JP2012070295A - Video processing apparatus and video processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce blurred moving images in a video.SOLUTION: A video processing apparatus is provided with: a frequency conversion part for frequency conversion of a video included in an input signal; a motion vector detection part which detects a motion of the video included in the input signal; and a blurred moving image reduction part which performs correction of reducing the blurred moving image on the video, on the basis of the result of the frequency conversion by the frequency conversion part and the motion vector detected by the motion vector detection part.

Description

  The present invention relates to a video processing technique for analyzing video characteristics and reducing moving image blur.

  In general, a video signal captured by a video camera or the like often includes a moving speed blur (exposure time) and a moving image blur that occurs at the time of imaging according to the movement of the subject or the camera.

  In order to reduce this motion blur, the motion vector is detected from continuously input video frames or recorded video frames, and the resolution is improved according to the motion of the video, thereby reducing blur. There are things to do. See, for example, US Pat.

2007-20140

  However, in the moving image blur reduction technique using the motion vector, erroneous correction of the motion vector in the direction and size different from the motion of the video, for example, the resolution is improved more than necessary due to erroneous detection of the motion vector. Therefore, it is difficult to improve the blur reduction effect so that no side effect occurs in the video. In addition, motion blur does not always exist in the moving image part. Even if the blur is corrected, the blur due to the focus difference of the camera or the like is corrected by simply reducing the blur. The feeling is lost.

  The present invention has been made in view of the above problems, and an object of the present invention is to appropriately suppress image blur and correct image quality.

  In order to solve the above problems, an embodiment of the present invention includes a frequency conversion unit that converts a frequency of a video included in an input signal, a motion vector detection unit that detects a motion of the video included in the input signal, What is necessary is just to comprise so that the moving image blur reduction part which performs the moving image blur reduction correction | amendment with respect to the said image | video based on the frequency conversion result by the said frequency conversion part and the motion vector which the said motion vector detection part detected may be provided.

  According to the present invention, it is possible to suitably suppress moving image blur and correct image quality.

It is explanatory drawing of an example of a structure of the image processing apparatus which concerns on Example 1 of this invention. It is explanatory drawing of an example of a structure of the image processing apparatus which concerns on Example 3 of this invention. It is explanatory drawing of an example of a structure of the image processing apparatus which concerns on Example 4 of this invention. It is explanatory drawing of the image processing which concerns on the Example of this invention. It is explanatory drawing of the image processing which concerns on the Example of this invention. It is explanatory drawing of the image processing which concerns on the Example of this invention. It is explanatory drawing of the image processing which concerns on the Example of this invention. It is explanatory drawing of the image processing which concerns on the Example of this invention. It is explanatory drawing of an example of the frequency distribution which concerns on the Example of this invention. It is explanatory drawing of an example of the frequency distribution which concerns on the Example of this invention. It is explanatory drawing of an example of the frequency distribution which concerns on the Example of this invention. It is explanatory drawing of an example of the frequency characteristic determination process by the frequency which concerns on the Example of this invention. It is explanatory drawing of an example of the frequency characteristic determination process which concerns on the Example of this invention. It is explanatory drawing of an example of the frequency characteristic determination process which concerns on the Example of this invention. It is explanatory drawing of an example of the motion vector detection process which concerns on the Example of this invention. It is an example of the flowchart of the moving image blur reduction process which concerns on Example 1 of this invention. It is an example of the flowchart of the moving image blur reduction process which concerns on Example 2 of this invention. It is explanatory drawing of an example of a structure of the moving image blur reduction part which concerns on Example 1 of this invention. It is explanatory drawing of an example of a structure of the moving image blur reduction part which concerns on Example 2 of this invention. It is explanatory drawing of an example of a structure of the moving image blur reduction part which concerns on Example 4 of this invention. It is explanatory drawing of the animation blur correction intensity | strength which concerns on the Example of this invention. It is explanatory drawing of the animation blur correction range which concerns on the Example of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that in the drawings for describing the embodiments, the same components are denoted by the same reference symbols in principle, and the repetitive description thereof will be omitted.

  First, an example of a video processing apparatus according to Embodiment 1 of the present invention will be described with reference to the configuration diagram of FIG.

  The video processing apparatus according to the present embodiment includes, for example, a frequency conversion unit 3, a motion vector detection unit 5, and a moving image blur reduction unit 7. Here, the frequency conversion unit 3 and the motion vector detection unit 5 may constitute one block called the detection unit 2.

  The input video signal 1 may be, for example, a video signal after decoding using a broadcast signal received by a built-in tuner as an input signal, or a video signal after image quality correction such as noise reduction processing or contrast processing. The output video signal 8 may be an input signal to a video display device such as an LCD panel, an input signal to a video storage device such as an HDD or a BD (Blu-ray Disc), and an input signal to the image quality correction signal. But you can.

  The frequency converter 3 uses the input video signal 1 as an input signal, for example, frequency-converts the video in an 8 × 8 pixel or 16 × 16 pixel block by FFT (Fast Fourier Transform), DCT (Discrete Cosine Transform), etc., A spectrum signal 4 having the block size frequency band quantized in the block is output.

  The motion vector detection unit 5 uses the input video signal 1 as an input signal to determine the motion of the video and outputs the vector, that is, the motion vector signal 6. In the case of a video processing apparatus that performs motion compensation type frame rate conversion, the motion vector detection unit 5 may be shared with the block of the motion vector detection unit at the time of motion compensation type frame rate conversion.

  For example, there are a gradient method, a phase difference limited correlation method, a block matching method, and the like as a method for obtaining a motion vector of an image. Hereinafter, the block matching method will be described as an example, but other motion vector detection methods can also be applied to the present invention.

  FIG. 7 is a diagram for explaining a vector detection method based on the block matching method. The block configuration may be, for example, a block size such as the above-mentioned 8 × 8 pixels or 16 × 16 pixels, but here, in order to simplify the description, a 3 × 3 pixel block will be described.

  In the detection of the motion vector, the search range is, for example, a range of 21 × 21 pixels centered on the same position as the pixel to be corrected in the current frame. Here, in the search range, the 3 × 3 pixel block centering on the pixel to be corrected in the 1v delay frame and the 3 × 3 pixel block having the lowest absolute value of the luminance difference are detected and become the center of the detected block. A relative position with respect to the pixel is detected as a motion vector.

Further, as expressed in Equation 1, (ξ, η) when the luminance difference absolute value Sad (x, y) of each block is minimum in the search range is a motion vector.

  Next, the operation of the moving image blur reduction unit 7 will be described with reference to FIG. The moving image blur reduction unit 7 includes, for example, a frequency characteristic determination / correction intensity calculation unit 71, a motion region detection / correction range calculation unit 72, and a moving image blur reduction processing unit 76.

For the sake of simplicity, the motion blur reduction process will be described using a one-dimensional DoG (Difference of Gaussian) as an example. DoG is expressed by the difference of Gaussian as shown in Equation 2, and the characteristic is as shown in FIG.

  The motion region detection / correction range calculation unit 72 receives the motion vector signal 6 as an input signal, and detects a motion region based on the input motion vector. Furthermore, assuming that the magnitude of the input motion vector is the correction width and the direction of the input motion vector is the correction direction, the correction range when the area with the motion is moving image blur is as shown in FIG. I want. Here, the correction direction is further replaced with a correction angle 73a, and the correction width is converted into a dispersion value 74a, which is used as a parameter for the DoG. Note that the conversion from the correction width to the variance value can be obtained by a proportional expression of the correction width having a fixed coefficient, for example. However, when the variance value is 0, DoG is also 0.

  The frequency characteristic determination / correction intensity calculation unit 71 uses the spectrum signal 4 as an input signal, determines the frequency characteristic from the input spectrum signal, detects a region having the frequency characteristic of moving image blur, and according to the detected moving image blur amount Thus, the correction strength is calculated and the gain signal 75a is output.

  The moving image blur reduction processing unit 76 uses the correction angle 73a calculated by the motion region detection / correction range calculation unit 72, the dispersion value 74a, and the gain signal 75a acquired by the frequency characteristic determination / correction intensity calculation unit 71 according to the following equation (2). A blur reduction process using the DoG filter is performed.

  Here, in order to reduce only the moving image blur by the moving image blur reduction processing unit 76, the frequency characteristic determining process and the moving image blur amount calculating process in the frequency characteristic determining / correcting intensity calculating unit 71 are important. An example of the frequency characteristic determination process and the moving image blur amount calculation method will be described with reference to FIGS. 4, 5, and 6. 5A to 5C are diagrams in which the vertical axis represents the frequency component in the vertical direction, the horizontal axis represents the frequency component in the horizontal direction, and the hatched portion represents the spectrum distribution. FIG. 5A shows a frequency distribution of a general still image including a texture. The image having such a frequency distribution corresponds to, for example, an image such as a tree in the image of FIG. 4A. FIG. 5b is an image in which the spectrum is distributed in the center, that is, the low frequency component. An image having such a frequency distribution corresponds to, for example, the frequency distribution of balloons, the sky in the background, and the wall of the house in the image of FIG. 4A. In the spectrum of FIG. 5c, the vertical direction is distributed in the high frequency component and the horizontal direction is distributed in the low frequency component. An image having such a frequency distribution corresponds to, for example, a horizontal stripe pattern on a roof or a moving image blur of a bus extending in the horizontal direction by horizontal movement in the image of FIG. 4A.

  6A to 6C represent regions obtained by dividing the frequency band of the quantized frequency space shown in FIGS. 5A to 5C, and the bold rectangles are ranges for determining the frequency distribution. For example, a threshold value for a spectrum value is provided for determination, and if the spectrum value is equal to or greater than the threshold value, it is counted as being distributed in the corresponding frequency band. In addition, you may use the logarithm of a spectrum at the time of determination.

  When the frequency distribution of FIG. 5a is determined by the rectangle of FIG. 6b, the spectrum is counted in the frequency band division region outside the range of the rectangle, but in the case of the rectangle of FIG. The region is included inside the rectangle of FIG. 6a and is not counted in the frequency band division region outside the rectangle. Thus, it is determined that there is a match in the case of the smallest rectangular region that includes all frequency band division regions whose spectral values are equal to or greater than the threshold value. In addition, although the said match determination was demonstrated as the case of the minimum rectangular area including all the frequency band division areas whose spectrum value is more than the said threshold value, among the frequency band division areas whose spectrum value is more than the said threshold value, The minimum rectangular area including a predetermined ratio or more may be used. For example, the predetermined ratio may be set according to the design, such as 80% or 90%.

  Since moving image blur is one of images distributed in one direction as shown in FIG. 5c, a plurality of moving image blur determination patterns as shown in FIG. 6c distributed in one direction are created, and pattern matching is performed by the above-described matching determination. By doing so, it is possible to detect a region having the frequency characteristics of moving image blur, that is, a candidate region where moving image blur may exist. As shown in FIG. 6c, the moving image blur amount can be calculated according to the width of the rectangle of the determination region used for pattern matching. Thereby, the gain signal 75a may be determined.

  Further, when the surrounding count values are calculated as in the broken-line rectangle in FIG. 6c, when the horizontal blur is weak, many spectra are counted in the broken-line rectangle. The correction intensity can be adjusted by the distribution amount.

  A flow of moving image blur reduction processing in the video processing apparatus according to the first embodiment described above will be described with reference to FIG. As shown in FIG. 8a, a moving area is detected from the magnitude and direction of the motion vector, and a correction range is calculated when the moving area is moving image blur (S11). FIG. 4b shows the detection result of the moving region with respect to FIG. 4a. In FIG. 4b, the motion area is shown as white and the stationary area is shown as black. Further, a region having the frequency characteristic of moving image blur is detected by frequency characteristic determination with respect to the frequency distribution, and a correction intensity corresponding to the moving image blur amount is calculated (S12). Among the areas with motion detected in S11, the blur reduction process of the correction intensity calculated in S12 is performed on the area having the frequency characteristics of the moving image blur detected in S12 using the correction range calculated in S11. Thus, the blur reduction correction can be performed on the range where the moving image blur exists as shown in FIG. 4C (S13).

  In the moving image blur reduction process in the video processing apparatus according to the first embodiment described above, a motion area in a video is detected from a motion vector detected from an input video, and the motion area is a motion blur. Calculate the correction range (correction direction, correction width), determine the frequency characteristics for the frequency conversion result of the input video, detect the area that has the frequency characteristics of the motion blur and calculate the correction strength. Using the calculated information of the correction range and the correction intensity, the blur reduction process is performed on the area having the frequency characteristic of the detected moving image blur. Thereby, a moving image blur area | region can be correct | amended suitably.

  Next, an example of a video processing apparatus according to the second embodiment of the present invention will be described.

  The video processing apparatus according to the present embodiment has the configuration shown in FIG. For example, a frequency conversion unit 3, a motion vector detection unit 5, and a moving image blur reduction unit 7 are included. However, in the second embodiment, the configuration of the moving image blur reduction unit 7 is different from that of the first embodiment and has the configuration shown in FIG. Since the configuration other than the moving image blur reduction unit 7 operates in the same manner as in the embodiment, the description thereof is omitted.

  The moving image blur reduction unit 7 according to the second embodiment includes a frequency characteristic determination / correction range calculation unit 77, a correction strength calculation unit 78, and a moving image blur reduction processing unit 76.

  As in the first embodiment, the moving image blur reduction process will be described using a one-dimensional DoG (Difference of Gaussian) as an example.

  The frequency characteristic determination / correction range calculation unit 77 receives the spectrum signal 4 and determines the frequency characteristic to detect a region having the frequency characteristic of moving image blur, and perform correction when performing the moving image blur reduction process on the region. A correction angle 73b indicating a range and a dispersion value 74b are output. Specifically, the following processing is performed.

  As described in the first embodiment, the moving image blur is an image distributed in one direction as shown in FIG. 5C, and therefore, moving image blur can be detected by creating a pattern distributed in one direction and performing pattern matching. For example, when the pattern of FIG. 5c is rotated every 30 degrees and pattern matching is performed, the direction of the blur is known with an accuracy of 30 degrees. Further, the blur spread becomes larger as the elliptical pattern as shown in FIG. Using this property, the correction range is obtained.

  In addition, for the detection of the correction direction, a pattern for a necessary angle may be prepared in advance, even by detection by the rotation.

  Also, the correction width can be calculated based on pattern matching by preparing a plurality of elliptical patterns having different widths for which a corresponding correction width is set in advance.

  As described above, after calculating the correction direction and the correction width, as shown in FIG. 12B, the correction direction is used as a correction angle, and the correction width is converted into a dispersion value to be a DoG parameter.

  The correction intensity calculation unit 78 uses the motion vector signal 6 as an input signal, and outputs, for example, a gain value proportional to the magnitude of the motion vector as the gain signal 75b.

  A flow of moving image blur reduction processing in the video processing apparatus according to the second embodiment described above will be described with reference to FIG. As shown in FIG. 8b, a region having the frequency characteristic of moving image blur is detected from the spread and direction of the frequency distribution, and a correction range for the detected region having the frequency characteristic of moving image blur is calculated (S21). FIG. 4d shows the detection result of the area having the frequency characteristic of the motion blur with respect to FIG. 4a. In FIG. 4d, the area | region area | region which has the frequency characteristic of a moving image blur is shown as white, and another area | region is shown as black. Further, the correction strength corresponding to the amount of motion is calculated from the motion vector (S22). In S21, as shown in FIG. 4d, in addition to moving image blur, a region having a horizontal stripe pattern on the roof is also detected in the region having moving image blur frequency characteristics. A region having no motion has a correction strength of 0 in the calculation of the correction strength in S22. Therefore, by performing the blur reduction of the correction intensity calculated in S22 using the correction intensity calculated in S21 for the area having the frequency characteristic of the moving image blur detected in S21 or a frequency characteristic equivalent to the moving image blur, As shown in FIG. 4e, the blur reduction correction can be performed on the range where the moving image blur exists (S23).

  In the moving image blur reduction process in the video processing apparatus according to the second embodiment described above, the frequency characteristic determination is performed on the frequency conversion result of the input video to detect the region having the moving image blur frequency characteristic, and the region having the motion Calculates the correction range (correction direction, correction width) when the video is blurred, calculates the correction strength based on the motion vector detected from the input video, and the correction range and correction strength information calculated by the above processing Is used to perform a blur reduction process on a region having the frequency characteristics of the detected moving image blur. Thereby, a moving image blur area | region can be correct | amended suitably.

  Next, an example of a video processing apparatus according to the third embodiment of the present invention will be described. Will be described with reference to the block diagram of FIG.

  The video processing apparatus according to the present embodiment has, for example, the configuration shown in FIG. Specifically, for example, it includes a frequency conversion unit 3, a motion vector detection unit 5, a moving image blur reduction unit 7, and the like. The moving image blur reduction unit 7 may have either the configuration of the first embodiment or the configuration of the second embodiment. Here, the frequency conversion unit 3 and the motion vector detection unit 5 may constitute one block called the detection unit 2.

  Hereinafter, differences from the first embodiment and the second embodiment will be described.

The frequency conversion unit 3 uses frequency conversion such as FFT and performs frequency conversion on, for example, a 32 × 32 pixel range of the current frame and the 1V delay frame. The pixel range is a range in which a motion vector is detected.
The frequency-converted spectrum signal 4 becomes an input signal for motion vector detection.

The motion vector detection unit 5 obtains a motion vector (ξ, η) based on a ratio of frequency distributions after frequency conversion such as FFT as shown in Equation 3.

  By configuring the detection unit as described above, the video processing apparatus according to the third embodiment can perform frequency conversion used for motion vector detection in addition to the effects of the video processing apparatuses according to the first and second embodiments. Is also used for moving picture blur reduction, which eliminates the need for motion vector detection by block matching and has the effect of reducing the circuit scale.

  Next, an example of a video processing apparatus according to the fourth embodiment of the present invention will be described. The video processing apparatus of the present embodiment has, for example, the configuration shown in FIG.

  In the first embodiment of the present invention, an example is shown in which a correction range is calculated based on the magnitude and direction of a motion vector for an input video signal, and a correction strength is calculated by frequency characteristic determination regarding a frequency distribution (hereinafter, method 1) . The method 1 is suitable for, for example, a scroll video in which a motion vector can be detected uniformly on the screen.

  On the other hand, in the second embodiment of the present invention, an example is shown in which the correction range is calculated from the frequency distribution by pattern matching with respect to the blur region in a certain direction, and the correction strength is calculated from the motion amount (hereinafter, method 2). The method 2 is less sensitive to the correction range than the method 1, but there are few erroneous corrections with respect to the detected moving image blur region. Therefore, the method 2 is suitable for, for example, an image with little movement or an image with fine movement.

  On the other hand, in the fourth embodiment of the present invention, the method 1 and the method are adaptively adapted according to external information such as content-related information and program-related information, and signal processing information such as movement and frequency distribution. This is a method that enables switching to 2. The details will be described below.

  As shown in FIG. 3, the video processing apparatus according to the fourth embodiment includes, for example, a detection unit 2, a moving image blur reduction unit 15, an analysis unit 11, and the like. The detection unit 2 may have any of the configurations of the first and second embodiments and the configuration of the third embodiment.

  The detection unit 2 uses the input video signal 1 as an input signal, and outputs a spectrum signal 4 by frequency conversion and a motion vector signal 6 by motion vector detection.

  The analysis unit 11 receives the spectrum signal 4 and the motion vector signal 6 as input signals, analyzes the characteristics of video motion, frequency distribution, etc. in one frame, and outputs a control signal for the processing method of the motion blur reduction unit 15 To do. For example, when a motion vector histogram in the video screen is generated from the motion vector signal 6 and a frequency histogram is generated from the spectrum signal 4 and the motion vector histogram indicates that the video is scrolling, or the frequency component in the frequency histogram If the signal is biased in one direction, the method 1 may be selected. Otherwise, the selection signal 9 for selecting the method 2 may be output.

  As another output example of the control signal, a frequency histogram is generated from the spectrum signal 4, and when the high frequency component is distributed in the frequency histogram, the correction strength is lowered, and when the frequency histogram is distributed in the low frequency component, the correction strength is decreased. You may output the gain signal 10 which performs operation, such as raising.

  As another output example of the control signal, the analysis unit 11 selects a motion blur reduction processing method based on content-related information such as genre information of the content program, program-related information, and the like, and outputs a selection signal. Also good. For example, in the case of a sports program, the selection signal 9 for selecting the method 1 may be output, and in the case of a song program, the selection signal 9 for selecting the method 2 may be output. In this case, the analysis unit 11 may use the information 14 as an input signal, and the information 14 may include content related information such as genre information of the content broadcast together with the input video signal 1 and program related information.

  Even if the gain signal 10 is output in accordance with the resolution of the video, for example, the correction strength is lowered for full HD video such as BD or terrestrial digital broadcasting, and the correction strength is increased for SD video such as DVD. Good. The analysis unit 11 may use the information 14 as an input signal and include information on the resolution of the video in the information 14. The configuration of the moving image blur reduction unit 15 of the video processing apparatus according to the fourth embodiment of the present invention will be described with reference to FIG. The moving image blur reduction unit 15 receives as input control signals such as an input video signal 1, a spectrum signal 4, a motion vector signal 6, a selection signal 9 for selecting method 1 or method 2, and a gain signal 10 for adjusting the overall correction strength, An output video signal 8 subjected to moving image blur reduction is output.

  The frequency characteristic determination / correction range calculation / correction intensity calculation unit (frequency) 79 receives the spectrum signal 4 and outputs a gain signal 75a used in the processing of the method 1. In order to output the gain signal 75a in the frequency characteristic determination / correction range calculation / correction intensity calculation unit (frequency) 79, the correction intensity calculation processing is the gain signal by the frequency characteristic determination / correction intensity calculation unit 71 described in the first embodiment. Since 75a is output, the process is the same as the correction intensity calculation process, and a description thereof will be omitted. Further, the frequency characteristic determination / correction range calculation / correction intensity calculation unit (frequency) 79 outputs a correction angle 73b and a variance value 74b used for the method 2 processing. The calculation process of the correction angle 73b and the dispersion value 74b in the frequency characteristic determination / correction range calculation / correction intensity calculation unit (frequency) 79 is performed in the correction angle 73b by the frequency characteristic determination / correction range calculation unit 77 described in the second embodiment. Since it is the same as the calculation process of the dispersion value 74b, description is abbreviate | omitted.

  The motion region detection / correction range calculation / correction intensity calculation unit (vector) 80 receives the vector signal 6 and outputs a correction angle 73a and a variance value 74a used in the processing of method 1. The calculation process of the correction angle 73a and the variance value 74a in the motion region detection / correction range calculation / correction intensity calculation unit (vector) 80 is the correction angle 73a performed by the motion region detection / correction range calculation unit 72 described in the first embodiment. Since this is the same as the calculation process of the variance 74a, the description is omitted. Further, the motion region detection / correction range calculation / correction intensity calculation unit (vector) 80 outputs a gain signal 75b used for the method 2 processing. In order to output the gain signal 75b in the motion region detection / correction range calculation / correction intensity calculation unit (vector) 80, the correction intensity calculation process outputs the gain signal 75b by the correction intensity calculation unit 78 described in the second embodiment. Therefore, since it is the same as the calculation process of the correction strength, the description is omitted.

  The selector unit selects the signals 73a to 75a used for the processing of the method 1 and the signals 73b to 75b used for the processing of the method 2 according to the input selection signal 9, and selects the parameters of the moving image blur reduction processing, the correction angle 73, The variance value 74 and the gain signal 75 are output to the moving image blur reduction processing unit 76. Here, if the gain signal 75 is multiplied by the input gain signal 10 and input to the moving image blur reduction process, the correction intensity according to the analysis result of the analysis unit 11 can be adjusted.

  In the moving image blur reduction process in the video processing apparatus according to the fourth embodiment described above, a plurality of moving image blurs are performed according to video motion characteristics, frequency distribution characteristics, content related information and program related information, or video resolution information. By switching the correction characteristics and the correction amount of the reduction process, it is possible to more suitably reduce the motion blur.

DESCRIPTION OF SYMBOLS 1 Input video signal 2 Detection part 3 Frequency conversion part 4 Spectral signal 5 Motion vector detection part 6 Motion vector signal 7 Moving image blur reduction part 8 Output video signal 9 Selection signal 10 Gain 11 Analysis part 14 Information 15 Moving picture blur reduction part 71 Frequency characteristic Judgment / correction range calculator (frequency)
72 Motion region detection / correction intensity calculation unit (vector)
73 Correction angle 73a Correction angle (vector)
73b Correction angle (frequency)
74 Variance value 74a Variance value (vector)
74b Dispersion value (frequency)
75 Gain 75a Gain (frequency)
75b Gain (vector)
76 Movie blur reduction processing unit 77 Frequency characteristic determination / correction strength calculation unit (frequency)
78 Correction range calculator (vector)
79 Frequency characteristics determination / correction range calculation / correction intensity calculation unit (frequency)
80 Motion region detection / correction range calculation / correction intensity calculation unit (vector)

Claims (20)

A frequency converter that converts the frequency of the video included in the input signal;
A motion vector detection unit for detecting the motion of the video included in the input signal;
An image processing apparatus comprising: a moving image blur reduction unit that performs moving image blur reduction correction on the image based on a frequency conversion result by the frequency conversion unit and a motion vector detected by the motion vector detection unit.   The moving image blur reduction unit calculates a correction range, a correction direction, and a correction strength for blur reduction correction based on a frequency conversion result by the frequency conversion unit and a motion vector detected by the motion vector detection unit, and calculates the calculated The video processing apparatus according to claim 1, wherein the parameter is used for blur reduction correction. The video processing apparatus according to claim 1, wherein:
The moving image blur reduction unit determines a frequency characteristic of a frequency conversion result by the frequency conversion unit, and determines an image region having a frequency distribution spreading in one direction in a frequency space as a candidate region where the moving image blur may exist. The video processing apparatus according to claim 1, wherein: The video processing apparatus according to claim 3,
The moving image blur reduction unit detects the frequency distribution spreading in the one direction by pattern matching with a frequency distribution pattern prepared in advance in the determination of the frequency characteristic.   The moving image blur reduction unit calculates a correction range and a correction direction of blur reduction correction based on the motion vector detected by the motion vector detection unit, and performs blur reduction correction from the frequency distribution that is a frequency conversion result by the frequency conversion unit. The video processing apparatus according to claim 1, wherein a correction intensity is calculated, and blur reduction correction is performed using the correction range, the correction direction, and the correction intensity.   The moving image blur reduction unit calculates a correction strength of blur reduction correction based on the motion vector detected by the motion vector detection unit, and a correction range of blur reduction correction from the frequency distribution that is a frequency conversion result by the frequency conversion unit The video processing apparatus according to claim 1, wherein blur correction is performed using the correction strength, the correction range, and the correction direction. The video blur reduction unit
Based on the frequency conversion result by the frequency conversion unit, the correction strength of the blur reduction correction is calculated, and the correction range and correction direction of the blur reduction correction are calculated based on the motion vector detected by the motion vector detection unit. A first blur reduction correction used for the reduction correction;
Based on the frequency conversion result by the frequency conversion unit, a correction range and a correction direction for blur reduction correction are calculated, and a correction strength for blur reduction correction is calculated based on the motion vector detected by the motion vector detection unit. A second blur reduction correction used for the reduction correction,
The video processing apparatus according to claim 1, wherein the video processing apparatus is switchable.   The first blur reduction correction and the second blur reduction correction of the moving image blur reduction unit are switched according to a frequency conversion result by the frequency conversion unit or a motion vector detected by the motion vector detection unit. The video processing apparatus according to claim 7.   8. The first blur reduction correction and the second blur reduction correction of the moving image blur reduction unit are switched according to content-related information or program-related information of a content or program including the video. The video processing apparatus described in 1. A video processing device according to any one of claims 1 to 9,
The video processing apparatus, wherein the motion vector detection unit detects a motion vector of the video using a frequency conversion result by the frequency conversion unit. A frequency conversion step for converting the frequency of the video included in the input signal;
A motion vector detection step of detecting a motion of a video included in the input signal;
A video processing method comprising: a moving image blur reduction step of performing moving image blur reduction correction on the video based on a frequency conversion result obtained by the frequency conversion step and a motion vector detected by the motion vector detection step.   In the moving image blur reduction step, a correction range, a correction direction, and a correction strength for blur reduction correction are calculated based on the frequency conversion result in the frequency conversion step and the motion vector detected in the motion vector detection step, and the calculation is performed. The video processing method according to claim 11, wherein the processed parameter is used for blur reduction correction. A video processing method according to any one of claims 11 and 12,
In the moving image blur reduction step, the frequency characteristic of the frequency conversion result in the frequency conversion step is determined, and an image region having a frequency distribution spreading in one direction in the frequency space is determined as a candidate region in which moving image blur may exist. The video processing method according to claim 11 or 12, wherein: The video processing method according to claim 13,
In the moving image blur reduction step, in the determination of the frequency characteristic, the frequency distribution spreading in one direction is detected by pattern matching with a frequency distribution pattern prepared in advance.   In the moving image blur reduction step, a correction range and a correction direction of blur reduction correction are calculated based on the motion vector detected in the motion vector detection step, and blur reduction correction is performed from the frequency distribution that is a frequency conversion result in the frequency conversion step. The image processing method according to claim 11, wherein a blur reduction correction is performed using the correction range, the correction direction, and the correction strength.   In the moving image blur reduction step, the blur reduction correction intensity is calculated based on the motion vector detected in the motion vector detection step, and the blur reduction correction is corrected from the frequency distribution that is a frequency conversion result in the frequency conversion step. 12. The video processing method according to claim 11, wherein a range and a correction direction are calculated, and blur reduction correction is performed using the correction intensity, the correction range, and the correction direction. In the video blur reduction step,
Based on the frequency conversion result in the frequency conversion step, calculates the correction strength of the blur reduction correction, calculates the correction range and correction direction of the blur reduction correction based on the motion vector detected in the motion vector detection step, A first blur reduction correction used for blur reduction correction;
Based on the frequency conversion result in the frequency conversion step, calculate the correction range and correction direction of the blur reduction correction, calculate the correction strength of the blur reduction correction based on the motion vector detected in the motion vector detection step, A second blur reduction correction used for blur reduction correction,
The video processing method according to claim 11, wherein switching is possible.   The first blur reduction correction and the second blur reduction correction in the moving image blur reduction step are switched according to a frequency conversion result in the frequency conversion step or a motion vector detected in the motion vector detection step. The video processing method according to claim 17.   18. The first blur reduction correction and the second blur reduction correction in the moving picture blur reduction step are switched according to content-related information or program-related information of a content or program including the video. The video processing method described in 1. A video processing method according to any one of claims 11 to 19, wherein
In the motion vector detection step, a motion vector of the video is detected using a frequency conversion result in the frequency conversion step.

Images