点的像素值求均值来更新所述像素点的像素值；
本申请实施例中，对于所述第四边缘区域，即第四等边直角三角形区域N2内的每一像素点，获取所述像素点在所述填充区域的B侧边上坐标对应点的像素值作为第七像素值，获取所述像素点在所述填充区域的R侧边上坐标对应点的像素值作为第八像素值，用所述第七像素值和第八像素值的均值作为所述像素点的像素值。
本步骤中，由于所述第四等边直角三角形位于所述第二矩形内，而所述第二矩形已经由所述填充区域的B侧边上的像素值进行填充即所述第四等边直角三角形区域内的像素值为所述填充区域的B侧边上的像素值，因此，此处在填充所述第四等边直角三角形时也可以先读出所述第四等边直角三角形区域内的每一像素点的像素值，再与所述填充区域的R侧边上坐标相对应的像素点的像素值求均值来更新所述像素点的像素值；
如图3所示，以所述水印区域的几何中心为圆心建立平面直角坐标系。下半个区域两侧宽度为h/2的两个三角形阴影区域，对于区域M2，利用R1_1左边L上与当前像素点纵坐标相同的像素点的值与下边B上与当前像素点横坐标相同的像素点值的均值来替代当前像素值；对于区域N2，利用R1_1右边R上与当前像素点纵坐标相同的像素点的值与下边B上与当前像素点横坐标相同的像素点值的均值来替代当前像素值。
M2区域中填充的计算过程如公式3所示
P(x，y)＝(Ly+Bx)/2         公式3
公式3中，Ly是纵坐标为y时对应的L边上的像素值，Bx是横坐标为x时对应的B边上的像素值，其中-w/2≤x≤-w/2+h/2&&-h/2≤y≤0。
N2区域中填充的计算过程如公式4所示
P(x，y)＝(Ry+Bx)/2          公式4
公式4中，Ry是纵坐标为y时对应的R边上的像素值，Bx是横坐标为x时对应的B边上的像素值，其中w/2-h/2≤x≤w/2&&-h/2≤y≤0。
本申请实施例二的步骤203、步骤204、步骤205、步骤206也可以采用先填充第一等边直角三角形区域M1、第二等边直角三角形区域N1、第三等边直角三角形区域M2和第四等边直角三角形区域N2，再填充除上述四个等边直角三角形区域之外的所述水印区域方法仍如上述步骤，但本申请实施例的填充顺序并不仅限于此。
实施例三
根据图4所示，在所述视频帧上选定水印扩展区域，拷贝出所述水印扩展区域内部的所有像素，将拷贝出的像素形成的区域作为掩膜区域，其中，所述水印扩展区域是将所述水印区域的四条侧边向外扩展第一像素范围选定的，进一步包括如下步骤：
步骤401：以所述水印区域的宽度和高度中较小值的N分之一为所述第一像素范围，将所述水印区域的四条侧边以所述第一像素范围扩展为所述水印扩展区域,其中N是正整数且保证所述第一像素范围大于1。
如图5，选定了所述水印区域R1后，以所述水印区域R1的矩形区域为基础，向外扩展一定范围形成新的矩形区域，作为所述水印扩展区域，拷贝出所述水印扩展区域内部的所有像素，形成一个新的区域作为掩膜区域。本申请实施例中，为了视觉效果，扩展范围是选定的所述第一像素范围是所述水印区域的宽度和高度中较小值的大小的1/10。
步骤402：利用高斯模糊对所述掩膜区域进行模糊处理，其中，高斯的模糊半径越大，模糊效果越平滑。
高斯模糊将正态分布(即"高斯分布")用于图像处理，将高斯模板“中心点”的每一个像素点的值都用周边像素点的平均值代替，即"中间点"取"周 围点"的平均值。在数值上，这是一种"平滑化"，在图形上，就相当于产生"模糊"效果，"中间点"失去细节。
显然，计算平均值时，高斯模板邻域的取值范围越大，"模糊效果"越强烈。因为图像都是连续的，越靠近的点关系越密切，越远离的点关系越疏远。因此，此处的计算平均值是指计算加权平均值，正态分布是一种钟形曲线，越接近中心，取值越大，越远离中心，取值越小，即距离中心点越近的点权重越大，距离越远的点权重越小。
将模糊处理后的所述掩膜区域作为R2，为了实现较好的模糊效果，本实施例中采用模糊半径r为10，其中计算加权平均权重的高斯公式如公式5所示：
Figure PCTCN2016083055-appb-000001
公式5中，x2+y2＝r2，r是模糊半径，即选择的高斯模板邻域的大小，x是距当前被模糊像素点的距离为模糊半径的像素点与当前被模糊像素点的横坐标的差，y是距当前被模糊像素点距离为模糊半径的像素点与当前被模糊像素点的纵坐标的差；其中，σ为正态分布的标准偏差，一般取值为σ＝1，通过公式5即可计算高斯模板覆盖内的当前像素点周围的像素点在所述加权平均中的权重。
步骤403：以所述掩膜区域中当前像素点到所述掩膜区域的矩形中心的距离为第一距离，以交点至所述矩形中心的距离为第二距离，以所述第一距离与所述第二距离的比值为叠加参数，其中，所述交点是经过所述矩形中心且经过所述当前像素点的直线与所述掩膜区域矩形边界的交点。
如图6所示，所述掩膜区域中当前像素点的位置为P(x，y)，过所述掩膜区域的中心O与P点作一条直线，该直线与所述掩膜区域的矩形的四条侧边上会产生一个交点。P点至矩形中心的距离为dxy，所述交点至矩形中心的距离为d，则所述叠加参数的计算如公式6所示：
q＝dxy/d         公式6
公式6中，0≤q≤1。
步骤404：叠加所述填充视频帧和模糊处理后的所述掩膜区域，进一步包括：
根据所述视频帧中所述水印扩展区域的位置，将模糊处理后的所述掩膜区域叠加到所述填充视频帧的相应位置，叠加大小以模糊处理后的所述掩膜区域的大小为基准，以叠加参数的整数次幂作为第一权值，以所述第一权值与所述填充视频帧的像素值相乘作为第一叠加项；
以1减第一权值作为第二权值，以所述第二权值与模糊处理后的所述掩膜区域的像素值相乘作为第二叠加项，以所述第一叠加项与所述第二叠加项相加得到新的像素值替代所述水印填充区域的像素值，其中，所述相加的过程是针对所述填充视频帧与处理后的所述掩膜区域中相同像素点的位置坐标进行的，所述叠加参数是根据所述掩膜区域当前位置处需叠加的像素点距所述掩膜区域中心点的距离进行计算。
叠加过程中，由于R1区域比R2区域小，最终叠加大小以R2区域的大小为基准。设R1区域及其周围的像素为PR1，R2区域的像素为PR2。叠加公式如公式7所示：
y＝qm*PR1+(1-qm)*PR2      公式7
公式7中，q为叠加参数，qm为第一权值，1-qm为第二权值，y为叠加后得到的新的像素值。
其中，m的取值根据叠加效果自行设定，本实施例中m＝10时，叠加效果最优。通过叠加，包含有水印的区域与周围区域实现了平滑地过度。
实施例四
如图7所示，本申请实施例的视频去水印装置，包括如下模块：区域选择模块701、水印填充模块702、模糊处理模块703、叠加模块704，其中：
所述区域选择模块7501，用于在视频帧中选定包含水印的矩形区域为水印区域，还用于在所述视频帧上选定水印扩展区域并拷贝出所述水印扩展区域内部的所有像素，将拷贝出的像素形成的区域作为掩膜区域，其中，所述水印扩展区域是将所述水印区域的四条侧边向外扩展第一像素范围选定的；
所述区域选择模块701进一步用于，将所述水印区域的高度作为水印区域的高度，将所述水印区域的T侧边向上扩展第一数量同时将所述水印区域的B侧边向下扩展第一数量得到所述填充区域；将所述水印区域的宽度作为水印区域的宽度，将所述水印区域的L侧边向左扩展第一数量同时将所述水印区域的R侧边向右扩展第一数量得到所述填充区域；
用分界线将所述水印区域在水平方向上平均分成两个部分，在上的一部分为第一矩形，在下的一部分为第二矩形，其中所述第一矩形的高与所述第二矩形的高相等且为所述水印区域的高度的二分之一。
所述水印填充模块702，用于在所述视频帧上，将所述水印区域的四条侧边向外扩展第二像素范围形成填充区域，用所述填充区域的矩形的四条侧边上的像素值填充所述水印区域内部的所有区域，形成填充视频帧；
所述水印填充模块702还用于，将所述第一矩形的每一列的像素值都用所述填充区域矩形的T侧边横坐标相对应的列的像素值代替，其中：
对于第一等边直角三角形区域内的每一像素点，获取所述像素点的横坐标在所述填充区域的T侧边的对应点的像素值作为第一像素值，获取所述像素点的纵坐标在所述填充区域的L侧边的对应点的像素值作为第二像素值，用所述第一像素值和第二像素值的均值作为所述像素点的像素值；
对于第二等边直角三角形区域内的每一像素点，获取所述像素点的横坐标在所述填充区域的T侧边的对应点的像素值作为第三像素值，获取所述像素点的纵坐标在所述填充区域的R侧边的对应点的像素值作为第四像素值，用所述第三像素值和第四像素值的均值作为所述像素点的像素值；
其中，所述第一矩形中的所述第一等边直角三角形区域和所述第二等边直角三角形区域以所述水印区域的高度的二分之一为底和高，所述第一等边直角三角形区域的高与所述水印区域的左侧的边重合，所述第一等边直角三角形区域的底与所述分界线重合，所述第二等边直角三角形区域的高与所述水印区域的右侧的边重合，所述第二等边直角三角形区域的底与所述分界线重合；
所述水印填充模块702还用于，将所述第二矩形的每一列的像素值都用所述填充区域矩形的B侧边横坐标相对应的列的像素值代替，进一步包括：
对于第三等边直角三角形区域内的每一像素点，获取所述像素点的横坐标在所述填充区域的B侧边的对应点的像素值作为第五像素值，获取所述像素点的纵坐标在所述填充区域的L侧边的对应点的像素值作为第六像素值，用所述第五像素值和第六像素值的均值作为所述像素点的像素值；
对于第四等边直角三角形区域内的每一像素点，获取所述像素点的横坐标在所述填充区域的B侧边的对应点的像素值作为第七像素值，获取所述像素点的纵坐标在所述填充区域的R侧边的对应点的像素值作为第八像素值，用所述第七像素值和第八像素值的均值作为所述像素点的像素值。
其中，所述第二矩形中的所述第三等边直角三角形区域和所述第四等边直角三角形区域以所述水印区域的高度的二分之一为底和高，其中所述第三等边直角三角形区域的高与所述水印区域的左侧的边重合，所述第三等边直角三角形区域的底与所述分界线重合，所述第四等边直角三角形区域的高与所述水印区域的右侧的边重合，所述第四等边直角三角形区域的底与所述分界线重合。
所述模糊处理模块703，用于对所述掩膜区域进行模糊处理；
利用高斯模糊对所述掩膜区域进行模糊处理，其中，模糊半径越大，模糊效果越平滑。
所述叠加模块704，用于根据所述视频帧中所述水印扩展区域的位置，将模糊处理后的所述掩膜区域叠加到所述填充视频帧的相应位置。
所述叠加模块704进一步用于，叠加所述水印区域和所述掩膜区域，叠加大小以所述掩膜区域的大小为基准，以第一叠加参数的整数次幂作为第一权值，以所述第一权值与所述填充视频帧的像素值相乘作为第一叠加项；
叠加大小以模糊处理后的所述掩膜区域的大小为基准，以第一叠加参数的整数次幂作为第一权值，以所述第一权值与所述水印填充区域的像素值相乘作为第一叠加项；
以1减第一权值作为第二权值，以所述第二权值与模糊处理后的所述掩膜区域的像素值相乘作为第二叠加项，以所述第一叠加项与所述第二叠加项相加得到新的像素值替代所述水印填充区域的像素值，其中，所述相加的过程是针对所述填充视频帧与处理后的所述掩膜区域中相同像素点的位置坐标进行的，所述第一叠加参数和所述第二叠加参数是根据所述掩膜区域当前位置处需叠加的像素点距所述掩膜区域中心点的距离进行计算。
应用实例
结合图8、图9所示，本实施例将结合实际应用场景来进一步阐述本申请实施例，本实施例中假设水印区域像素大小为400*200。
矩形B1D1D2B2是选定的视频帧中带水印的区域，即所述水印区域。
所述水印区域中，长和宽的最小值为200，因此以最小值200的十分之1即20作为扩展的第一像素范围。
以所述水印区域为基础，将D1D2、B1D1、B1B2、B2D2这四条侧边分别向外扩展20像素，形成新的区域C1E1E2C2，即所述水印扩展区域，其大小为440*240。
在原视频帧上拷贝出所述水印扩展区域内部的所有像素，将拷贝出的像素形成的新的区域作为掩膜区域，所述掩膜区域的大小为440*240；
以所述水印区域为基础，将D1D2、B1D1、B1B2、B2D2这四条侧边分别向外扩展一个像素，形成新的区域C1C2E1E2，即所述填充区域，填充区域的大小为402*202。
图8中，A1A4是所述水印区域的宽度方向上的分界线，将所述水印区域分为上下两部分，在上的A1 D1D2 A4为第一矩形，在下的B1A1A4 B2为第二矩形。其中，A1D1＝A1B1＝h/2。
用边E1E2上的像素值填充A2D1D2 A3区域，即A2D1D2 A3中每一列的像素值都用与其横坐标相等的E1E2上的像素值来代替。例如图8中Q点的像素值用与其横坐标相等的F2点的像素值来代替。
如图8中，第一矩形的三角形阴影部分A1A2D1，用边C1E1和边E1E2上与当前需填充像素值横纵坐标相对应的点的像素值的均值来代替。图8中的S点的像素值就用所述填充区域中上边E1E2上与其横坐标相同的S1和所述填充区域中左边C1E1上与其纵坐标相同的S2这两点的像素值的均值来代替；
对于第一矩形的三角形阴影部分A3A4D2，用边C2E2和边E1E2上与当前需填充像素值横纵坐标相对应的点的像素值的均值来代替。图8中的P点的像素值就用所述填充区域中上边E1E2上与其横坐标相同的P1和所述填充区域中右边C2E2上与其纵坐标相同的P2这两点的像素值的均值来代替；
对于第二矩形，填充方法与第一矩形相似，不同点在于利用所述填充区域的下边C1C2代替上半个区域计算过程中的上边E1E2，此处不赘述。
本申请实施例中，高斯模糊处理的模糊半径设置为10，此时能达到最优的模糊效果。如图9所示，G点的坐标为(100，50)，用以G(100，50)为中心，以10为半径形成二维高斯模板，G的像素值为所述二维高斯模板中所有像素的平均值。
如9所示，OG＝dxy，OM＝d，则叠加参数q＝dxy/d，本实施例中取叠加参数的10次幂为第一权值时叠加效果最优。第二权值为1减第一权值。
以处理后的所述掩膜区域大小为基准，叠加所述填充视频帧与模糊处理后的所述掩膜区域。对所述图8中Q点的坐标为Q(100，50)，图9中G点坐标为G(100，50)，因此，两块区域叠加时，坐标相同的像素点进行叠加，即用第一权值乘以Q点的像素值加上第二权值乘以G点的像素值作为新的像素值来代替所述水印区域中该坐标位置的像素值。
以上所描述的装置实施例仅仅是示意性的，可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性的劳动的情况下，即可以理解并实施。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到各实施方式可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件。基于这样的理解，上述技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如ROM/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行各个实施例或者实施例的某些部分所述的方法。
最后应说明的是：以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围。
工业实用性
本申请实施例提供的视频去水印方法及装置与现有技术相比，本申请实施例既可以完全去除水印，又能保证去除水印区域与周围区域平滑过渡。本 申请实施例的去除水印的新方法对于视频传播者来说，既可以轻松擦除视频中多余的水印，在去除水印的同时，能够保证原视频观看的舒适性，对于观众观看体验有重要的提升作用。

Claims (11)

  1. 一种视频去水印方法，其特征在于，该方法包括：
    在视频帧中选定包含水印的矩形区域为水印区域并选定水印扩展区域，其中，所述水印扩展区域是将所述水印区域的四条侧边向外扩展第一像素范围选定的；
    拷贝出所述水印扩展区域内部的所有像素，将拷贝出的像素形成的区域作为掩膜区域，对所述掩膜区域进行模糊处理；
    在所述视频帧上，将所述水印区域的四条侧边向外扩展第二像素范围形成填充区域，用所述填充区域的四条侧边上的像素值填充所述水印区域内部的所有区域，形成填充视频帧，其中，所述第二像素范围小于所述第一像素范围；
    根据所述视频帧中所述水印扩展区域的位置，将模糊处理后的所述掩膜区域叠加到所述填充视频帧的相应位置。
  2. 根据权利要求1所述的视频去水印方法，其特征在于，
    用所述填充区域的四条侧边上的像素值填充所述水印区域内部的所有区域，形成填充视频帧，进一步包括：
    用分界线将所述水印区域在宽度方向上平均分成两个新的矩形分别为第一矩形和第二矩形，所述第一矩形和所述第二矩形的宽度等于所述水印区域的宽度的二分之一；
    分别将所述第一矩形和第二矩形沿长度方向划分为两个相同的边缘区域和一个中间区域，其中，每个所述边缘区域有一条边分别于与水印区域的不同宽边重合，所述中间区域有一条边与水印区域的长边重合；
    用与水印区域的宽边和长边对应的填充区域的侧边上的像素点填充所述边缘区域，以及，用与水印区域的长边对应的填充区域的侧边上的像素点填充所述中间区域。
  3. 根据权利要求1所述的视频去水印方法，其特征在于，所述水印扩展区域是将所述水印区域的四条侧边向外扩展第一像素范围选定的，进一步包括：
    以所述水印区域的宽度和高度中较小值的N分之一作为所述第一像素范围，将所述水印区域的四条侧边以所述第一像素范围扩展为所述掩膜区域，其中N是正整数且保证所述第一像素范围大于1。
  4. 根据权利要求1所述的视频去水印方法，其特征在于，对所述掩膜区域进行模糊处理，进一步包括：
    利用高斯模糊对所述掩膜区域进行模糊处理，高斯的模糊半径越大，模糊效果越平滑。
  5. 根据权利要求1所述的视频去水印方法，其特征在于，根据所述视频帧中所述水印扩展区域的位置，将模糊处理后的所述掩膜区域叠加到所述填充视频帧的相应位置，进一步包括：
    叠加大小以模糊处理后的所述掩膜区域的大小为基准，以叠加参数的整数次幂作为第一权值，以所述第一权值与所述填充视频帧中的像素值相乘作为第一叠加项；
    以1减第一权值作为第二权值，以所述第二权值与模糊处理后的所述掩膜区域的像素值相乘作为第二叠加项，以所述第一叠加项与所述第二叠加项相加得到新的像素值替代所述水印填充区域的像素值，其中，所述相加的过程是针对所述填充视频帧与处理后的所述掩膜区域中位置坐标相同的像素点 进行的，所述叠加参数是根据所述掩膜区域当前位置处需叠加的像素点距所述掩膜区域中心点的距离进行计算。
  6. 根据权利要求5所述的视频去水印方法，其特征在于，所述叠加参数是由所述掩膜区域当前位置需叠加的像素点距所述掩膜区域中心点的位置进行计算，进一步包括：
    以所述掩膜区域中当前像素点到所述掩膜区域的矩形中心的距离为第一距离，以交点至所述矩形中心的距离为第二距离，以所述第一距离与所述第二距离的比值为所述叠加参数，其中，所述交点是经过所述矩形中心且经过所述当前像素点的直线与所述掩膜区域矩形边界的交点。
  7. 一种在其上记录有用于执行权利要求1-6中任一项所述方法的程序的计算机可读记录介质。
  8. 一种视频去水印装置，其特征在于，该装置包括：区域选择模块、水印填充模块、模糊处理模块、叠加模块，其中，
    所述区域选择模块，用于在视频帧中选定包含水印的矩形区域为水印区域，用于在所述视频帧上选定水印扩展区域并拷贝出所述水印扩展区域内部的所有像素，将拷贝出的像素形成的区域作为掩膜区域，其中，所述水印扩展区域是将所述水印区域的四条侧边向外扩展第一像素范围选定的；
    所述区域选择模块，还用于在所述视频帧上，将所述水印区域的四条侧边向外扩展第二像素范围形成填充区域；
    水印填充模块，用于用所述填充区域的四条侧边上的像素值填充所述水印区域内部的所有区域，形成填充视频帧；
    模糊处理模块，用于对所述掩膜区域进行模糊处理；
    叠加模块，用于根据所述视频帧中所述水印扩展区域的位置，将模糊处理后的所述掩膜区域叠加到所述填充视频帧的相应位置。
  9. 根据权利要求8所述的视频去水印装置，其特征在于，
    所述水印填充模块进一步用于用分界线将所述水印区域在宽度方向上平均分成两个新的矩形分别为第一矩形和第二矩形，所述第一矩形和所述第二矩形的宽度等于所述水印区域的宽度的二分之一；
    所述水印填充模块还用于，分别将所述第一矩形和第二矩形沿长度方向划分为两个相同的边缘区域和一个中间区域，其中，每个所述边缘区域有一条边分别于与水印区域的不同宽边重合，所述中间区域有一条边与水印区域的长边重合；用与水印区域的宽边和长边对应的填充区域的侧边上的像素点填充所述边缘区域，以及，用与水印区域的长边对应的填充区域的侧边上的像素点填充所述中间区域。
  10. 根据权利要求8所述的视频去水印装置，其特征在于，
    所述模糊处理模块还用于，利用高斯模糊对所述掩膜区域进行模糊处理，其中，高斯的模糊半径越大，模糊效果越平滑。
  11. 根据权利要求8所述的视频去水印装置，其特征在于，
    所述叠加模块进一步用于，叠加所述水印区域和所述掩膜区域，叠加大小以所述掩膜区域的大小为基准，以叠加参数的整数次幂作为第一权值，以所述第一权值与所述填充视频帧中的像素值相乘作为第一叠加项；
    以1减第一权值作为第二权值，以所述第二权值与模糊处理后的所述掩膜区域的像素值相乘作为第二叠加项，以所述第一叠加项与所述第二叠加项相加得到新的像素值替代所述水印填充区域的像素值，其中，所述相加的过程是针对所述填充视频帧与处理后的所述掩膜区域中相同像素点的位置坐标进行的，所述叠加参数是根据所述掩膜区域当前位置处需叠加的像素点距所述掩膜区域中心点的距离进行计算。
PCT/CN2016/083055 2015-07-24 2016-05-23 一种视频去水印方法及装置 Ceased WO2017016294A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/242,339 US20170024843A1 (en) 2015-07-24 2016-08-19 Method and device for removing video watermarks

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510441679.2 2015-07-24
CN201510441679.2A CN105898322A (zh) 2015-07-24 2015-07-24 一种视频去水印方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/242,339 Continuation US20170024843A1 (en) 2015-07-24 2016-08-19 Method and device for removing video watermarks

Publications (1)

Publication Number Publication Date
WO2017016294A1 true WO2017016294A1 (zh) 2017-02-02

Family

ID=57002008

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/083055 Ceased WO2017016294A1 (zh) 2015-07-24 2016-05-23 一种视频去水印方法及装置

Country Status (2)

Country Link
CN (1) CN105898322A (zh)
WO (1) WO2017016294A1 (zh)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111583087A (zh) * 2020-05-06 2020-08-25 杭州趣维科技有限公司 一种图像去水印方法
CN112667576A (zh) * 2020-12-22 2021-04-16 珠海豹趣科技有限公司 水印内容处理方法、装置、电子设备及存储介质
CN113643173A (zh) * 2021-08-19 2021-11-12 广东艾檬电子科技有限公司 水印去除方法、装置、终端设备及可读存储介质
CN114092300A (zh) * 2020-08-24 2022-02-25 中国电信股份有限公司 水印消除方法、装置和计算机可读存储介质
CN114862988A (zh) * 2022-05-12 2022-08-05 咪咕视讯科技有限公司 视频图像的遮盖方法、装置、设备和介质
CN115988259A (zh) * 2021-10-14 2023-04-18 腾讯科技（深圳）有限公司 一种视频处理方法、装置、终端、介质及程序产品

Families Citing this family (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106651805B (zh) * 2016-12-28 2020-08-25 贵州大学 基于机器学习的图像水印去除方法
CN106934780B (zh) * 2017-03-15 2019-11-01 北京思创科泰科技有限公司 一种基于图像修复的自动水印去除方法
CN107945126B (zh) * 2017-11-20 2022-02-18 杭州登虹科技有限公司 一种图像中眼镜框消除方法、装置和介质
CN108320265B (zh) * 2018-01-31 2021-09-21 努比亚技术有限公司 一种图像处理方法、终端及计算机可读存储介质
CN108257082B (zh) * 2018-02-01 2021-08-17 北京维山科技有限公司 基于固定区域去除图像手指的方法和装置
CN108550099B (zh) * 2018-04-13 2022-03-01 北京焦点新干线信息技术有限公司 一种去除图像中水印的方法和装置
CN108765349A (zh) * 2018-05-31 2018-11-06 四川斐讯信息技术有限公司 一种带有水印的图像修复方法及系统
CN108769803B (zh) * 2018-06-29 2021-06-22 北京字节跳动网络技术有限公司 带边框视频的识别方法、裁剪方法、系统、设备及介质
CN110278439A (zh) * 2019-06-28 2019-09-24 北京云摄美网络科技有限公司 基于帧间预测的去水印算法
CN110827189B (zh) * 2019-11-01 2023-12-05 山东浪潮科学研究院有限公司 一种数字图像或视频的水印去除方法及系统
CN111062903A (zh) * 2019-12-06 2020-04-24 携程计算机技术（上海）有限公司 图像水印的自动处理方法、系统、电子设备及存储介质
CN111179159B (zh) * 2019-12-31 2024-02-20 北京金山云网络技术有限公司 消除视频中目标影像的方法、装置、电子设备及存储介质
CN111263242B (zh) * 2020-02-12 2021-03-19 山东三宏信息科技有限公司 帧序列处理方法、系统和存储介质
CN111277914B (zh) * 2020-03-05 2021-03-19 山东三宏信息科技有限公司 帧序列处理方法、系统和存储介质
CN111626912B (zh) * 2020-04-09 2024-07-12 智者四海（北京）技术有限公司 水印去除方法及装置
CN111488881A (zh) 2020-04-10 2020-08-04 杭州睿琪软件有限公司 文本图像中手写内容去除方法、装置、存储介质
CN112233055B (zh) * 2020-10-15 2021-09-10 北京达佳互联信息技术有限公司 视频去痕方法和视频去痕装置
CN112235650A (zh) * 2020-10-19 2021-01-15 广州酷狗计算机科技有限公司 视频处理方法、装置、终端及存储介质
CN113744150A (zh) * 2021-08-31 2021-12-03 杭州隆埠科技有限公司 一种图像去水印的方法及装置
CN114862643A (zh) * 2022-03-31 2022-08-05 阿里巴巴（中国）有限公司 视频水印处理方法、装置、存储介质及系统
CN116055656A (zh) * 2022-12-26 2023-05-02 北京新奥特集团有限公司 基于人工智能生成内容的视频制品制作方法、装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102511047A (zh) * 2010-05-14 2012-06-20 联发科技（新加坡）私人有限公司 视频节目的字幕消除方法以及相关视频显示系统
EP2492843A1 (en) * 2011-02-22 2012-08-29 Thomson Licensing Graphic user interface for adaption of parameters for an automatic logo removal algorithm and method for adaption of these parameters
CN103618905A (zh) * 2013-12-09 2014-03-05 清华大学深圳研究生院 一种视频中的台标区域的内容补绘方法
CN104581111A (zh) * 2013-10-09 2015-04-29 奥多比公司 利用变换的目标区域填充

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
AU6131899A (en) * 1998-08-31 2000-03-21 Digital Video Express, L.P. Watermarking system and methodology for digital multimedia content
CN102930499A (zh) * 2012-11-16 2013-02-13 崔得龙 一种用于向数字图像信号加入和去除数字水印的方法
CN103116628A (zh) * 2013-01-31 2013-05-22 新浪网技术(中国)有限公司 图像文件数字签名、以及重复图像文件判断方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102511047A (zh) * 2010-05-14 2012-06-20 联发科技（新加坡）私人有限公司 视频节目的字幕消除方法以及相关视频显示系统
EP2492843A1 (en) * 2011-02-22 2012-08-29 Thomson Licensing Graphic user interface for adaption of parameters for an automatic logo removal algorithm and method for adaption of these parameters
CN104581111A (zh) * 2013-10-09 2015-04-29 奥多比公司 利用变换的目标区域填充
CN103618905A (zh) * 2013-12-09 2014-03-05 清华大学深圳研究生院 一种视频中的台标区域的内容补绘方法

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LEAWO.: "Video Local Fuzzy Function, How To Process A Local Part of A Video, Such As Facial Fuzzy Processing of A Person, Fuzzy Processing of A Logo of A Video Picture Or A Title", 3 December 2014 (2014-12-03), Retrieved from the Internet <URL:http://www.leawo.cn/space-172th.do.387561.id.read793.html> *
ZHANG, JING ET AL.: "Detecting Logo-Removal Forgery by Inconsistencies of Blur", 2009 INTERNATIONAL CONFERENCE ON INDUSTRIAL MECHATRONICS AND AUTOMATION, 31 December 2009 (2009-12-31) *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111583087A (zh) * 2020-05-06 2020-08-25 杭州趣维科技有限公司 一种图像去水印方法
CN111583087B (zh) * 2020-05-06 2023-09-05 杭州小影创新科技股份有限公司 一种图像去水印方法
CN114092300A (zh) * 2020-08-24 2022-02-25 中国电信股份有限公司 水印消除方法、装置和计算机可读存储介质
CN112667576A (zh) * 2020-12-22 2021-04-16 珠海豹趣科技有限公司 水印内容处理方法、装置、电子设备及存储介质
CN113643173A (zh) * 2021-08-19 2021-11-12 广东艾檬电子科技有限公司 水印去除方法、装置、终端设备及可读存储介质
WO2023019682A1 (zh) * 2021-08-19 2023-02-23 广东艾檬电子科技有限公司 水印去除方法、装置、终端设备及可读存储介质
CN115988259A (zh) * 2021-10-14 2023-04-18 腾讯科技（深圳）有限公司 一种视频处理方法、装置、终端、介质及程序产品
CN114862988A (zh) * 2022-05-12 2022-08-05 咪咕视讯科技有限公司 视频图像的遮盖方法、装置、设备和介质
CN114862988B (zh) * 2022-05-12 2025-10-24 咪咕视讯科技有限公司 视频图像的遮盖方法、装置、设备和介质

Also Published As

Publication number Publication date
CN105898322A (zh) 2016-08-24

Similar Documents

Publication Publication Date Title
WO2017016294A1 (zh) 一种视频去水印方法及装置
Avidan et al. Seam carving for content-aware image resizing
Huang et al. Temporally coherent completion of dynamic video
CN108876718B (zh) 图像融合的方法、装置及计算机存储介质
He et al. Rectangling panoramic images via warping
JP5058341B2 (ja) 画像処理方法
JP5692980B2 (ja) 深さマップ生成を用いた変換方法および装置
Liu et al. Bundled camera paths for video stabilization
TWI539813B (zh) 影像合成裝置及方法
US20170024843A1 (en) Method and device for removing video watermarks
CN109934772B (zh) 一种图像融合方法、装置及便携式终端
US20250095158A1 (en) Methods and Systems for Automatically Generating Backdrop Imagery for a Graphical User Interface
TWI595443B (zh) 影像處理方法、電子裝置以及非暫態電腦可讀取媒體
JP6683307B2 (ja) 多数のカメラを用いた最適の球形映像獲得方法
Liu et al. A hybrid approach for near-range video stabilization
CN107358609B (zh) 一种用于增强现实的图像叠加方法及装置
KR20200022334A (ko) 영상 처리 방법과 장치
Zhang et al. Simultaneous camera path optimization and distraction removal for improving amateur video
WO2023103684A1 (zh) 人像头发流动特效处理方法、装置、介质和电子设备
CN104822030B (zh) 一种基于图像变形的不规则视频矩形化矫正方法
JP5973572B2 (ja) 映像セグメンテーション方法
JP6528540B2 (ja) 画像処理装置、画像処理方法及びプログラム
Choi et al. Automated video editing for aesthetic quality improvement
JP6508874B2 (ja) 映像投影装置、映像投影方法、およびコンピュータプログラム
Herling Advanced real-time manipulation of video streams

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: 16829661

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16829661

Country of ref document: EP

Kind code of ref document: A1