JP2009296124A - Image processing apparatus and method and decoding device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce noise generated as pulsations in video images. <P>SOLUTION: An image processing apparatus includes: a three-dimensional noise reduction processing means (120) for performing three-dimensional noise reduction processing on input video signals; a pulsation detection means (110) for detecting a pulsation phenomenon that an object projected in the video image, corresponding to the input video signal, is pulsed; and a control means (130) for controlling the three-dimensional noise reduction processing means so that the strength of the three-dimensional noise reduction processing on a pulsating part where the pulsation phenomenon is detected in the input video signals is higher than the strength of the three-dimensional noise reduction processing on a non-pulsating part that excludes the pulsation part in the input video signals. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and method for reducing image noise, and a technical field of a decoding apparatus.

  As this type of image processing device, for example, a device that is provided in a digital broadcast receiver and performs two-dimensional noise reduction processing or three-dimensional noise reduction processing as processing for reducing noise of an image or video is known ( For example, see Patent Document 1).

  The two-dimensional noise reduction process is a process of removing noise included in the one frame using the video signal of one frame, and the three-dimensional noise reduction process is a video signal of a plurality of consecutive frames (or fields). Are compared, and the extracted difference signal is recognized as noise and removed.

  On the other hand, as a video signal encoding method, MPEG (Moving Picture Experts Group) -2 method or H.264 is used. The H.264 system is known. In these encoding methods, image data is divided into I picture (intra-frame encoded image: Intra-Picture), P picture (inter-frame forward predictive encoded image: Predictive-Picture) and B picture (both in each frame). It is encoded into any one of the data of Bidirectionally predictive-picture. Furthermore, the image data is configured in units of GOP (Group of Picture). The GOP is composed of a P picture and a B picture with the I picture at the head, such as IBBPBBPBBPBBPBB.

  In such an encoding method, an object reflected in a video is originally stationary, but its outline part is deformed or its luminance is changed. There is a case. Such deformation or change propagates from the I picture that is the intra-frame encoded image to the P picture or B picture that is the predictive encoded image, and as a result, corresponds to the period of the I picture (in other words, the GOP). The object appearing in the image is perceived as pulsating. That is, in such an encoding method, a pulsation phenomenon or pulsation noise in which an object shown in the image pulsates at a period corresponding to GOP (hereinafter, this pulsation phenomenon or pulsation noise is also simply referred to as “pulsation”) occurs. There is a risk of it. Such pulsation is sometimes called “breathing” (see, for example, Patent Document 2).

  Therefore, for example, Patent Document 2 discloses a technique for reducing breathing (ie, pulsation) by intentionally degrading an image of a frame with a low degradation level to the same level as an image of a frame with a large degradation level. Has been. For example, Patent Document 3 discloses a technique for intentionally blurring a set of frames in a GOP so as to achieve a smooth visual transition between frames.

JP 2005-150903 A Japanese translation of PCT publication No. 2002-514023 JP 2005-528856 A

  However, according to the techniques disclosed in Patent Documents 2 and 3 described above, the image quality of a frame with a small image quality degradation is intentionally degraded in order to reduce pulsation in the image, so that the overall image quality of the entire image is improved. There is a technical problem that it may decrease.

  The present invention has been made in view of the above-described problems, for example, and provides an image processing apparatus and method capable of reducing noise generated as pulsation in an image, and a decoding apparatus including such an image processing apparatus. Is an issue.

  In order to solve the above-mentioned problem, an image processing apparatus according to claim 1 is directed to an input video signal in which image data encoded on the transmission side is transmitted to the reception side and decoded on the reception side. An image processing apparatus for performing image processing, wherein three-dimensional noise reduction processing means for performing three-dimensional noise reduction processing as the image processing, and a pulsation phenomenon in which an object reflected in an image corresponding to the input video signal pulsates is detected The intensity of the three-dimensional noise reduction processing for the pulsation part in which the pulsation phenomenon is detected in the input video signal, and the three-dimensional for the non-pulsation part excluding the pulsation part in the input video signal. Control means for controlling the three-dimensional noise reduction processing means so as to be higher than the intensity of the noise reduction processing.

  In order to solve the above problem, an image processing method according to claim 7 is directed to an input video signal in which image data encoded on a transmission side is transmitted to a reception side and decoded on the reception side. An image processing method for performing image processing, comprising: a three-dimensional noise reduction processing step for performing three-dimensional noise reduction processing as the image processing; and detecting a pulsation phenomenon in which an object shown in a video corresponding to the input video signal pulsates The intensity of the three-dimensional noise reduction process for the pulsation part in which the pulsation phenomenon is detected in the input video signal, and the three-dimensional part for the non-pulsation part excluding the pulsation part in the input video signal. A control step of controlling the three-dimensional noise reduction processing means so as to be higher than the intensity of the noise reduction processing.

  In order to solve the above-described problem, a decoding device according to claim 8 decodes the encoded image data and the image processing device according to any one of claims 1 to 6 to perform the input. Decoding means for outputting as a video signal.

  The operation and other advantages of the present invention will become apparent from the best mode for carrying out the invention described below.

  Hereinafter, an image processing apparatus and method, and a decoding apparatus according to embodiments of the present invention as the best mode for carrying out the invention will be described in order.

(Embodiment of Image Processing Device)
An embodiment of the image processing apparatus according to the present invention is an image processing apparatus for performing image processing on an input video signal obtained by transmitting image data encoded on the transmission side to the reception side and decoding on the reception side. A three-dimensional noise reduction processing means for performing a three-dimensional noise reduction process as the image processing; a pulsation detection means for detecting a pulsation phenomenon in which an object shown in an image corresponding to the input video signal pulsates; The intensity of the three-dimensional noise reduction process for the pulsation part where the pulsation phenomenon is detected in the input video signal is greater than the intensity of the three-dimensional noise reduction process for the non-pulsation part excluding the pulsation part of the input video signal. Control means for controlling the three-dimensional noise reduction processing means so as to be higher.

  According to the embodiment of the image processing apparatus of the present invention, during the operation, the 3D noise reduction process is performed on the input video signal by the 3D noise reduction unit under the control of the control unit.

  The input video signal is transmitted on the transmission side, for example, MPEG-2, This is a signal obtained by transmitting image data encoded by an encoding method such as the H.264 method to the reception side via a transmission path, and decoding the transmitted image data on the reception side.

  The three-dimensional noise reduction process generates a difference between frames (or between fields) by subtracting the video signal of one frame (or one field) and the video signal of the current input frame (or input field), This difference is used to remove noise included in the current input frame (or input field). More specifically, the three-dimensional noise reduction process subtracts the video signal of the previous frame and the video signal of the current input frame to obtain a difference between frames, and multiplies the difference between the frames by a certain coefficient. Or a process of removing noise contained in the video signal by adding or subtracting a specific frequency band extracted from the difference between the frames to the video signal of the current input frame.

  In this embodiment, the pulsation phenomenon in the video corresponding to the input video signal is detected by the pulsation detecting means. The “pulsation phenomenon” according to the present invention means that an object shown in an image is almost or completely stationary in position, and the shape or brightness level of the object which does not change originally changes periodically. The image data is, for example, MPEG-2 format, H.264, etc. It may be generated by encoding with an encoding method with motion compensation such as the H.264 method. The pulsation phenomenon is typically noise generated in a period corresponding to a GOP (that is, a GOP period) in the outline of an object where an object that is not originally moving is visually recognized as if pulsing. is there.

  In the present embodiment, in particular, the control means is configured so that the intensity of the three-dimensional noise reduction process for the pulsating part of the input video signal is higher than the intensity of the three-dimensional noise reduction process for the non-pulsating part of the input video signal. Controls the three-dimensional noise reduction processing means. Therefore, the three-dimensional noise reduction process is performed on the pulsating portion of the input video signal with higher intensity than the non-pulsating portion of the input video signal. Therefore, since the video corresponding to the pulsation portion can be smoothed with time, the pulsation phenomenon in the video can be reliably reduced. Furthermore, since the three-dimensional noise reduction process is performed on the non-pulsation part with a weaker intensity than the pulsation part, an adverse effect on the image due to excessively high intensity of the three-dimensional noise reduction process (that is, image quality) (Degradation of the image) hardly or practically occurs.

  As described above, according to the present embodiment, noise generated as pulsation in an image can be reduced.

  In one aspect of the embodiment of the image processing apparatus of the present invention, the pulsation detecting means determines whether a difference signal between two frames or fields having different times in the input video signal is greater than a predetermined threshold value. A difference detecting means for detecting; a motion vector detecting means for detecting a motion vector based on the difference signal; and the difference detecting means detecting that the difference signal is greater than the predetermined threshold, and the motion vector detecting means And a pulsation determining means for determining that the pulsation phenomenon has occurred when the magnitude of the motion vector detected by the step is smaller than a predetermined size.

  According to this aspect, the pulsation phenomenon can be suitably detected. That is, when it is detected that the difference signal is larger than the predetermined threshold value and the magnitude of the motion vector is smaller than the predetermined magnitude, it is determined by the pulsation determining means that the pulsation phenomenon has occurred. The pulsation phenomenon in which the position of the object being reflected is almost or completely stationary (that is, hardly or not moving at all) and the shape or luminance level of the object changes periodically is detected with high accuracy. It becomes possible.

  In the aspect in which the pulsation detecting unit includes the difference detecting unit, the motion vector detecting unit, and the pulsation determining unit, the control unit is configured to perform the pulsation based on at least one of the signal level of the difference signal and the magnitude of the motion vector. You may determine the intensity | strength of the said three-dimensional noise reduction process with respect to a part.

  In this case, the control means determines the strength of the three-dimensional noise reduction process for the pulsation part based on at least one of the signal level of the difference signal and the magnitude of the motion vector. Therefore, the strength of the three-dimensional noise reduction process can be determined according to the degree of the pulsation phenomenon (that is, the pulsation level). That is, the intensity of the three-dimensional noise reduction process can be determined according to the pulsation level estimated based on at least one of the signal level of the difference signal and the magnitude of the motion vector. Therefore, noise generated as pulsation in the video can be effectively reduced.

  In another aspect of the embodiment of the image processing apparatus of the present invention, the control means decodes the encoded image data and outputs the decoded image data as the input video signal and also outputs the encoding information related to the encoding. The strength of the three-dimensional noise reduction process for the pulsation portion is determined based on the coding information.

  According to this aspect, for example, encoding information (eg, MPEG-2, H.264, VC-1, etc.), average bit rate, quantization step, loop filter coefficient, GOP period, picture mode, etc. Based on the above, the strength of the three-dimensional noise reduction processing for the pulsation portion is determined. Therefore, the strength of the three-dimensional noise reduction process can be determined according to the possibility of occurrence of the pulsation phenomenon and the degree of the pulsation phenomenon. Therefore, noise generated as pulsation in the video can be effectively reduced.

  In another aspect of the embodiment of the image processing apparatus of the present invention, the pulsation detecting unit decodes the encoded image data and outputs the decoded image data as the input video signal, and the encoded information related to the encoding is output. The pulsation phenomenon is detected based on the encoded information.

  According to this aspect, for example, the pulsation phenomenon is detected based on encoding information such as a GOP cycle, a picture mode, and a prediction mode in intra prediction encoding. Therefore, the pulsation phenomenon can be detected with high accuracy, in other words, the pulsation portion and the non-pulsation portion in the input video signal can be distinguished with high accuracy. Therefore, it is possible to more reliably perform the three-dimensional noise reduction process with a high intensity on the pulsating part, and more reliably perform the three-dimensional noise reduction process with a lower intensity on the non-pulsation part. .

  In another aspect of the embodiment of the image processing apparatus of the present invention, the image processing apparatus further includes a two-dimensional noise reduction processing unit that performs a two-dimensional noise reduction process on the input video signal, and the control unit controls the pulsation portion. The two-dimensional noise reduction processing means is controlled so that the strength of the two-dimensional noise reduction processing is lower than the strength of the two-dimensional noise reduction processing for the non-pulsation portion.

  According to this aspect, the pulsation portion of the input video signal is subjected to the three-dimensional noise reduction process with higher intensity than the two-dimensional noise reduction process, and therefore noise generated as pulsation in the video is more effective. Can be reduced.

(Embodiment of Image Processing Method)
An embodiment of the image processing method according to the present invention is an image processing method for performing image processing on an input video signal obtained by transmitting image data encoded on the transmission side to the reception side and decoding on the reception side. A three-dimensional noise reduction processing step of performing a three-dimensional noise reduction process as the image processing, a pulsation detection step of detecting a pulsation phenomenon in which an object reflected in an image corresponding to the input video signal pulsates, The intensity of the three-dimensional noise reduction process for the pulsation part where the pulsation phenomenon is detected in the input video signal is greater than the intensity of the three-dimensional noise reduction process for the non-pulsation part excluding the pulsation part of the input video signal. And a control step of controlling the three-dimensional noise reduction processing means so as to increase.

  According to the embodiment of the image processing method of the present invention, it is possible to receive the same benefits as the benefits of the above-described embodiment of the image processing device of the present invention.

  Incidentally, in response to the various aspects of the embodiment of the image processing apparatus of the present invention described above, the embodiment of the image processing method of the present invention can also adopt various aspects.

(Embodiment of decoding apparatus)
An embodiment of the decoding device of the present invention includes the above-described embodiment of the image processing device of the present invention (including various aspects), and decodes the encoded image data as the input video signal. Decoding means for outputting.

  According to the embodiment of the decoding device of the present invention, it is possible to receive the same benefits as those of the above-described embodiment of the image processing device of the present invention.

  The effect | action and other gain of embodiment of this invention are clarified from the Example demonstrated below.

  As described above, according to the embodiment of the image processing apparatus of the present invention, the image processing apparatus includes the three-dimensional noise reduction processing unit, the pulsation detection unit, and the control unit. According to the embodiment of the image processing method of the present invention, the image processing method includes a three-dimensional noise reduction processing step, a pulsation detection step, and a control step. According to the embodiment of the decoding apparatus of the present invention, the above-described embodiment of the image processing apparatus of the present invention and decoding means are provided. Therefore, noise generated as pulsation in the video can be reduced.

  Embodiments of the present invention will be described below with reference to the drawings.

<First embodiment>
An image processing apparatus according to the first embodiment will be described with reference to FIGS. In the present embodiment, the case where the image processing apparatus according to the present invention is provided on the digital broadcast receiving side will be described as an example.

  FIG. 1 is a block diagram illustrating the configuration of the image processing apparatus according to the first embodiment.

  In FIG. 1, the image processing apparatus 100 according to the present embodiment includes a pulsation detection unit 110, a three-dimensional noise reduction processing unit 120, and a control unit 130.

  The image processing apparatus 100 performs a three-dimensional noise reduction process on an input video signal input from the outside (that is, a decoder 810, which will be described later with reference to FIG. 2) as described later, and outputs an output video signal. Output. Note that the image processing apparatus 100 may be configured to perform other image processing such as two-dimensional noise reduction processing on the input video signal in addition to the three-dimensional noise reduction processing.

  FIG. 2 is a block diagram illustrating a configuration of a digital broadcasting system including the image processing apparatus according to the present embodiment.

  In FIG. 2, an encoder 910 and a modulator 920 are provided on the transmission side of the digital broadcast system, and a decoder 810, a demodulator 820, an image processing device 100, and a display device are provided on the reception side of the digital broadcast system. 850 is provided.

  On the transmission side of the digital broadcasting system, image data to be transmitted to the reception side is transmitted by an encoder 910, for example, MPEG-2 system, H.264, etc. It is encoded by an encoding method with motion compensation such as the H.264 method. The encoded image data is modulated by the modulator 920 and transmitted to the receiving side via the transmission line 700 at a bit rate within a predetermined range.

  On the receiving side of the digital broadcasting system, image data transmitted from the transmitting side via the transmission path 700 is demodulated by the demodulator 820 and then decoded by the decoder 810 and input to the image processing apparatus 100 as an input video signal. The The image processing apparatus 100 performs image processing on the input video signal and outputs it to the display device 850 as an output video signal. The display device 850 displays video based on the output video signal from the image processing device 100. The image processing apparatus 100 constitutes an example of the “decoding apparatus” according to the present invention together with the decoder 810.

  As described above, the input video signal input to the image processing apparatus 100 is, for example, MPEG-2 format, H.264, or the like on the transmission side. A signal obtained by transmitting image data encoded by an encoding method with motion compensation such as the H.264 method to the receiving side via the transmission path 700 and decoding the transmitted image data by the decoder 810 on the receiving side. It is. For this reason, the input video signal may include a pulsation part in which a pulsation phenomenon occurs. That is, the pulsation phenomenon (that is, the object shown in the image is almost or completely stationary in position and the shape or brightness level of the object that does not change originally is present in the image corresponding to the input image signal. For example, there is a possibility that a phenomenon that periodically changes in a GOP cycle or the like) occurs. As described above, the pulsation phenomenon is caused by, for example, the MPEG-2 system, H.264, or the like. This occurs when the image data is encoded by an encoding method such as the H.264 method. Furthermore, since the encoded image data needs to be transmitted from the transmission side to the reception side via the transmission path 700 at a bit rate within a predetermined range, the high frequency of the image is suppressed in order to keep the bit amount within the predetermined range. Band components may be lost. If the high-frequency band component is lost in this way, the contour of the object shown in the image corresponding to the image data is likely to be deformed or the luminance is likely to change, and as a result, the pulsation phenomenon is visually recognized. It becomes easy.

  In FIG. 1 again, the pulsation detecting unit 110 detects a pulsation phenomenon in which an object shown in an image corresponding to the input video signal pulsates. The pulsation detection unit 110 outputs a pulsation flag, which is information indicating whether or not a pulsation phenomenon is detected, to the control unit 130. When detecting a pulsation phenomenon, the pulsation detection unit 110 outputs the pulsation flag to an on state (in other words, raising the pulsation flag) to the control unit 130, and on the other hand, when no pulsation phenomenon is detected. Then, the pulsation flag is turned off (in other words, the pulsation flag is not raised) and output to the control unit 130.

  FIG. 3 is a block diagram illustrating the configuration of the pulsation detecting unit according to the present embodiment.

  In FIG. 3, the pulsation detection unit 110 includes a motion detection unit 111, a motion vector detection unit 112, a pulsation determination unit 113, and a frame memory 114. The motion detection unit 111 is an example of the “difference detection unit” according to the present invention.

  The input video signal is input to the frame memory 114, the motion detection unit 111, and the motion vector detection unit 112.

  The frame memory 114 is configured to hold an input video signal for one frame. The frame memory 114 delays the input video signal by one frame and outputs it to the motion detector 111 and the motion vector detector 112.

  The motion detection unit 111 calculates an absolute value of a difference signal (that is, a difference absolute value) that is a difference between a luminance signal between the input video signal and the input video signal delayed by one frame input from the frame memory 114. . The motion detection unit 111 outputs a difference flag, which is information indicating whether or not the calculated difference absolute value is greater than a predetermined threshold value, to the pulsation determination unit 113. When the calculated difference absolute value is larger than the predetermined threshold, the motion detection unit 111 sets the difference flag to the on state (in other words, sets the difference flag) and outputs it to the pulsation determination unit 113, while the calculated difference absolute value When the value is not larger than the predetermined threshold value, the pulsation flag is turned off (in other words, the difference flag is not set) and output to the pulsation determination unit 113.

  FIG. 4 is a block diagram illustrating the configuration of the motion detection unit according to the present embodiment.

  As illustrated in FIG. 4, the motion detection unit 111 includes a difference absolute value processing unit 1111 and a threshold processing unit 1112.

  The difference absolute value processing unit 1111 calculates the difference absolute value from the input video signal and the input video signal delayed by one frame input from the frame memory 114. The difference absolute value processing unit 1111 outputs the calculated difference absolute value to the threshold processing unit 1112.

  The threshold processing unit 1112 compares the difference absolute value input from the difference absolute value processing unit 1111 with a predetermined threshold value, and outputs the comparison result to the pulsation determination unit 113 as a difference flag. For example, the predetermined threshold value may be set to zero, or may be set to a value corresponding to the magnitude of a change in luminance level to be detected as a pulsation phenomenon. In addition, when the difference flag is on for a part of the input video signal, at least one of “random noise”, “motion”, and “pulsation phenomenon” exists in the video corresponding to the part. Means.

  In FIG. 3 again, the motion vector detection unit 112 detects a motion vector by the block matching method based on the input video signal and the input video signal delayed from the frame memory 114 by one frame. In addition, as a method for the motion vector detection unit 112 to detect the motion vector, another method such as a gradient method may be used.

  The motion vector detection unit 112 outputs a motion vector flag, which is information indicating whether or not the size of the detected motion vector is zero, to the pulsation determination unit 113. When the magnitude of the motion vector is zero, the motion vector detection unit 112 sets the motion vector flag to the on state (in other words, sets the motion vector flag) and outputs it to the pulsation determination unit 113. On the other hand, when the magnitude of the detected motion vector is not zero, the motion vector detection unit 112 outputs the motion vector flag to the pulsation determination unit 113 with the motion vector flag turned off (in other words, without setting the motion vector flag).

  When the magnitude of the motion vector is smaller than the predetermined magnitude, the motion vector detection unit 112 outputs the motion vector flag to the pulsation determination unit 113 with the motion vector flag turned on and the magnitude of the detected motion vector is equal to the predetermined magnitude. If it is greater than or equal to the size, the motion vector flag may be output to the pulsation determining unit 113 as an off state.

  FIG. 5 is a block diagram illustrating the configuration of the motion vector detection unit according to the present embodiment.

  As shown in FIG. 5, the motion vector detection unit 112 includes a correlation detection unit 1121 and a correlation determination unit 1122.

  Based on the input video signal and the input video signal delayed by one frame input from the frame memory 114, the correlation detection unit 1121 determines the previous one frame for the input video signal corresponding to each block of the current frame. The absolute value of the difference from the video signal corresponding to each block is calculated, and correlation information indicating the position of the block one frame before giving the smallest absolute difference among the absolute differences is detected (ie, the block is detected). By detecting a motion vector by searching for a motion vector as a unit, correlation information that is information on the motion vector is detected). The correlation detection unit 1121 outputs the minimum difference absolute value as the difference level to the correlation determination unit 1122 and outputs the detected correlation information to the correlation determination unit 1122.

  Based on the correlation information, correlation determination section 1122 determines whether the deviation between the position of the block one frame before giving the minimum difference absolute value and the position of the block in the corresponding current frame is within a predetermined range. It is determined whether or not the magnitude of the motion vector is zero by determining whether or not the difference level is equal to or less than a predetermined threshold. The correlation determination unit 1122 has a predetermined difference between the position of the block one frame before giving the minimum difference absolute value and the position of the block in the corresponding current frame, and the difference level is When it is determined that the value is equal to or less than the predetermined threshold value, the motion vector flag is set to an off state and output to the pulsation determining unit 113.

  FIG. 6 is a conceptual diagram illustrating an example of motion vector detection by the motion vector detection unit according to the present embodiment.

  FIG. 6 shows an image of the current frame (right side in FIG. 6) and an image of the previous frame (left side in FIG. 6) for the video image of the object C1. In the detection of the motion vector by the motion vector detection unit 112, each frame is divided into m × n blocks, and a motion vector search is performed in units of the blocks.

  As shown in FIG. 6, when the position of the object C1 is changed without changing the shape of the object C1 between the image of the previous frame and the image of the current frame (that is, the object C1 When the object C1 is moving without being deformed (in other words, when no pulsation phenomenon occurs), a motion vector that is an example of a motion vector detected by the motion vector detection unit 112 The magnitude of V1 is not zero. Therefore, in this case, the motion vector detection unit 112 outputs the motion vector flag to the pulsation determination unit 113 as an off state.

  FIG. 7 is a conceptual diagram illustrating another example of motion vector detection by the motion vector detection unit according to the present embodiment.

  FIG. 7 shows an image of the current frame (right side in FIG. 7) and an image of the previous frame (left side in FIG. 7) for the video image of the object C2.

  As shown in FIG. 7, when the shape or luminance level of the object C2 slightly changes between the image of the previous frame and the image of the current frame, and the position of the object C2 does not change (that is, When the contour portion of the object C2 is slightly deformed and the object C2 does not move (in other words, when a pulsation phenomenon may occur), the motion detected by the motion vector detection unit 112 The magnitude of the motion vector V2, which is an example of the vector, is zero. Therefore, in this case, the motion vector detection unit 112 outputs the motion vector flag to the pulsation determination unit 113 as an on state.

  In FIG. 3 again, the pulsation determination unit 113 determines whether a pulsation phenomenon has occurred in the video based on the difference flag input from the motion detection unit 111 and the motion vector flag input from the motion vector detection unit 112. (That is, whether the currently input portion of the input video signal is a pulsation portion) and outputs the determination result to the control unit 130 as a pulsation flag.

  More specifically, when both the difference flag and the motion vector flag are in the on state, the pulsation determining unit 113 determines that a pulsation phenomenon has occurred in the video, and sets the pulsation flag to the on state to control the control unit. To 130. On the other hand, when at least one of the difference flag and the motion vector flag is in the off state, the pulsation determining unit 113 determines that no pulsation phenomenon has occurred in the video, and outputs the pulsation flag to the control unit 130 as an off state. To do.

  Thus, the pulsation detection unit 110 detects a pulsation phenomenon in an image corresponding to the input video signal (in other words, detects a pulsation part and a non-pulsation part in the input video signal), and controls the detection result as a pulsation flag. To the unit 130.

  Particularly in the present embodiment, as described above, the pulsation detection unit 110 includes the pulsation determination unit 113 that determines whether or not a pulsation phenomenon occurs in the video based on the difference flag and the motion vector flag. Therefore, the pulsation phenomenon can be detected with high accuracy.

  In FIG. 1 again, the three-dimensional noise reduction processing unit 120 performs a three-dimensional noise reduction process on the input video signal under the control of the control unit 130, and the input video signal subjected to the three-dimensional noise reduction process. Is output as an output video signal. As described above with reference to FIG. 2, the output video signal is input to the display device 850.

  FIG. 8 is a block diagram illustrating the configuration of the three-dimensional noise reduction processing unit according to the present embodiment together with the control unit according to the present embodiment.

  In FIG. 8, the three-dimensional noise reduction processing unit 120 includes a subtractor 121, an adder 122, a frame memory 123, a nonlinear processing unit 124, and a multiplier 125.

  The input video signal is input to the subtracter 121 and the adder 122.

  The subtractor 121 generates a difference signal N1 by subtracting the input video signal from the output video signal delayed by one frame by the frame memory 123, and outputs the difference signal N1 to the nonlinear processing unit 124.

  The non-linear processing unit 124 passes only the signal level component defined by the coefficient β of the difference signal N1 output from the subtractor 121 and outputs the signal N2 to the multiplier 125. The coefficient β is controlled by the control unit 130.

  The multiplier 125 multiplies the signal N2 output from the nonlinear processing unit 124 by the coefficient α and outputs the result to the adder 122. The coefficient α is controlled by the control unit 130.

  The adder 122 adds the input video signal and the signal output from the multiplier 125 and outputs the result as an output video signal.

  The three-dimensional noise reduction processing unit 120 configured as described above generates a difference signal N1 between frames by subtracting the input video signal of the previous frame and the input video signal of the current frame by the subtractor 121, The signal N2 obtained by extracting the signal level component defined by the coefficient β from the difference signal N1 and multiplied by the coefficient α is added to the input video signal of the current frame, thereby removing noise contained in the input video signal. 3D noise reduction processing is performed.

  In FIG. 1 and FIG. 8, the control unit 130 corresponds to the three-dimensional noise reduction processing unit 120 (more specifically, the multiplier of the three-dimensional noise reduction processing unit 120) according to the pulsation flag output from the pulsation detection unit 110. The coefficient α in 125 and the coefficient β in the nonlinear processing unit 124 of the three-dimensional noise reduction processing unit 120 are controlled.

  Particularly in the present embodiment, when the control unit 130 performs the three-dimensional noise reduction process on the pulsation part according to the pulsation flag (that is, when the pulsation flag is on), the control unit 130 performs the three-dimensional process on the non-pulsation part. The coefficient β is controlled and the signal N2 is multiplied by the multiplier 125 so that the signal level passed through by the non-linear processing unit 124 becomes larger than when the noise reduction process is performed (that is, when the pulsation flag is off). The coefficient α is controlled so as to increase (or increase) the coefficient α. In other words, the control unit 130 controls the coefficients α and β in this way so that the intensity of the three-dimensional noise reduction process for the pulsating part is higher than the intensity of the three-dimensional noise reduction process for the non-pulsating part. The three-dimensional noise reduction processing unit 120 is controlled.

  Next, the control of the coefficient β by the control unit according to the present embodiment will be described more specifically with reference to FIG.

  FIG. 9 shows the passing signal level in each case when the value of the coefficient β is 1 and when the value of the coefficient β is 0 in the nonlinear processing unit according to the present embodiment.

  In FIG. 9, the non-linear processing unit 124 is configured such that when the value of the coefficient β is 1, the signal level that allows the difference signal N1 to pass is higher than when the value of the coefficient β is 0. Further, the controller 130 sets the value of the coefficient β to 1 when performing the three-dimensional noise reduction process on the pulsation part, and performs the three-dimensional noise reduction process on the non-pulsation part. Set the value of the coefficient β to 0. As described above, when the control unit 130 performs the three-dimensional noise reduction processing on the pulsating portion, the signal level that the nonlinear processing unit 124 passes is higher than when the three-dimensional noise reduction processing is performed on the non-pulsating portion. The coefficient β is controlled so as to increase (in other words, the nonlinear processor 124 is controlled). That is, the non-linear processing unit 124 is controlled by the control unit 130 so that the three-dimensional noise reduction process is performed on the pulsating part with higher intensity than that on the non-pulsating part.

  FIG. 10 is a flowchart for explaining the flow of the operation of the control unit according to the present embodiment. This process is executed in units of pixels or blocks.

  In FIG. 10, first, it is determined by the pulsation detection unit 110 (see FIG. 1) whether or not the pulsation part has started (step S10). In other words, the pulsation determination unit 113 (see FIG. 3) in the pulsation detection unit 110 determines whether or not a pulsation phenomenon has occurred in the video corresponding to the input video signal, thereby starting the pulsation part in the input video signal. It is determined whether or not it has been done. The pulsation detection unit 110 outputs the determination result to the control unit 130 as a pulsation flag. That is, when the pulsation detection unit 110 determines that the pulsation part has started, the pulsation flag is turned on and is output to the control unit 130. On the other hand, when the pulsation part determines that the pulsation part has not been started, The flag is output to the control unit 130 as an off state.

  When it is determined that the pulsation part has not started (step S10: No), it is determined whether or not the pulsation part has started with the next pixel or block in the input video signal as the current pixel or block. .

  On the other hand, when it is determined that the pulsation portion has started (step S10: Yes), the control unit 130 sets the strength of the three-dimensional noise reduction process to be higher (step S20). That is, the control unit 130 controls the coefficient β in the non-linear processing unit 124 of the three-dimensional noise reduction processing unit 120 so that the value β becomes 1 (see FIG. 9) and the multiplication of the three-dimensional noise reduction processing unit 120. The coefficient α in the device 125 is controlled to be larger than the coefficient α when the three-dimensional noise reduction process is performed on the non-pulsation portion. That is, when the pulsation part is started, the control unit 130 determines that the intensity of the three-dimensional noise reduction process performed on the input video signal is higher than the intensity of the three-dimensional noise reduction process performed on the non-pulsation part. The three-dimensional noise reduction processing unit 120 is controlled so as to be higher.

  Next, it is determined by the pulsation detection unit 110 whether or not the pulsation part has ended (step S30). In other words, the pulsation determination unit 113 in the pulsation detection unit 110 determines whether or not the pulsation phenomenon has occurred in the video corresponding to the input video signal, thereby determining whether or not the pulsation part has ended. The pulsation detection unit 110 outputs the determination result to the control unit 130 as a pulsation flag. That is, if the pulsation detecting unit 110 determines that the pulsation part has ended (that is, no pulsation phenomenon has occurred), the pulsation detection unit 110 outputs the pulsation flag to the control unit 130 as an off state, and the pulsation part has ended. When it is determined that there is no pulsation phenomenon (that is, the pulsation phenomenon has occurred), the pulsation flag is turned on and output to the control unit 130.

  If it is determined that the pulsation portion has not ended (step S30: No), it is determined whether the pulsation portion has ended with the next pixel or block in the input video signal as the current pixel or block. .

  On the other hand, when it is determined that the pulsation portion has ended (step S30: Yes), the control unit 130 sets the strength of the three-dimensional noise reduction process to be low (step S40). That is, the control unit 130 controls the coefficient β in the non-linear processing unit 124 of the three-dimensional noise reduction processing unit 120 so that the value β becomes 0 (see FIG. 9), and the multiplication of the three-dimensional noise reduction processing unit 120. The coefficient α in the unit 125 is controlled to be smaller than the coefficient α when the three-dimensional noise reduction process is performed on the pulsation portion. That is, when the pulsation part is completed, the control unit 130 determines that the intensity of the 3D noise reduction process performed on the input video signal is lower than the intensity of the 3D noise reduction process performed on the pulsation part. Thus, the three-dimensional noise reduction processing unit 120 is controlled.

  Subsequently, it is determined whether or not the pulsation portion has started with the next pixel or block in the input video signal as the current pixel or block (step S10).

  In this way, particularly in the present embodiment, the control unit 130 performs the three-dimensional noise reduction process so that the intensity of the three-dimensional noise reduction process for the pulsation part is higher than the intensity of the three-dimensional noise reduction process for the non-pulsation part. The unit 120 is controlled. Therefore, the three-dimensional noise reduction process is performed on the pulsating portion of the input video signal with higher intensity than the non-pulsating portion of the input video signal. Therefore, since the video corresponding to the pulsation portion can be smoothed with time, the pulsation phenomenon in the video can be reliably reduced. Furthermore, since the three-dimensional noise reduction process is performed on the non-pulsation part with a weaker intensity than the pulsation part, an adverse effect on the image due to excessively high intensity of the three-dimensional noise reduction process (that is, image quality) (Degradation of the image) hardly or practically occurs.

  In this embodiment, the image processing apparatus 100 is configured to perform 3D noise reduction processing in units of frames, but is configured to perform 3D noise reduction processing in units of fields instead of frames. May be. Also in this case, noise generated as pulsation in the video can be reduced as in the present embodiment.

  In the present embodiment, the control unit 130 is configured to control both the coefficients α and β, but may be configured to control either the coefficient α or β. Even in this case, noise generated as pulsation in the video can be reduced. However, from the viewpoint of controlling the intensity of the three-dimensional noise reduction process with higher accuracy, the control unit 130 is preferably configured to control both the coefficients α and β as in the present embodiment. .

  When the image processing apparatus 100 is configured to perform two-dimensional noise reduction processing on the input video signal in addition to the three-dimensional noise reduction processing, the strength of the two-dimensional noise reduction processing for the pulsation portion is high. The intensity of the two-dimensional noise reduction process for the non-pulsation portion may be lower. In this case, since the three-dimensional noise reduction process is performed on the pulsation portion with higher intensity than the two-dimensional noise reduction process, it is possible to more effectively reduce noise generated as pulsation in the video.

  As described above, according to the image processing apparatus 100 according to the present embodiment, the control unit 130 determines that the intensity of the three-dimensional noise reduction process for the pulsating part is higher than the intensity of the three-dimensional noise reduction process for the non-pulsating part. As described above, since the three-dimensional noise reduction processing unit 120 is controlled, noise generated as pulsation in an image can be effectively reduced.

<Second embodiment>
An image processing apparatus according to the second embodiment will be described with reference to FIGS.

  FIG. 11 is a block diagram illustrating the configuration of the image processing apparatus according to the second embodiment. In FIG. 11, the same components as those in the first embodiment shown in FIGS. 1 to 10 are denoted by the same components, and the description thereof is omitted as appropriate.

  In FIG. 11, the image processing apparatus 200 according to the second embodiment includes the control unit 130 b instead of the control unit 130 in the first embodiment described above, and the image processing apparatus 100 according to the first embodiment described above. In other respects, the configuration is substantially the same as that of the image processing apparatus 100 according to the first embodiment described above.

  As illustrated in FIG. 11, the control unit 130 b is configured such that encoding information related to encoding is input from the decoder 810. It should be noted that the encoded image data transmitted from the transmission side described above with reference to FIG. 2 is input to the decoder 810 as an input encoded signal. The decoder 810 decodes this input encoded signal based on the encoded information and outputs it as an input video signal to the image processing apparatus 200.

  The encoded information includes an encoding method (for example, MPEG-2 method, H.264 method, VC-1 method, etc.) and an average bit rate in the transmission line 700.

  The encoding information includes not only the encoding method and the average bit rate, but also the quantization step and the encoding method are H.264. In the case of H.264, at least one of a loop filter coefficient, a GOP cycle, and a picture mode may be included. The loop filter coefficient is H.264. In the H.264 system, the ratio of the block boundary on which filter processing is performed among the block boundaries existing on one screen is shown.

  In the present embodiment, in particular, the control unit 130b determines the strength of the three-dimensional noise reduction processing for the pulsation part (more specifically, the three-dimensional noise reduction processing unit 120 based on the encoded information input from the decoder 810). The coefficient α) in the multiplier 125 is determined. In other words, the control unit 130b controls the coefficient α according to the encoded information input from the decoder 810.

  FIG. 12 is a graph illustrating an example of the control of the coefficient α according to the encoded information by the control unit according to the second embodiment.

  In FIG. 12, a control line L1 indicates control by the control unit 130b of the coefficient α corresponding to the average bit rate of the transmission line 700 (see FIG. 2) when the encoding method is the MPEG-2 method. The control line L2 has an encoding method of H.264. The control by the control unit 130b of the coefficient α corresponding to the average bit rate of the transmission line 700 in the case of the H.264 system is shown.

  The control unit 130b controls the coefficient α according to the control lines L1 and L2. For example, when the encoding method is the MPEG-2 method in a predetermined average bit rate interval, the control unit 130b determines that the encoding method is H.264. The coefficient α is set higher than in the H.264 system. The control unit 130b sets the coefficient α higher when the average bit rate is lower, and sets the coefficient α lower when the average bit rate is higher (see the control lines L1 and L2). .

  Here, the possibility of occurrence of pulsation varies depending on the encoding method and the average bit rate. That is, for example, when the encoding method is the MPEG-2 method, the encoding method is H.264. Since the compression rate for compressing image data at the time of encoding is lower than in the case of the H.264 system, a large amount of high-frequency band components of the image are lost to suppress the bit amount within a predetermined range, and a pulsation phenomenon may occur. Will increase. Also, for example, the lower the average bit rate, the more the high frequency band components of the image are lost in order to keep the bit amount within a predetermined range, and the possibility of occurrence of a pulsation phenomenon increases.

  However, as described above, particularly in the present embodiment, the control unit 130b controls the coefficient α in accordance with the encoded information input from the decoder 810. Therefore, the strength of the three-dimensional noise reduction process can be determined according to the possibility that the pulsation phenomenon occurs. In other words, it is possible to determine the strength of the three-dimensional noise reduction process so that the deterioration of the image quality in the pulsation portion is assumed to some extent and the assumed deterioration of the image quality is almost or completely eliminated. Therefore, noise generated as pulsation in an image can be reduced more effectively.

  That is, in the present embodiment, in particular, the control unit 130b, for example, when the average bit rate is high or the encoding method is H.264. In the case of the H.264 system, it is assumed that the noise generated as a pulsation phenomenon is small. Therefore, the coefficient α is set low, and conversely, when the average bit rate is low or the encoding system is the MPEG-2 system. In this case, since it is assumed that there is a lot of noise generated as a pulsation phenomenon, the coefficient α is set higher. In other words, the control unit 130b increases the strength of the three-dimensional noise reduction process for a pulsation part that is assumed to have a lot of noise generated as a pulsation phenomenon, and image quality deterioration due to noise generated as the pulsation phenomenon does not occur much. The three-dimensional noise reduction processing unit 120 is controlled so as to reduce the strength of the three-dimensional noise reduction processing for the assumed pulsation portion. Therefore, noise generated as pulsation in an image can be reduced more effectively.

<Third embodiment>
An image processing apparatus according to the third embodiment will be described with reference to FIGS.

  FIG. 13 is a block diagram illustrating the configuration of the image processing apparatus according to the third embodiment. In FIGS. 13 to 16, the same components as those in the first embodiment shown in FIGS. 1 to 10 are denoted by the same components, and the description thereof is omitted as appropriate.

  In FIG. 13, the image processing apparatus 300 according to the third embodiment includes a pulsation detection unit 110 c instead of the pulsation detection unit 110 in the first embodiment described above, and the control unit 130 in the first embodiment described above. Instead, the image processing apparatus 100 according to the first embodiment is different from the image processing apparatus 100 according to the first embodiment described above in that a control unit 130c is provided, and the other points are substantially the same as those of the image processing apparatus 100 according to the first embodiment described above. Yes.

  In FIG. 13, the pulsation detecting unit 110c detects a pulsation phenomenon in which an object shown in an image corresponding to an input video signal pulsates. The pulsation detection unit 110c outputs a pulsation level that is information indicating the degree of the detected pulsation phenomenon (that is, the degree of strength) to the control unit 130c.

  FIG. 14 is a block diagram illustrating a configuration of a pulsation detecting unit according to the third embodiment.

  In FIG. 14, the pulsation detection unit 110c includes a motion detection unit 111c, a motion vector detection unit 112c, a pulsation determination unit 113c, and a frame memory 114. The motion detection unit 111c is an example of the “difference detection unit” according to the present invention.

  The input video signal is input to the frame memory 114, the motion detection unit 111c, and the motion vector detection unit 112c.

  The motion detection unit 111c calculates the absolute value of the difference signal (that is, the difference absolute value) that is the difference between the luminance signal between the input video signal and the input video signal delayed by one frame input from the frame memory 114. . When the calculated difference absolute value is equal to or less than the predetermined threshold, the motion detection unit 111c outputs a difference level that is information indicating the calculated difference absolute value to the pulsation determination unit 113c.

  FIG. 15 is a block diagram illustrating a configuration of a motion detection unit according to the third embodiment.

  As illustrated in FIG. 15, the motion detection unit 111 c includes the threshold processing unit 1112 c in place of the threshold processing unit 1112 in the first example described above with reference to FIG. 4, according to the first example described above. Unlike the motion detection unit 111c, the other points are substantially the same as those of the motion detection unit 111c according to the first embodiment described above.

  The threshold processing unit 1112c compares the difference absolute value input from the difference absolute value processing unit 1111 with a predetermined threshold, and if the difference absolute value is equal to or less than the predetermined threshold, the difference level indicating the difference absolute value Is output to the pulsation determining unit 113c. On the other hand, when the difference absolute value is larger than the predetermined threshold, the threshold processing unit 1112c does not output the difference level to the pulsation determining unit 113c. Therefore, the difference level indicates the difference absolute value when the difference absolute value is equal to or smaller than the predetermined threshold, and also functions as a difference flag in the first embodiment described above. The threshold processing unit 1112c may be configured to output a difference flag to the pulsation determining unit 113c in addition to the difference level.

  In FIG. 14 again, the motion vector detection unit 112c detects a motion vector by the block matching method based on the input video signal and the input video signal delayed by one frame input from the frame memory 114. Note that other methods such as a gradient method may be used as a method by which the motion vector detection unit 112c detects the motion vector.

  The motion vector detection unit 112c outputs motion vector information indicating the magnitude of the detected motion vector to the pulsation determination unit 113c.

  FIG. 16 is a block diagram illustrating a configuration of a motion vector detection unit according to the third embodiment.

  As shown in FIG. 16, the motion vector detection unit 112c is different from the correlation determination unit 1122c in the first example described above with reference to FIG. 5 in that it includes a correlation determination unit 1122c. Unlike the motion vector detection unit 112c, the other configuration is substantially the same as that of the motion vector detection unit 112c according to the first embodiment described above.

  Based on the correlation information, correlation determination section 1122c determines whether the deviation between the position of the block one frame before giving the minimum difference absolute value and the position of the block in the corresponding current frame is within a predetermined range. A motion vector is detected by determining whether or not the difference level is equal to or less than a predetermined threshold. The correlation determination unit 1122c outputs motion vector information indicating the magnitude of the detected motion vector to the pulsation determination unit 113c.

  In FIG. 14 again, the pulsation determination unit 113c is based on the difference level input from the motion detection unit 111c and the motion vector information input from the motion vector detection unit 112c. And the estimated pulsation level is output to the control unit 130c.

  FIG. 17 is a graph illustrating an example of a pulsation level output by the pulsation determining unit according to the third embodiment.

  As shown in FIG. 17, the pulsation determining unit 113c decreases the pulsation level as the magnitude of the motion vector (that is, the motion vector scalar quantity) increases based on the motion vector information input from the motion vector detection unit 112c. Output. That is, the pulsation determining unit 113c estimates that the possibility that the pulsation phenomenon has occurred is lower as the magnitude of the motion vector is larger, and outputs a pulsation level having a smaller value.

  FIG. 18 is a graph illustrating another example of the pulsation level output by the pulsation determining unit according to the third embodiment.

  FIG. 18 shows changes in the pulsation level output from the pulsation determining unit 113c with respect to the difference level in four cases where the magnitudes of the motion vectors are different from each other.

  Of the control lines L3a, L3b, L3c, and L3d, the control line L3a shows the case where the magnitude of the motion vector (that is, the motion vector scalar quantity) is the largest, and the control line L3b has the magnitude of the motion vector. The second largest case is shown, the control line L3c shows the third largest motion vector, and the control line L3d shows the smallest motion vector magnitude.

  Based on the difference level input from the motion detection unit 111c and the motion vector information input from the motion vector detection unit 112c, the pulsation determination unit 113c increases and outputs the pulsation level as the difference level increases. As the vector size decreases, the pulsation level is raised and output (see control lines L3a, L3b, L3c and L3d).

  In FIG. 13 again, the control unit 130c determines whether the three-dimensional noise reduction processing unit 120 (more specifically, in the multiplier 125 of the three-dimensional noise reduction processing unit 120) according to the pulsation level output from the pulsation detection unit 110c. The coefficient α and the coefficient β) in the nonlinear processing unit 124 of the three-dimensional noise reduction processing unit 120 are controlled.

  Particularly in the present embodiment, the control unit 130c determines the strength of the three-dimensional noise reduction process for the pulsation part based on the pulsation level output from the pulsation detection unit 110c. Therefore, the intensity of the three-dimensional noise reduction process can be determined according to the degree of the pulsation phenomenon. That is, as described above with reference to FIGS. 17 and 18, the three-dimensional noise is determined according to the pulsation level estimated based on at least one of the difference level and the magnitude of the motion vector (that is, the motion vector scalar quantity). The intensity of the reduction process can be determined. In other words, the strength of the three-dimensional noise reduction process for the pulsation part can be controlled with higher accuracy. Therefore, noise generated as pulsation in an image can be reduced more effectively.

<Fourth embodiment>
An image processing apparatus according to the fourth embodiment will be described with reference to FIGS.

  FIG. 19 is a block diagram illustrating the configuration of the image processing apparatus according to the fourth embodiment. 19 and 20, the same components as those in the first embodiment shown in FIGS. 1 to 10 are denoted by the same components, and the description thereof is omitted as appropriate.

  In FIG. 19, the image processing apparatus 400 according to the fourth embodiment includes the pulsation detecting unit 110d instead of the pulsation detecting unit 110 in the first embodiment described above, and the image processing apparatus according to the first embodiment described above. Unlike the image processing apparatus 100, the other configuration is substantially the same as that of the image processing apparatus 100 according to the first embodiment described above.

  As illustrated in FIG. 19, the pulsation detection unit 110 d is configured such that encoding information related to encoding is input from a decoder 810.

  The encoded information includes at least one of a GOP cycle, a picture mode, and a prediction mode in intra-frame prediction encoding.

  The encoding information includes GOP cycle, picture mode, prediction mode in intra prediction encoding, encoding method, average bit rate, quantization step, and encoding method. At least one of the loop filter coefficients in the case of the H.264 system may be included.

  The pulsation detection unit 110d detects a pulsation phenomenon in the video corresponding to the input video signal, and outputs the detection result to the control unit 130 as a pulsation flag. Here, particularly in the present embodiment, the pulsation detecting unit 110d detects the pulsation phenomenon based on the encoded information in addition to the input video signal.

  FIG. 20 is a block diagram illustrating a configuration of a pulsation detecting unit according to the fourth embodiment.

  In FIG. 20, the pulsation detecting unit 110d according to the fourth example includes the pulsation determining unit 113d in place of the pulsation determining unit 113 in the first example described above with reference to FIG. Unlike the pulsation detection unit 110 according to the example, the other points are configured in substantially the same manner as the pulsation detection unit 110 according to the first embodiment described above.

  Particularly in the present embodiment, the pulsation determining unit 113d includes the encoded information input from the decoder 810 in addition to the difference flag input from the motion detecting unit 111 and the motion vector flag input from the motion vector detecting unit 112. Based on the above, it is determined whether or not a pulsation phenomenon has occurred in the video (that is, whether or not the currently input portion of the input video signal is a pulsation portion), and the determination result is controlled as a pulsation flag. To the unit 130.

  More specifically, the pulsation determination unit 113d determines whether the pulsation determination unit 113 according to the first embodiment described above has a determination method (that is, if both the difference flag and the motion vector flag are in an on state) In addition to determining that a phenomenon has occurred and determining that at least one of the difference flag and the motion vector flag is in an off state, a pulsation phenomenon has not occurred in the video) The pulsation phenomenon is detected by a determination method based on

  For example, the pulsation phenomenon is likely to occur in the GOP cycle, and the degree or level of each I picture tends to increase. Therefore, the pulsation determination unit 113d causes the change corresponding to the characteristic to occur in the input video signal. Is determined based on the GOP cycle and the picture mode included in the encoded information.

  Alternatively, for example, when the prediction mode of intra prediction encoding is used in encoding at the encoder 910 (see FIG. 2), the pulsation phenomenon occurs when the prediction mode of intra prediction encoding changes. Since there is a tendency to occur, the pulsation determining unit 113d determines whether or not a pulsation phenomenon has occurred in the video based on whether or not the prediction mode of intra prediction encoding included in the encoding information changes. judge.

  As described above, in this embodiment, in particular, in addition to the difference flag and the motion vector flag, for example, a pulsation phenomenon is detected based on encoding information such as a GOP cycle, a picture mode, and a prediction mode in intra prediction encoding. Therefore, the pulsation phenomenon can be detected with high accuracy, in other words, the pulsation portion and the non-pulsation portion in the input video signal can be distinguished with high accuracy.

  Therefore, according to the image processing apparatus 400 according to the present embodiment, the three-dimensional noise reduction process is more reliably performed on the pulsating portion with high intensity, and the lower intensity is more reliably performed on the non-pulsating portion. Thus, it is possible to perform a three-dimensional noise reduction process.

<Fifth embodiment>
An image processing apparatus according to the fifth embodiment will be described with reference to FIG.

  FIG. 21 is a block diagram illustrating the configuration of the three-dimensional noise reduction processing unit according to the fifth embodiment together with the control unit according to the fifth embodiment.

  The image processing apparatus according to the fifth embodiment includes a three-dimensional noise reduction processing section 120e instead of the three-dimensional noise reduction processing section 120 in the first embodiment described above, and the control section 130 in the first embodiment described above. Instead of the image processing apparatus 100 according to the first embodiment described above in that a control unit 130e is provided instead, the other configuration is substantially the same as the image processing apparatus 100 according to the first embodiment described above. ing.

  The three-dimensional noise reduction processing unit 120e includes five frame memories 121e to 125e, a moving average filter unit 126e, a first multiplier 127e, a second multiplier 128e, and an adder 129e.

  The input video signal is input to the first frame memory 121e and the moving average filter unit 126e.

  The first frame memory 121e delays the input video signal by one frame and outputs it to the moving average filter unit 126e and the second frame memory 122e.

  The second frame memory 122e delays the video signal input from the first frame memory 121e by one frame and outputs it to the moving average filter unit 126e and the third frame memory 122e.

  The third frame memory 123e delays the video signal input from the second frame memory 122e by one frame and outputs it to the moving average filter unit 126e and the fourth frame memory 124e.

  The fourth frame memory 124e delays the video signal input from the third frame memory 123e by one frame and outputs it to the moving average filter unit 126e and the fifth frame memory 125e.

  The fifth frame memory 125e delays the video signal input from the fourth frame memory 124e by one frame, and outputs it to the moving average filter unit 126e.

  As described above, the input video signal is sequentially delayed by one frame by the five frame memories 121e to 125e, and the input video signal delayed by one frame is output from the first frame memory 121e to the moving average filter unit 126e. Then, the input video signal delayed by 2 frames is output from the second frame memory 122e to the moving average filter unit 126e, and the input video signal delayed by 3 frames is output from the third frame memory 123e by the moving average filter. The input video signal output to the unit 126e, delayed by 4 frames from the fourth frame memory 124e, is output to the moving average filter unit 126e, and input video signal delayed by 5 frames from the fifth frame memory 125e. Is output to the moving average filter unit 126e.

  The moving average filter unit 126e includes an input video signal, an input video signal delayed by one frame, an input video signal delayed by two frames, an input video signal delayed by three frames, and four frames. Calculating the average of the input video signals related to the six frames having different time from the input video signal delayed by the amount of time and the input video signal delayed by the amount of 5 frames, and the first multiplier as the average signal S1 To 127e.

  The first multiplier 127e multiplies the average signal S1 output from the moving average filter 126e by the coefficient k1 and outputs the result to the adder 129e. The coefficient k1 is controlled by the control unit 130e.

  The second multiplier 128e multiplies the signal S2 output from the second frame memory 122e (ie, the input video signal delayed by two frames) S2 by the coefficient k2 and outputs the result to the adder 129e. The coefficient k2 is controlled by the control unit 130e. The coefficients k1 and k2 are set so that the relational expression k2 = 1−k1 is established between the coefficient k1 and the coefficient k2 (that is, the sum of the value of the coefficient k1 and the value of the coefficient k2 is 1. To be controlled).

  The adder 129e adds the signal output from the first multiplier 127e and the signal output from the second multiplier 128e, and outputs the result as an output video signal.

  As described above, the three-dimensional noise reduction processing unit 120e adds the signal obtained by multiplying the average signal S1 by the coefficient k1 and the signal S2 delayed by two frames by the coefficient k2, thereby adding the input video signal. 3D noise reduction processing is performed to remove noise included in the.

  In accordance with the pulsation flag output from the pulsation detection unit 110, the control unit 130e is a three-dimensional noise reduction processing unit 120e (more specifically, a coefficient k1 in the first multiplier 127e of the three-dimensional noise reduction processing unit 120e, And the coefficient k2) in the second multiplier 128e of the three-dimensional noise reduction processing unit 120e is controlled.

  Particularly in the present embodiment, when the control unit 130e performs the three-dimensional noise reduction process on the pulsation part according to the pulsation flag (that is, when the pulsation flag is on), the control unit 130e performs the three-dimensional process on the non-pulsation part. The coefficient k1 multiplied by the average signal S1 by the first multiplier 127e is larger (in other words, by the second multiplier 128e than when the noise reduction process is performed (that is, when the pulsation flag is off). The coefficients k1 and k2 are controlled so that the coefficient k2 multiplied by the signal S2 becomes smaller. In other words, the control unit 130e controls the coefficients k1 and k2 in this manner so that the intensity of the three-dimensional noise reduction process for the pulsation part is higher than the intensity of the three-dimensional noise reduction process for the non-pulsation part. The three-dimensional noise reduction processing unit 120e is controlled.

  Therefore, since the video corresponding to the pulsation portion can be smoothed with time, the pulsation phenomenon in the video can be reliably reduced.

  The present invention includes, for example, a digital broadcast receiver and an electronic device (for example, a liquid crystal television, a PDP television, a DVD recorder, a blue-ray recorder, an HDD recorder, a cable broadcast receiving terminal, a satellite broadcast receiving terminal, It can be applied to an IP broadcast receiving terminal, a car navigation system, a mobile phone, a personal computer, a one-segment receiving device, etc.), decoding software, and the like.

  The present invention is not limited to the above-described embodiments, and can be appropriately changed without departing from the gist or concept of the invention that can be read from the claims and the entire specification. The method and the decoding device are also included in the technical scope of the present invention.

It is a block diagram which shows the structure of the image processing apparatus which concerns on 1st Example. It is a block diagram which shows the structure of the digital broadcasting system provided with the image processing apparatus which concerns on 1st Example. It is a block diagram which shows the structure of the pulsation detection part which concerns on 1st Example. It is a block diagram which shows the structure of the motion detection part which concerns on 1st Example. It is a block diagram which shows the structure of the motion vector detection part which concerns on 1st Example. It is a conceptual diagram which shows an example of the detection of the motion vector by the motion vector detection part which concerns on 1st Example. It is a conceptual diagram which shows the other example of the detection of the motion vector by the motion vector detection part which concerns on 1st Example. It is a block diagram which shows the structure of the three-dimensional noise reduction process part which concerns on 1st Example with a control part. In the non-linear processing unit according to the first embodiment, the passband is shown in each case when the value of the coefficient β is 1 and when the value of the coefficient β is 0. It is a flowchart for demonstrating the flow of operation | movement of the control part which concerns on 1st Example. It is a block diagram which shows the structure of the image processing apparatus which concerns on 2nd Example. It is a graph which shows an example of control of coefficient alpha according to coding information by the control part concerning the 2nd example. It is a block diagram which shows the structure of the image processing apparatus which concerns on 3rd Example. It is a block diagram which shows the structure of the pulsation detection part which concerns on 3rd Example. It is a block diagram which shows the structure of the motion detection part which concerns on 3rd Example. It is a block diagram which shows the structure of the motion vector detection part which concerns on 3rd Example. It is a graph which shows an example of the pulsation level which the pulsation determination part which concerns on 3rd Example outputs. It is a graph which shows the other example of the pulsation level which the pulsation determination part which concerns on 3rd Example outputs. It is a block diagram which shows the structure of the image processing apparatus which concerns on 4th Example. It is a block diagram which shows the structure of the pulsation detection part which concerns on 4th Example. It is a block diagram which shows the structure of the three-dimensional noise reduction process part which concerns on 5th Example with a control part.

Explanation of symbols

100, 200, 300, 400 Image processing device 110, 110c, 110d Pulsation detection unit 111, 111c Motion detection unit 112, 112c Motion vector detection unit 113, 113c, 113d Pulsation determination unit 120, 120e Three-dimensional noise reduction unit 130, 130b , 130c, 130e Control unit 810 Decoder 910 Decoder

Claims (8)

An image processing device that transmits image data encoded on a transmission side to a reception side and performs image processing on an input video signal decoded on the reception side,
Three-dimensional noise reduction processing means for performing three-dimensional noise reduction processing as the image processing;
Pulsation detecting means for detecting a pulsation phenomenon in which an object reflected in an image corresponding to the input video signal pulsates;
The intensity of the three-dimensional noise reduction process for the pulsation part where the pulsation phenomenon is detected in the input video signal is greater than the intensity of the three-dimensional noise reduction process for the non-pulsation part excluding the pulsation part of the input video signal. An image processing apparatus comprising: control means for controlling the three-dimensional noise reduction processing means so as to be higher. The pulsation detecting means includes
Difference detection means for detecting whether or not a difference signal between two frames or fields having different times in the input video signal is greater than a predetermined threshold;
Motion vector detecting means for detecting a motion vector based on the difference signal;
When the difference detection unit detects that the difference signal is greater than the predetermined threshold and the magnitude of the motion vector detected by the motion vector detection unit is smaller than a predetermined magnitude, the pulsation phenomenon The image processing apparatus according to claim 1, further comprising: a pulsation determining unit that determines that the pulsation occurs. The said control means determines the intensity | strength of the said three-dimensional noise reduction process with respect to the said pulsation part based on at least one of the signal level of the said difference signal, and the magnitude | size of the said motion vector. Image processing apparatus. The control means includes
The encoded image data is decoded and output as the input video signal and connected to decoding means for outputting the encoded information related to the encoding,
The image processing apparatus according to any one of claims 1 to 3, wherein an intensity of the three-dimensional noise reduction process for the pulsation portion is determined based on the encoded information. The pulsation detecting means includes
The encoded image data is decoded and output as the input video signal and connected to decoding means for outputting the encoded information related to the encoding,
The image processing apparatus according to any one of claims 1 to 4, wherein the pulsation phenomenon is detected based on the encoded information. Two-dimensional noise reduction processing means for performing two-dimensional noise reduction processing on the input video signal;
The control means controls the two-dimensional noise reduction processing means so that the strength of the two-dimensional noise reduction processing for the pulsating portion is lower than the strength of the two-dimensional noise reduction processing for the non-pulsating portion. The image processing apparatus according to claim 1, wherein: An image processing method in which image data encoded on a transmission side is transmitted to a reception side, and image processing is performed on an input video signal decoded on the reception side,
A three-dimensional noise reduction processing step for performing a three-dimensional noise reduction process as the image processing;
A pulsation detecting step of detecting a pulsation phenomenon in which an object reflected in an image corresponding to the input video signal pulsates;
The intensity of the three-dimensional noise reduction process for the pulsation part where the pulsation phenomenon is detected in the input video signal is greater than the intensity of the three-dimensional noise reduction process for the non-pulsation part excluding the pulsation part of the input video signal. And a control step for controlling the three-dimensional noise reduction processing means so as to be higher. The image processing apparatus according to any one of claims 1 to 6,
A decoding device comprising: decoding means for decoding the encoded image data and outputting the decoded image data as the input video signal.

Images