JP2008227928A - Frame rate converter and video display device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a frame rate converter, along with a video display device, capable of generating an appropriate interpolating video for smooth video reproduction even at such points as video content significantly changes, for example at a scene change. <P>SOLUTION: The frame rate converter comprises a correlation judging means which judges whether there exists correlation between adjoining, or succeeding, frames, a first interpolating frame generating means which, if judged that there exists correlation, generates interpolation information for generating an interpolating frame that is inserted between the frames based on that frames, and generates the interpolating frame based on the acquired interpolation information, and a second interpolating frame generating means which, if judged there exists no correlation, predicts the interpolation information for generating a current interpolating frame based on the interpolation information used for the previous interpolating frame, and generates an interpolating frame based on the predicted interpolation information. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a frame rate conversion device and a video display device.

  Video display devices capable of displaying content at a frame rate higher than that of existing content have been developed. For example, in a liquid crystal television, in order to prevent a moving image from being blurred, an image of 60 frames per second is converted into an image of 120 frames per second, and the obtained 120 frames of images are displayed on a liquid crystal display. Has been developed. When displaying content on such a video display device, the video of the same frame is not simply output a plurality of times, but interpolated video between frames is generated by signal processing and inserted between the frames to smoothly To get a good playback video.

  The interpolated video is normally generated using the frame video before and after the interpolated video, but there is a problem that a good interpolated video cannot be obtained at a location where the video content changes greatly, such as a scene change.

  In the “image rate conversion method and image rate conversion device” described in Japanese Patent Laid-Open No. 2006-270823, (1) the image of the previous frame is applied as it is, or (2) where there is no correlation between the previous and next frames. By using motion vectors between the previous and next frames, motion compensation is performed in only one direction from the previous frame to generate an interpolation frame.

However, in (1) above, rattling occurs at the moment of the scene change. In the above (2), since an inaccurate motion vector generated using a frame across scene changes is used, even if an interpolation frame is generated only from the image of the previous frame, the image is distorted. End up.
JP 2006-270823 A

  An object of the present invention is to provide a frame rate conversion device and a video display device capable of generating a suitable interpolation frame and displaying a smooth reproduced video even at a location where video content changes greatly, such as a scene change. .

  An object of the present invention is to provide a frame rate conversion device and a video display device capable of generating a smooth interpolation frame even in a scene such as fade-in or fade-out.

  According to the first aspect of the present invention, in the frame rate conversion apparatus, a correlation determination unit that determines whether there is a correlation between adjacent front and rear frames. First interpolation frame generation means for generating interpolation information for generating an interpolation frame to be inserted between the two frames and generating an interpolation frame based on the obtained interpolation information, and determined to have no correlation In this case, second interpolation information is generated based on the interpolation information used to generate the current interpolation frame based on the interpolation information used for generating the previous interpolation frame, and the interpolation frame is generated based on the predicted interpolation information. The interpolation frame generating means is provided.

  According to a second aspect of the present invention, in the frame rate conversion apparatus according to the first aspect, the second interpolation frame generation means generates an interpolation frame using only the previous frame of the two frames. It is characterized by being.

  In the third aspect of the present invention, when it is determined that there is no correlation by the first correlation determination unit and the first correlation determination unit that determine whether or not there is a correlation between adjacent frames. After the brightness levels of both frames are aligned, the second correlation determining unit that determines whether there is a correlation between the frames and the second correlation determining unit determine that there is a correlation. Interpolation information for generating an interpolation frame to be inserted between both frames after the level has been aligned is generated, and an interpolation frame is generated based on the obtained interpolation information It is characterized by that.

  According to a fourth aspect of the present invention, in the frame rate conversion apparatus according to the third aspect, when the first correlation determining unit determines that there is a correlation, the frame is converted from the two frames to the interval between the two frames. When it is determined that there is no correlation by means for generating interpolation information for generating an interpolation frame to be inserted into the frame, and by means for generating an interpolation frame based on the obtained interpolation information and the second correlation determination means And means for predicting the interpolation information for generating the current interpolation frame based on the interpolation information used for generating the previous interpolation frame, and generating the interpolation frame based on the predicted interpolation information. It is characterized by that.

  According to a fifth aspect of the present invention, in the video display device, the frame rate conversion device according to any one of the first to fourth aspects is provided.

  According to the present invention, it is possible to generate a suitable interpolation frame and display a smooth reproduced video even at a location where the video content changes greatly, such as a scene change.

  Also, according to the present invention, a smooth interpolation frame can be generated even in a scene such as fade-in or fade-out.

  DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments in the case where the present invention is applied to a frame rate conversion device for converting 60 frames per second video into a 120 frames per second provided in a liquid crystal display device will be described below with reference to the drawings. .

[1] Explanation of basic concept of Example 1

  In this specification, as shown in FIG. 1, when an interpolation frame Fc (N) is generated based on two frames F (N) and F (N + 1) before and after, the temporally old frame (N) Is referred to as a previous frame or a current frame, and a temporally new frame F (N + 1) is referred to as a subsequent frame or a reference frame.

  With reference to FIG. 2, the basic concept of the first embodiment will be described.

  In the first embodiment, the interpolation frame generation method is changed between the case where there is a correlation between two adjacent frames and the case where there is no correlation between two adjacent frames. FIG. 2 shows an example in which two adjacent frames F (N) and F (N + 1) have a correlation and two adjacent frames F (N + 1) and F (N + 2) have no correlation. .

  When the interpolated frame Fc (N) is generated from the two frames F (N) and F (N + 1) that are correlated with each other, there is motion based on the images of both the frames F (N) and F (N + 1). An area (motion area: an area indicated by diagonal lines in the middle diagram in FIG. 2) is specified, and a motion vector (in the middle diagram in FIG. 2) is set for each of a plurality of blocks set by dividing the video area. , 0 or →). The information on the motion area and the motion vector become interpolation information for generating an interpolation frame. The movement area is specified by determining whether or not there is a video change for each pixel, and specifying an aggregate of pixels having a video change as the movement area.

  For pixels included in an area where there is no motion, the image of the corresponding target pixel in one of the previous and subsequent frames is used as an interpolated image as it is. For the pixels included in the non-motion area, the average of the corresponding target pixel image in the previous frame F (N) and the corresponding target pixel image in the subsequent frame F (N + 1) may be used as the interpolation image. . For pixels included in an area with motion, an interpolated image is extracted from one or both of the preceding and following frames F (N) and F (N + 1) by extracting an image at a pixel position determined based on the motion vector for the target pixel. Use as

  When generating an interpolated frame Fc (N + 1) from two frames F (N + 1) and F (N + 2) that are not correlated with each other, an interpolated frame Fc (N + 1) is generated from both frames F (N + 1) and F (N + 2). The motion area and the motion vector for generating the interpolation frame Fc (N + 1) are predicted based on the motion area and the motion vector obtained last time, instead of obtaining the motion area and the motion vector for performing the motion.

  Then, an interpolated frame Fc (N + 1) is generated based on the predicted motion area and motion vector. That is, for the pixels included in the non-motion area, the corresponding target pixel image of the previous frame F (N + 1) is used as an interpolation image as it is. For pixels included in an area with motion, an image at a pixel position determined based on the motion vector for the target pixel is extracted from the previous frame F (N + 1) and used as an interpolated image.

As described above, when the interpolation frame Fc (N + 1) is generated from the two frames (N + 1) and (N + 2) that are not correlated with each other, the interpolation frame Fc (N) is generated only from the previous frame.
[2] Description of electrical configuration of frame rate conversion device

  FIG. 3 shows an electrical configuration of the frame rate conversion apparatus.

  The input video signal is sent to the delay circuit 1, the frame correlation determination unit 2, the motion area generation unit 3, the motion vector generation unit 4, and the interpolation frame synthesis unit 6. The input video signal sent to the delay circuit 1 is delayed by one frame period, and then sent to the frame correlation determination unit 2, motion area generation unit 3, motion vector generation unit 4 and interpolation frame synthesis unit 6.

  A frame input to the delay circuit 1 is represented by F (N + 1), and a frame output from the delay circuit 1 is represented by F (N). Since the interpolated frame Fc (N) is generated from F (N) and F (N + 1), F (N) is referred to as the previous frame (current frame), and F (N + 1) is referred to as the subsequent frame (reference frame). To do.

The frame correlation determination unit 2 determines whether or not there is a correlation between F (N + 1) and F (N). Whether there is a correlation between F (N + 1) and F (N) can be performed using a known technique. For example, as described in JP-A-2006-270823, by using the values of one or both of the inter-block correlation D _v of the block matching error amount D _M and MV, F (N + 1) and F (N) It may be determined whether or not there is a correlation.

  The frame correlation determination unit 2 sends a signal r representing the determination result of whether or not there is a correlation to the motion area / motion vector correction unit 5 and the interpolated frame synthesis unit 6. In this example, r = 1 when there is a correlation, and r = 0 when there is no correlation.

  The motion area generation unit 3 identifies a pixel having a change in the video between F (N + 1) and F (N). That is, the absolute value of the difference between F (N + 1) and F (N) is obtained for each pixel, and when the absolute value of the difference is equal to or greater than a predetermined threshold, the pixel is changed to a pixel with a change (motion A certain pixel). The motion area generation unit 3 generates motion area information M ′ (N) including information indicating whether or not each pixel has a motion and sends the motion area information M ′ (N) to the motion area / motion vector correction unit 5.

  The motion vector generation unit 4 calculates a motion vector between F (N + 1) and F (N). Specifically, the motion vector generation unit 4 obtains a motion vector for each of a plurality of blocks set by dividing a video area using an existing method such as block matching. The motion vector generation unit 4 generates a motion vector V ′ (N) obtained for each block and sends it to the motion area / motion vector correction unit 5.

  FIG. 4 shows an electrical configuration of the motion area / motion vector correction unit 5.

  The motion area / motion vector correction unit 5 includes a signal switching circuit 51, a delay circuit 52, and a motion area / motion vector prediction unit 53.

  The signal switching circuit 51 includes, as motion area information, motion area information M ′ (N) generated by the motion area generation unit 3 and motion area prediction information M ″ generated by the motion area / motion vector prediction unit 53 ( Further, the signal switching circuit 51 receives the motion vector V ′ (N) generated by the motion vector generation unit 4 and the motion area / motion vector prediction unit 53 as a motion vector for each block. The predicted value V ″ (N) of the motion vector generated by is input.

  The signal switching circuit 51 switches the input signal based on the signal r sent from the frame correlation determination unit 2. That is, when the frame correlation determination unit 2 determines that there is a correlation between the two frames F (N) and F (N + 1) (r = 1), the signal switching circuit 51 is The generated motion area information M ′ (N) is output as motion area information M (N), and the motion vector V ′ (N) generated by the motion vector generation unit 4 is output as a motion vector V (N). .

  When the frame correlation determination unit 2 determines that there is no correlation between the two frames F (N) and F (N + 1) (r = 0), the motion area generated by the motion area / motion vector prediction unit 53 Is output as motion area information M (N), and the motion vector prediction value V ″ (N) generated by the motion area / motion vector prediction unit 53 is used as the motion vector V (N). Output as.

  The motion area information M (N) and motion vector V (N) output from the signal switching circuit 51 are sent to the delay circuit 52 and also sent to the interpolation frame synthesis unit 6. The interpolation frame synthesis unit 6 generates an interpolation frame Fc (N) based on the motion area information M (N) and the motion vector V (N) sent from the signal switching circuit 51. Details of the processing performed by the interpolation frame synthesis unit 6 will be described later.

  The delay circuit 52 delays the motion area information M (N) and the motion vector V (N) sent from the signal switching circuit 51 by one frame period, and then sends them to the motion area / motion vector prediction unit 53. The motion area / motion vector prediction unit 53 receives the motion area information M (N−1) for the (N−1) th frame and the motion vector V ((N−1) th frame) sent from the delay circuit 52 ( Based on (N-1), the motion area and motion vector for the Nth frame are predicted and output as M ″ (N) and V ″ (N). In this case, the motion area information M (N−1) and the motion vector V (N−1) correspond to “interpolation information used for generating the previous interpolation frame” in claim 1.

  FIG. 5 shows a prediction processing procedure executed by the motion area / motion vector prediction unit 53.

  As shown in FIG. 6, it is assumed that the video area is divided into m × n blocks, and a motion vector V ′ (N) is calculated for each block. The position of each block is represented by a variable b representing 1 to m × n. Each block in a certain frame number is represented by B (b, frame number) as shown in FIG. For example, B (1, N-1) indicates the first block of the (N-1) th frame.

  Further, the motion area information corresponding to each block in a certain frame number is represented by M (b, frame number) as shown in FIG. For example, M (1, N−1) indicates motion area information corresponding to the first block of the (N−1) th frame. The motion area information M (1, N-1) is configured by information indicating whether or not there is a motion area for each pixel in the block. In FIG. 7, hatching indicates a motion area represented by the motion area information M (N−1) of the (N−1) th frame.

  A motion vector corresponding to each block in a certain frame number is represented by V (b, frame number). For example, V (1, N−1) indicates a motion vector corresponding to the first block of the (N−1) th frame.

  Returning to FIG. 5, first, initial setting is performed (step S1). That is, the initial value of the motion vector prediction value V ″ (b, N) for each block is set to 0, and the initial value of the motion area prediction information M ″ (N) is set to 0 (no motion region). ) Is set.

  After the initial setting, 1 is set to the variable b representing the block position (step S2). Then, it is determined whether or not the condition that the processing target block B (b, N−1) includes a motion area and the motion vector V (b, N−1) of the block is not 0 is satisfied ( Step S3). Whether or not the processing target block B (b, N-1) includes a motion area is determined based on the motion area information M (b, N-1) corresponding to the block.

  If the above condition is not satisfied, it is determined whether or not the processing for all the block positions has been completed (step S9). That is, it is determined whether or not b = m × n. If b = m × n is not satisfied, it is determined that processing for all the block positions is not completed, b is incremented by 1 (step S10), and the process returns to step S3.

  In step S3, it is determined that the condition that the processing target block B (b, N-1) includes a motion area and the motion vector V (b, N-1) of the block is not 0 is satisfied. In such a case, the motion area information M (b, N-1) corresponding to the processing target block is shifted to the content of the motion vector V (b, N-1) by the magnitude of the motion vector. The motion area prediction information M ″ (N) is updated (step S4).

  That is, the updated prediction information M ″ (N) indicates the motion area in the processing target block B (b, N−1) in the direction of the motion vector V (b, N−1) corresponding to the block. This is information representing the motion area after being moved by the magnitude of the motion vector.

  Next, whether or not at least a part of the moving area in the processing target block B (b, N-1) has moved into a block adjacent to the processing target block B (b, N-1) by the shift. It is determined whether or not (step S5). When it is determined that the motion area has moved within the adjacent block, the motion vector prediction value V ″ (N) corresponding to the adjacent block to which the motion area has moved is added to the processing block B (b, N−1). The motion vector V (b, N-1) is set (step S6), and the process proceeds to step S7, and if it is determined in step S5 that the motion area has not moved to the adjacent block, step S7 is performed. Migrate to

  In step S7, it is determined whether or not the motion area is included in the prediction information M ″ (b, N) corresponding to the block position b among the motion area prediction information M ″ (N) updated in step S4. Is determined. That is, when the motion area information M (b, N-1) is shifted in step S4, it is determined whether or not a motion area remains in the block after the shift.

  When the motion area is included in the motion area prediction information M ″ (b, N), the motion vector prediction value V ″ (b, N) corresponding to the processing target block B (b, N−1). Then, the motion vector V (b, N-1) corresponding to the processing target block B (b, N-1) is set (step S8). Then, the process proceeds to step S9. If it is determined in step S7 that the motion area is not included in the motion area prediction information M ″ (b, N), the process proceeds to step S9. In this case, the processing target block B (b , N−1), the motion vector predicted value V ″ (b, N) is 0 (initial value), which is a vector different from V (b, N−1).

  In step S9, it is determined whether or not processing for all blocks has been completed. That is, it is determined whether or not b = m × n. If b = m × n is not satisfied, it is determined that processing for all the blocks has not been completed, b is incremented by 1 (step S10), and the process returns to step S3.

  FIG. 8 shows an interpolation frame synthesis process procedure performed by the interpolation frame synthesis unit 6.

  A pixel is represented by P (= (x, y)). First, the first pixel (0, 0) is selected (step S21).

  It is determined whether or not the target pixel P is included in the motion area (step S22). This determination is performed based on the motion area information M (N) given to the interpolated frame synthesis unit 6 by the motion area / motion vector correction unit 5.

  When the target pixel P is not included in the motion area, the interpolation pixel value Fc (P, N) corresponding to the target pixel P is calculated based on the following equation (1), and the interpolation pixel value Fc (P , N) (step S23).

  Fc (P, N) = F (P, N) (1)

In the above equation (1), F (P, N) is a pixel value F (P, N) corresponding to the target pixel P of the current frame (N).

  When the process of step S23 is completed, it is determined whether or not the process for all the pixels has been completed (step S28). If the processing for all the pixels has not been completed, the target pixel P is updated to the next pixel (step S29), and the process returns to step S22.

  If it is determined in step S22 that the target pixel P is included in the motion area, the motion vector V (N (N) for each block given to the interpolated frame synthesis unit 6 by the motion area / motion vector correction unit 5 is determined. ), The motion vector of the block including the target pixel P is set as a motion vector v (= (vx, vy)) corresponding to the target pixel P (step S24).

  Further, it is determined whether or not the correlation determination result r given to the interpolated frame synthesis unit 6 by the frame correlation determination unit 2 is 1 (step S25). When r = 1, that is, when it is determined that there is a correlation between the current frame and the reference frame, the interpolated pixel value Fc (corresponding to the target pixel P is calculated based on the following equation (2). P, N) is calculated and set as the interpolated pixel value Fc (P, N) (step S26). Then, the process proceeds to step S28.

Fc (P, N) = (1/2) · F (P−v / 2, N)
+ (1/2) · F (P + v / 2, N + 1) (2)

  In the above equation (2), F (P−v / 2, N) is ½ of the magnitude of the motion vector v corresponding to the target pixel P with respect to the target pixel P of the current frame (N). The pixel value at a position in the direction opposite to the direction of the motion vector v is represented. Further, F (P + v / 2, N + 1) is in the direction of the motion vector v by 1/2 of the magnitude of the motion vector v corresponding to the target pixel P with respect to the target pixel P of the reference frame (N + 1). It represents the pixel value of the position.

  If it is determined in step S25 that r = 0, that is, if it is determined that there is no correlation between the current frame and the reference frame, the target pixel P is calculated based on the following equation (3). The interpolation pixel value Fc (P, N) corresponding to is calculated and set as the interpolation pixel value Fc (P, N) (step S27). Then, the process proceeds to step S28.

  Fc (P, N) = F (P−v / 2, N) (3)

  In the above equation (3), F (P−v / 2, N) is ½ of the magnitude of the motion vector v corresponding to the target pixel P with respect to the target pixel P of the current frame (N), The pixel value at a position in the direction opposite to the direction of the motion vector v is represented.

  When the interpolated pixel value Fc (P, N) is set for all the pixels, the result of step S28 is YES, and the current interpolated frame synthesis process is terminated.

[1] Description of Basic Concept of Second Embodiment Also in the second embodiment, it is basically assumed that an interpolation frame is generated by the same method as in the first embodiment.

  There is a case where fade-out (the overall luminance is gradually decreased) immediately before the scene change, or fade-in (an overall luminance is gradually increased) immediately after the scene change.

  As shown in FIG. 9, during the fade-out (or fade-in), the overall luminance differs between two adjacent frames F (N) and F (N + 1) even if they are continuous images. Therefore, there is a high possibility that it is determined that there is no correlation. Then, even if there is a substantial correlation between both frames F (N) and F (N + 1), an interpolated frame is created using the correlation between both frames F (N) and F (N + 1). It will not be done.

  Therefore, in the second embodiment, when it is determined that there is no correlation between both frames F (N) and F (N + 1), the entire luminance of both frames F (N) and F (N + 1) is aligned, Check again whether there is a correlation. If it is determined that there is a correlation, an interpolated frame is generated based on both frames F (N) and F (N + 1) in which the overall luminance is uniform. Then, the overall luminance of the obtained interpolation frame is adjusted to an intermediate luminance between both frames F (N) and F (N + 1).

[2] Description of electrical configuration of frame rate conversion device

  FIG. 10 shows an electrical configuration of the frame rate conversion apparatus.

  The input video signal is sent to the delay circuit 1, the frame correlation determination unit 2 and the luminance correction unit 7. After the input video signal sent to the delay circuit 1 is delayed by one frame period, the frame correlation determination unit 2, the luminance correction unit 7, the motion area generation unit 3, the motion vector generation unit 4, and the interpolated frame synthesis unit 6 Sent to.

  An image input to the delay circuit 1 is represented by F (N + 1), and an image output from the delay circuit 1 is represented by F (N).

  The frame correlation determination unit 2 determines whether there is a correlation between F (N + 1) and F (N). The frame correlation determination unit 2 sends a determination result r on whether there is a correlation to the luminance correction unit 7. In this example, r = 1 when there is a correlation, and r = 0 when there is no correlation.

  FIG. 11 shows an electrical configuration of the luminance correction unit 7.

  The luminance correction unit 7 includes an overall luminance difference detection unit 71, an overall luminance adjustment unit 72, a frame correlation determination unit 73, an OR circuit 74, and a signal switching unit 75.

The overall brightness difference detection unit 71 detects a brightness difference between the frames F (N) and F (N + 1). Specifically, assuming that the average luminance of F (N) is Y _N and the average luminance of F (N + 1) is Y _{N + 1} , F (N) and F (N + 1) based on the following equation (4) The luminance difference Y ′ is calculated.

Y ′ = Y _N −Y _{N + 1} (4)

The overall luminance adjustment unit 72 adds a luminance difference Y ′ to the luminance of each pixel of F (N + 1) in order to align the average luminance of F (N + 1) with the average luminance Y _N of F (N).

  The frame correlation determination unit 73 determines whether or not there is a correlation between F (N) and F ″ (N + 1) after luminance adjustment. The frame correlation determination unit 73 determines the determination result r ″ as an OR circuit 74. Output to. In this example, r ″ = 1 when it is determined that there is a correlation, and r ″ = 0 when it is determined that there is no correlation.

  A determination result r from the frame correlation determination unit 2 is sent to the OR circuit 74. When r = 1, the addition result r ′ of the OR circuit 74 is always 1. When r = 0, the addition result r ′ of the OR circuit 74 has the same value as the determination result r ″ of the frame correlation determination unit 73.

  The signal switching unit 75 is input with a value of 0 and the luminance difference Y ′ detected by the overall luminance difference detecting unit 71 as an input signal representing the luminance difference, and as a signal representing F (N + 1), F (N + 1) before luminance adjustment and F ″ (N + 1) after luminance adjustment are input.

  The signal switching unit 75 switches the input signal based on the correlation determination result r sent from the frame correlation determination unit 2. That is, when the frame correlation determining unit 2 determines that there is a correlation between F (N) and F (N + 1) (r = 1), the signal switching unit 75 represents a luminance difference with a value of 0. While outputting as the signal Y, F (N + 1) before luminance adjustment is output as the image F ′ (N + 1) of the reference frame. On the other hand, when the frame correlation determination unit 2 determines that there is no correlation between F (N) and F (N + 1) (r = 0), the signal switching unit 75 is detected by the overall luminance difference detection unit 71. The luminance difference Y ′ thus output is output as a signal Y representing the luminance difference, and F ″ (N + 1) after the luminance adjustment is output as an image F ′ (N + 1) of the reference frame.

  Returning to FIG. 10, the motion area generation unit 3 identifies a pixel having a change in the image between F (N) and F ′ (N + 1) output from the luminance correction unit 7. That is, the absolute value of the difference between F (N) and F ′ (N + 1) is obtained for each pixel, and when the absolute value of the difference is equal to or greater than a predetermined threshold, the pixel is changed to a pixel (motion) Pixel). The motion area generation unit 3 generates motion area information M ′ (N) including information indicating whether or not each pixel has a motion and sends the motion area information M ′ (N) to the motion area / motion vector correction unit 5.

  The motion vector generation unit 4 calculates a motion vector between F (N) and F ′ (N + 1) output from the luminance correction unit 7. Specifically, the motion vector generation unit 4 obtains a motion vector for each of a plurality of blocks set by dividing a video area using an existing method such as block matching. The motion vector generation unit 4 generates a motion vector V ′ (N) obtained for each block and sends it to the motion area / motion vector correction unit 5.

  The configuration of the motion area / motion vector correction unit 5 is the same as the configuration described in the first embodiment (FIG. 4). However, in the first embodiment, the determination result r of the frame correlation determination unit 2 is given as the switching control signal to the signal switching circuit 51 in FIG. A signal r ′ generated by the luminance correction unit 7 is given as a switching control signal.

  The interpolated frame synthesis unit 6 performs the same process (the process shown in FIG. 8) as in the first embodiment to synthesize an interpolated frame. However, the reference frame F ′ (N + 1) output from the luminance correction unit 7 is used instead of the reference frame F (N + 1) of the first embodiment, and the luminance correction unit 7 replaces the determination result r of the first embodiment. The generated signal r ′ is used.

  The interpolated frame (denoted by Fc ′ (N) in the second embodiment) generated by the interpolated frame synthesizing unit 6 is sent to the overall luminance adjusting unit 8. A luminance difference Y is given to the overall luminance adjustment unit 8 from the luminance correction unit 7. The overall luminance adjustment unit 8 reduces the luminance of each pixel of the interpolation frame Fc ′ (N) generated by the interpolation frame synthesis unit 6 by ½ of the luminance difference Y, so that the final interpolation frame Fc (N )

If the average luminance of the current frame F (N) is Y _N and the average luminance of the reference frame F (N + 1) is Y _{N + 1} , the luminance difference Y is Y = Y _N −Y _{N + 1} . At the time of fade-in (fade-out), the inter-frame Fc ′ (N) is generated after the luminance is adjusted so that the overall luminance of the reference frame F (N + 1) becomes equal to the overall luminance Y _{N of the} current frame F (N). Therefore, the average luminance of the interpolation frame Fc ′ (N) is Y _N.

Therefore, in order to set the average luminance of the interpolation frame Fc ′ (N) to an intermediate luminance {(Y _N + Y _{N + 1} ) / 2} between the current frame F (N) and the reference frame F (N + 1), the interpolation frame The brightness of each pixel of Fc ′ (N) is reduced by ½ of the brightness difference Y (= Y _N −Y _{N + 1} ).

That is, assuming that the average luminance of the interpolation frame Fc ′ (N) is Y _N , the average luminance Yc of the interpolation frame Fc (N) after luminance correction is represented by {(Y _N + Y _{N +1} ) / 2}.

_{Yc = Y N -Y / 2 =} Y N - (Y N -Y N + 1) / 2 = (Y N + Y N + 1) / 2
(4)

It is a schematic diagram which shows the present flame | frame (N), the reference frame F (N + 1), and the interpolated frame Fc (N) produced | generated from those frames. FIG. 3 is a schematic diagram for explaining a basic concept of the first embodiment. It is a block diagram which shows the electrical structure of a frame rate conversion apparatus. 3 is a block diagram showing an electrical configuration of a motion area / motion vector correction unit 5. FIG. 11 is a flowchart showing a prediction processing procedure executed by a motion area / motion vector prediction unit 53. It is a schematic diagram showing m × n blocks set in the video area. It is a schematic diagram which shows that the motion area information corresponding to each block in a certain frame number is represented by M (b, frame number). 5 is a flowchart showing an interpolation frame synthesis processing procedure executed by an interpolation frame synthesis unit 6. FIG. 6 is a schematic diagram for explaining a basic concept of Example 2. It is a block diagram which shows the electric constitution of a frame rate conversion apparatus. 3 is a block diagram illustrating an electrical configuration of a luminance correction unit 7. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Delay circuit 2 Frame correlation determination part 3 Motion area production | generation part 4 Motion vector production | generation part 5 Motion area and motion vector correction | amendment part 6 Interpolation frame synthetic | combination part 7 Brightness correction part 8 Overall brightness adjustment part

Claims (5)

Correlation determination means for determining whether there is a correlation between adjacent front and rear frames,
When it is determined that there is a correlation, interpolation information for generating an interpolation frame to be inserted between the both frames is generated from the both frames, and an interpolation frame is generated based on the obtained interpolation information The first interpolation frame generation means, and when it is determined that there is no correlation, the interpolation information for generating the current interpolation frame is predicted based on the interpolation information used for the previous interpolation frame generation. Second interpolation frame generation means for generating an interpolation frame based on the predicted interpolation information,
A frame rate conversion apparatus comprising: 2. The frame rate conversion apparatus according to claim 1, wherein the second interpolation frame generation means generates an interpolation frame using only the previous frame of the two frames. First correlation determination means for determining whether or not there is a correlation between adjacent front and rear frames;
When the first correlation determining means determines that there is no correlation, the second correlation determining means for determining whether or not there is a correlation between both frames after aligning the luminance levels of the two frames, and If the correlation determination unit 2 determines that there is a correlation, interpolation information for generating an interpolation frame to be inserted between the frames is generated from both frames after the luminance levels are aligned. Means for generating an interpolation frame based on the obtained interpolation information;
A frame rate conversion apparatus comprising: When the first correlation determination unit determines that there is a correlation, the interpolation information for generating an interpolation frame to be inserted between the both frames is generated from the both frames, and the obtained interpolation information is obtained. If the second correlation determination unit determines that there is no correlation based on the interpolation information used for generating the previous interpolation frame, the current interpolation is generated. Means for predicting interpolation information for generating a frame and generating an interpolation frame based on the predicted interpolation information;
The frame rate conversion apparatus according to claim 3, comprising: An image display device comprising the frame rate conversion device according to claim 1.

Images