JP5147655B2 - Video signal processing device and video display device - Google Patents

Description

  The present invention relates to a technique for creating an interpolation frame from a video frame of an input video signal and performing frame rate conversion by combining this with a frame of an input video signal.

  In a video display device such as a television receiver, so-called frame rate conversion in which the frame rate (for example, 60 Hz) of an input video signal is upconverted (for example, 120 Hz) may be used to improve the display performance of moving images. is there. Such frame rate conversion is performed by creating an interpolation frame corresponding to the motion of the video from a plurality of video frames of the input video signal, and combining this with the frame of the input video signal. As a result, the number of frames per unit time of the video is increased and the resolution is increased in the time direction, thereby improving blurring and rattling in moving image display and improving moving image display performance.

  For example, as described in Patent Document 1, the interpolation frame detects a motion vector between frames using the current frame and the previous frame, and the frame at the position specified by the motion vector and the previous frame are detected. It is created by filtering video data. An interpolation frame created using such motion vectors may be referred to as a motion compensation type interpolation frame below.

JP 2006-165602 A

  As described in Patent Document 1, the motion compensated interpolation frame is created by filtering processing (for example, linear interpolation processing, weighted average processing) from the preceding and succeeding video frames. Drops to a blurred image.

  When frame rate conversion is used to improve moving image blur, if the interpolated frame becomes a blurred image as described above, the effect of improving the moving image blur may be reduced. Further, when converting a movie content having a frame rate of 24 Hz in 2-3 pull down format to, for example, 60 Hz in order to eliminate rattling, as shown in FIG. 17, for example, two key frames (video frames) 171, Four interpolation frames 172 to 176 are inserted between 176. In this case, for example, the high-frequency component of the moving tree portion 177 is reduced by the filtering process in the interpolation frames 172 to 176. As a result, an image quality difference from the key frame occurs, and surface flicker may be recognized. This is a phenomenon that appears more prominently as the high-frequency component contained in the video increases.

  The present invention has been made in view of the above-described problems, and provides a technique capable of improving the image quality by suppressing blurring or flicker of video that has undergone frame rate conversion using an interpolation frame. It is.

  The present invention includes an interpolation frame enhancer that performs enhancement processing for increasing high-frequency components on an interpolation frame created by filtering processing from the two video frames using motion vectors detected from the two frames. It is characterized by that. As a result, it is possible to correct the high-frequency component of the interpolation frame that has been reduced by the filtering process.

  It is preferable to control the enhancement amount (amplification amount of the high frequency component) in the enhancement processing according to the characteristics of the video. The features of this video are, for example, the amount of motion of the video, the direction of the motion vector, the number of pixels or blocks in which the motion vector is detected, the number of pixels having edge components greater than or equal to a predetermined value, Including at least one.

  According to the present invention, it is possible to improve the image quality by suppressing the blurring and flicker of the frame rate converted video.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that elements having the same function and operation are denoted by the same reference numerals, and redundant description is omitted.

  FIG. 1 is a block diagram showing a configuration example of a video processing apparatus according to the first embodiment of the present invention.

  In FIG. 1, an input video signal 1 is, for example, a television signal, a video signal reproduced from an external video reproduction device, and is input in a digital format. When the input video signal 1 is an analog signal, it is converted into a digital signal by an A / D converter (not shown). This input video signal 1 is supplied to the resolution converter 2. The resolution conversion unit 2 performs enlargement / reduction processing in the horizontal and / or vertical direction so that the image format of the input video signal 1 matches the definition of the display unit 9.

  An input video signal enlarged or reduced by the resolution converter 2 is input to a frame rate converter (FRC) 3 to convert a frame rate of the input video signal (hereinafter referred to as FRC). Called processing). Such FRC processing detects a motion vector from the current video frame (hereinafter referred to as the current frame) in the input video signal and the signal of the previous frame in time (for example, one frame before) from the current frame. Then, an interpolation frame is generated using the motion vector, the current frame, and the previous frame. The previous frame can be generated by delaying a signal for one frame by one frame period by an image memory connected to the FRC 3. Then, the FRC 3 increases the frame rate by combining the generated interpolation frame with the video frame sequence of the input video signal. For example, if the input video signal has a frame rate of 60 Hz, it is converted by FRC3 so as to increase to 120 Hz, which is twice as much. Here, the process of combining the video frame sequence of the input video signal and the interpolation frame includes, for example, a first combination process of inserting one or a plurality of interpolation frames between two adjacent video frames in the input video signal, and the input video It is assumed to include a second combination process in which one or a plurality of video frames in the signal are periodically replaced (replaced) with interpolation frames.

  In the present embodiment, the FRC 3 is capable of executing a plurality of types of FRC processing. For example, (1) processing for converting a video signal with a frame rate of 60 Hz to double 120 Hz as described above, and (2) frame Processing to convert a video signal with a rate of 60 Hz to 180 Hz, which is three times higher, (3) Processing to convert a video signal (telecine signal) with a frame rate of 24 Hz, such as movie content, into 60 Hz, and (4) 2-3 with movie content There are four FRC processing modes: processing for converting a video signal with a frame rate of 60 Hz in a pull-down format to 60 Hz.

  The FRC processes (1) to (3) are performed by the first combination process. In the process (1), one interpolation frame is inserted between adjacent video frames, and in the process (2), two interpolation frames are inserted between adjacent video frames. In the process (3), one interpolation frame is inserted between certain adjacent video frames, and two interpolation frames are inserted between adjacent video frames. The process (4) is performed by the second combination process, and a part of the same video frame, for example, one of the same video frame that is continued twice and two of the same video frames that are continued three times. Performs processing to replace the interpolation frame. In the process (4), the frame rate is not increased, but the motion sensation caused by the continuous appearance of the same video frame in the 2-3 pull-down method is reduced.

  One of these FRC processes (1) to (4) is selected and set by the FRC conversion mode signal 7 given from the outside. The FRC conversion mode signal 7 may be generated by a user operation, for example, and the type of video content to be displayed (for example, movie, sports) is added to information (for example, digital television) added to the input video signal. It may be discriminated from the EPG information added to the broadcast signal) and automatically generated. For example, the FRC conversion mode signal 7 is generated such that (2) is selected when the displayed video content is sports, and (3) is selected when the video content is a movie. The type of the FRC process is not limited to the above, and may be two or five or more.

  In the present embodiment, the interpolation frame generated by the FRC 3 is subjected to enhancer processing for increasing or strengthening the high-frequency component of the interpolation frame. Details of the interpolation frame generation and enhancer processing will be described later.

  The video signal from the FRC 3 is input to the timing controller 5 to generate an optimal timing signal (horizontal, vertical synchronization signal, and dot clock signal) for displaying the frame rate converted video signal and display it together with the video signal. Supply to part 6. The display unit 6 is configured using, for example, a liquid crystal panel or a plasma display panel, and displays a video based on the timing signal from the timing controller 5 and the video signal subjected to frame rate conversion.

  FIG. 2 shows an example of the FRC 3 according to the present embodiment. In FIG. 2, one frame of the output signal from the resolution conversion unit 2 (here, this frame is assumed to be the current frame F1) is input to the motion vector detection unit 31 and the memory I / F 35 that accesses the image memory 4. The The memory I / F 35 outputs and stores the current frame F1 to the image memory 4 connected to the memory I / F 35, while the video frame data stored in the image memory 4 in the previous frame F1 is stored in the previous frame F2. Read as. The read previous frame F2 is output to the motion vector detection unit 31 and the interpolation frame generation unit 33.

  The motion vector detection unit 31 uses the input data of the current frame F1 and the previous frame F2 data to specify the FRC processing mode (for example, the above-described (1) to (4) described above by the FRC conversion mode signal 7. ) Of the motion vector MV of the video (object) corresponding to the FRC process). As a motion vector detection method, a conventionally known block matching method, gradient method, phase-only correlation method, or the like is used. Here, it is assumed that the block matching method is used, and an example of detection of the motion vector will be described with reference to FIG.

  FIG. 3 shows how the motion vector is detected when the interpolation frame FI1 is inserted between the current frame F1 and the previous frame F2 temporally continuous thereto. The motion vector MV indicates to which block on the current frame F1 a certain target block 311 in the previous frame F2 has moved. These blocks are assumed to have N × N pixels. N is an integer, for example, a number such as 2, 4, 8, 16 is selected.

  First, a search range 314 having a predetermined number of blocks is set on the current frame F1, with reference to a block 315 that is spatially the same position as the target block 311 in the previous frame F2 (ie, a point-symmetric position). Here, the number of blocks in the search range 314 is assumed to be a total of 55 blocks of 11 in the horizontal direction and 5 in the vertical direction, and the coordinates of the position of the reference block 315 in the search range 314 are (0, 0). Next, a difference value between each block in the set search range 314 and the target block 311 in the previous frame F2 is calculated. Subsequently, as a result of this calculation, the block in the search range 314 having the smallest difference is determined as the movement destination block in the current frame F1 of the target block 311 in the previous frame F2. A straight line connecting the destination block and the target block 311 in the previous frame F2 is a motion vector indicated by a broken line in FIG. In FIG. 3, for example, if the block 313 is the block with the smallest difference, the target block 311 of the previous frame F2 has moved to the block 313 on the current frame F1 according to the direction of the motion vector MV indicated by the broken line. . The coordinates of the block 313 at this time are (5, 2), and data of this coordinate is acquired as information regarding the motion vector MV for the target block 311.

  The processing from the setting of the search range 314 to the acquisition of information on the motion vector MV is performed for all the blocks of the previous frame F2, and the motion vector MV is detected and set for each of all the blocks. In this embodiment, the motion vector is detected in units of blocks composed of a plurality of pixels. However, the motion vector may be detected for each pixel by the same method as the above processing.

  The motion vector MV detected in block units in this way is output to the interpolation frame generation unit 33 and the video feature extraction unit 32 for extracting video features. The interpolation frame generation unit 33 performs an interpolation frame generation process corresponding to the FRC process mode specified by the FRC conversion mode signal 7 using the motion vector MV, the current frame F1, and the previous frame F2. An example of this interpolation frame processing will be described again with reference to FIG.

  In FIG. 3, for example, when the FRC process (1) is performed, one interpolation frame FI1 is inserted between the current frame F1 and the previous frame F2. In this case, it is estimated that the target block 311 of the previous frame F2 moves to the block 313 of the current frame F1 through a block 312 (hereinafter referred to as an interpolation block) in the interpolation frame FI1. Therefore, the pixel data of the interpolation block 312 is a process for calculating an intermediate value or an average value between the pixel data of the target block 311 of the previous frame F2 and the pixel data of the block 313 indicated by the motion vector MV, for example, a linear interpolation process. Or a filtering process such as a weighted average process. By performing this calculation (filtering process) for all the blocks of the previous frame F2, pixel data of all the interpolation blocks of the interpolation frame FI1 is obtained, and a motion adaptive interpolation frame is created.

  When the FRC process (2) is performed, two interpolation frames FI1 are inserted between the current frame F1 and the previous frame F2. In this case, linear interpolation processing is performed according to the temporal distance between each interpolation frame FI1, the current frame F1, and the previous frame F2. For example, if the ratio of the temporal distance between one of the two interpolated frames FI1 and the current frame F1 to the temporal distance between the interpolated frame FI1 and the previous frame F2 is 2: 1, the current frame F1. The block 313 is multiplied by the coefficient 1/3, the block 311 of the previous frame F2 is multiplied by the coefficient 2/3, and the results are added to each other.

  The number of inserted interpolation frames, the insertion position, and the coefficient setting are controlled according to the FRC conversion mode signal 7 described above. The above-described detection of the motion vector MV and generation of the interpolation frame FI1 are executed by software processing, but may be executed by hardware processing.

  As described above, since the interpolation frame FI1 is created by the filtering process, the high-frequency component is reduced and the image becomes blurred. Therefore, in this embodiment, the motion compensation type interpolation frame created by the filtering process as described above is supplied to the adaptive enhancer processing unit 34 which is an interpolation frame enhancer unit, where the high frequency component of the interpolation frame is increased or strengthened. Therefore, an enhancer process is performed. Details of the adaptive enhancer processing unit 34 will be described later.

  On the other hand, the video feature extraction unit 32 extracts video features using the motion vector MV detected as described above. In this embodiment, it is assumed that a feature relating to the motion of the video is extracted as the feature of the video. Specifically, when a motion vector MV (a motion vector having coordinates other than (0, 0)) detected in the previous frame F2 exists in a predetermined number of blocks or more, there is motion in the video. If it is less than the predetermined value, it is determined that there is no motion, and feature information relating to the motion of the video is detected. Here, the predetermined number of blocks, which is a criterion for determining the presence or absence of motion, is about 80 to 90% of the total number of blocks in the previous frame F2, for example, but may be a value other than this. The control signal SC for controlling the adaptive enhancer processing unit 34 having a value of “1” when it is determined that there is motion and “0” when it is determined that there is no motion is the adaptive enhancer process. To the unit 34.

  The adaptive enhancer processing unit 34 performs enhancer processing for increasing or strengthening the high frequency component of the input interpolation frame FI1 based on the control signal SC. A specific example of the adaptive enhancer processing unit 34 will be described with reference to FIG.

  In FIG. 4, the interpolation frame FI1 output from the interpolation frame generation unit 33 is supplied to the vertical high-pass filter 341, the horizontal high-pass filter 342, and the delay unit 343. The vertical high-pass filter 341 extracts a high-frequency component in the vertical direction (edge portion such as an outline) from the interpolation frame FI1, and the horizontal high-pass filter 342 similarly extracts a high-frequency component in the horizontal direction from the interpolation frame FI1 to perform enhancement processing. Ingredients are removed. Here, the high-frequency component is extracted using the two high-pass filters of the vertical high-pass filter 341 and the horizontal high-pass filter 342, but either one may be used.

  The vertical and horizontal high-frequency components extracted by the high-pass filters 341 and 342 are input to the adder 344 and added to each other. The high frequency component added by the adder 344 is supplied to a nonlinear processing unit 345 which is an enhancer processing unit.

  For example, as shown in FIG. 5, the non-linear processing unit 345 amplifies the coring circuit 51 that performs noise removal processing for removing noise from the high-frequency component from the adder 344, and the output signal from the coring circuit 51. And an amplitude limiter circuit 53 for applying a predetermined restriction to the high-frequency component amplified by the gain adjustment circuit 52.

  Depending on the processing or operation of each circuit, the non-linear processing unit 345 has non-linear input / output characteristics. Processing of each circuit in the nonlinear processing unit 345 will be described with reference to FIG. 6 showing an example of the input / output characteristics.

  In FIG. 6, the horizontal axis indicates the level of the input signal of the nonlinear processing unit 345, and the vertical axis indicates the level of the output signal of the nonlinear processing unit 345. A range indicated as “CORE” on the horizontal axis in FIG. 6 indicates a range of levels to be subjected to coring processing for setting the level of a signal in the range to 0 in the coring circuit 51. . A signal in this level range is regarded as a noise component having a minute amplitude, and noise is removed by setting the level of the signal to 0. This prevents the noise component from being amplified by the subsequent gain adjustment circuit 52. Hereinafter, this range is referred to as “coring amount”.

  The gain adjustment circuit 52 amplifies the high frequency component from which noise has been removed in accordance with the predetermined gain amount GAIN shown in FIG. However, the amplitude limiter circuit 53 limits the output signal level so that signals whose input levels are in the range of LIM1 to LIM2 and -LIM1 to -LIM2 are not amplified beyond a predetermined limit level, and excessive enhancer processing is performed. Is suppressed. As a result, excessive amplification of the high-frequency component with a large amplitude near the maximum dynamic range by the gain adjustment circuit 52 is suppressed, and problems such as overexposure and blackout due to such excessive amplification are prevented.

  The various processes in the vertical high-pass filter 341, the horizontal high-pass filter 342, the coring circuit 51 in the nonlinear processing unit 345, the gain adjustment circuit 52, and the amplitude limiter circuit 53 are collectively referred to as “enhancer processing”. Hereinafter, the component will be referred to as an “enhancement signal”.

  On the other hand, the delay unit 343 is configured by a memory, for example, and delays the interpolation frame FI1 by a delay amount substantially equal to the time required for various processes by the vertical high-pass filter 341, the horizontal high-pass filter 342, and the nonlinear processing unit 345. The interpolated frame FI1 delayed by the delay unit 343 and the enhancement signal from the nonlinear processing unit 345 are added cells by the adder 346. As a result, an interpolation frame in which the high frequency component is increased or emphasized is generated.

  One of the interpolation frame FI2 and the interpolation frame FI1 delayed by the delay unit 343 is selected by the selector 347 according to the control signal SC and output. When the control signal SC is “1”, that is, when there is a video motion, the interpolation frame FI2 is selected. When the control signal SC is “0”, that is, when there is no video motion, the delayed interpolation frame FI1 is selected. Selected.

  Returning to FIG. 2 again, the interpolated frame FI1 / FI2 output from the selector 347 of the adaptive enhancer 34 is supplied to the memory I / F 35, where it is combined with the previous frame F2 stored in the image memory 4. The For example, in the FRC process (1) described above, the previous frame F2 stored in the image memory 4 and the interpolated frame output from the adaptive enhancer 34 are alternately output at a cycle that is half the vertical cycle of the input video signal. To do. In this way, the interpolated frame and the video frame that have been enhanced according to the motion of the video are combined and the frame rate converted signal is output. The frame rate converted signal is displayed on the display unit 6 via the timing controller 5 as described above.

  In this way, in this embodiment, when there is motion in the input video, the enhancer processing unit 345 increases or emphasizes the high-frequency component of the interpolation frame, so that the interpolation frame that becomes particularly large when there is motion. Blur and flicker are reduced. If there is no motion, the filtering process is not performed (in this case, the interpolated block 312 data is a copy of the target block of the previous frame F1), or even if it is performed, the influence on the high frequency component is small. The interpolated frame is output as it is without processing.

  For this reason, according to the present embodiment, in a video with a large motion, even when the high frequency component is reduced by the filtering process for generating the interpolation frame, the frame rate converted signal is increased by increasing the interpolation frame high frequency component. Blur and flicker can be reduced, and a high-quality frame rate converted video can be obtained.

  Next, a second embodiment of the present invention will be described with reference to FIGS.

  FIG. 7 shows a second embodiment of the adaptive enhancer processing unit according to the present invention, which is denoted by reference numeral 34-2 in the same drawing. In FIG. 7, the same components as those in the first embodiment described with reference to FIG.

  In the first embodiment, the presence / absence of use of the interpolated frame subjected to the enhancement process is controlled by the motion presence / absence signal from the video feature extraction unit 32. On the other hand, in the second embodiment, the video feature extraction unit 32 detects not only the presence / absence of motion but also the magnitude of motion as a video feature, and an adaptive enhancer processing unit based on a control signal SC indicating the magnitude of the motion amount. 34-2, the filter characteristics of the vertical and horizontal high-pass filters, the enhancement amount of the gain adjustment circuit 52 (high frequency component amplification factor) in the nonlinear processing unit, the coring amount of the noise removal circuit 51, and the limiter in the amplitude limiter circuit 53. The range is variably controlled. As described above, in the second embodiment, the configurations of the video feature extraction unit 32, the noise removal circuit 51, the gain adjustment circuit 52, and the amplitude limiter circuit 53 are different, but are not illustrated here.

  Here, the motion amount of the video can be obtained by creating a histogram of the detected motion vector MV (hereinafter referred to as a vector histogram) in the video feature extraction unit 32, for example. This vector histogram shows the frequency distribution of motion vectors in one frame of video, and an example thereof is shown in FIG.

  In the vector histogram of FIG. 8, the X axis corresponds to the horizontal direction of the search range 314 shown in FIG. 3, for example, and the Y axis corresponds to the vertical direction of the search range 314. The vector histogram uses the coordinate data of the motion vector MV detected for each block of the previous frame F2 described above, and indicates the number of blocks in which the motion vector having each coordinate data of the search range 314 is detected in the previous frame F2. It is obtained by counting (integrating) over the whole. Thereby, the appearance frequency in one frame of the motion vector corresponding to each coordinate can be calculated. From this vector histogram, the coordinates of a motion vector having a count number (appearance frequency) greater than or equal to a predetermined count value are obtained, and the magnitude of the motion amount is determined from this. For example, if the count value of a motion vector having an absolute value of x of coordinates of 4 or more is a predetermined count value (for example, 40000) or more, the motion amount of the frame is determined to be “large”. For example, if the count value of a motion vector having an absolute value of x of coordinates 2 to 3 or an absolute value of y of 2 is equal to or greater than a predetermined count value, the motion amount of the frame is determined to be “medium”. For example, if the count value of the motion vector whose absolute value of x of the coordinate is 1 and whose absolute value of y is 0 to 1 is equal to or greater than a predetermined count value, the motion amount of the frame is determined to be “small”. If a motion vector of 0 for both x and y is equal to or greater than a predetermined count value, the frame is determined to be “no motion”.

  The x and y values of the vector histogram and the search range 314 are ± 5 and ± 2 for simplification of explanation and illustration, respectively. However, when actually applied to the apparatus, x is ± 30. ~ 100, y is preferably about ± 10-50. Further, it is needless to say that the values of x and y for determining the magnitude of the motion amount are not limited to the above example, and can be appropriately changed according to application to an actual apparatus.

  In the example of the vector histogram shown in FIG. 8, the count numbers of the coordinates (−3, 0) and (−2, −1) exceed the predetermined count value (40000). Of these, for example, the amount of motion of the image of the frame is determined based on the value of the coordinate having the larger count number. Here, since the count number of the coordinates (−3, 0) is larger, the motion amount of the video of the frame is determined to be “medium”, and information on the determined motion amount is extracted as the feature of the video. .

  Information indicating the magnitude of the amount of motion of the video extracted by the video feature extraction unit 32 in this way is used as the control signal SC as the vertical high-pass filter 371 and the horizontal high-pass of the adaptive enhancer processing unit 34-2 shown in FIG. The data is output to the filter 372 and the nonlinear processing unit 375.

  Hereinafter, with respect to the adaptive enhancer processing unit 34-2 according to the second embodiment, the operation of parts different from the first embodiment described above, and FIG. 9 showing an example of the filter characteristics according to the present embodiment, and the enhancement processing An example of the switching operation will be described with reference to FIG.

  In FIG. 9, the vertical high-pass filter 371 and the horizontal high-pass filter 372 are configured such that their filter characteristics are varied by the control signal SC from the video feature extraction unit 34, respectively. For example, each of the vertical high-pass filter 371 and the horizontal high-pass filter 372 has two types of filter characteristics, a first filter characteristic shown in FIG. 9A and a second filter characteristic shown in FIG. Switching control is performed by a control signal SC from the feature extraction unit 34. Here, in the first filter characteristic, the slope of the gain characteristic curve with respect to a signal near the cutoff frequency (cutoff frequency) is gentler than that of the second filter characteristic.

  Here, in motion vector detection, there is a general tendency that a motion vector is more likely to be erroneously detected when the motion amount is large (the motion is fast) than when the motion amount is small (the motion is slow). is there. Therefore, when the amount of motion is small (for example, the speed required to cross the screen is 20 seconds or more: 20 [seconds / screen]), the false detection rate of the motion vector is low, and the above-described filtering in the interpolation frame generation unit 33 is performed. There is little deterioration of the high frequency component by processing. In this case, with respect to the characteristics of the horizontal and vertical high-pass filters 371 and 372, the characteristic curve is gentle (in other words, the cutoff characteristic is slow) in the vicinity of the cutoff frequency as shown in FIG. N) A high frequency portion is extracted using a filter having the first filter characteristic.

  At this time, the amplification degree (enhancement amount) in the gain adjustment circuit 51 of the nonlinear processing unit 375 is controlled to 0.5 as shown in FIG. 10 by the control signal SC. The coring circuit 52 and the amplitude limiter circuit 53 of the nonlinear processor 375 are also controlled by the control signal SC, and as shown in FIG. 10, the coring amount is “large” and the limiter range is also “wide”.

  In the present example, the above-described control is performed for a slow moving image for the following reason. In other words, as described above, a video with a small amount of motion has little deterioration of high-frequency components, and noise and overexposure are easily noticeable. This is because priority is given to suppressing the occurrence of noise and overexposure to the flat portion before and after the edge portion.

  When the amount of motion is large (for example, 9 [seconds / screen] or less), erroneous detection of the motion vector is high. However, since the original image at the time of shooting is already blurred at this speed, filtering in the interpolation frame generation unit 33 is performed. High-frequency degradation due to processing is not so great (not noticeable). Therefore, in this case, both the filter characteristics of the horizontal and vertical high-pass filters 371 and 372 and the enhancement amount in the gain adjustment circuit 51 may be the same control as in the case where the motion amount is small. In addition, since a fast moving image is not easily noticeable with noise and overexposure, as shown in FIG. 10, the coring amount is set to “small” and the limiter range is set to “narrow” by the control signal SC.

  On the other hand, when the amount of motion is medium (for example, 10 [seconds / screen] or more), the motion vector false detection rate is reasonably high, and the degradation of the high-frequency component due to the filtering processing in the interpolation frame generation unit 33 is the largest. Accordingly, the filter characteristics of the horizontal and vertical high-pass filters 371 and 372 have a steep characteristic curve in the vicinity of the cutoff frequency as shown in FIG. N) Use a filter to accurately extract high-frequency components.

  At this time, the amplification degree (enhancement amount) in the gain adjustment circuit 51 of the nonlinear processing unit 375 is controlled to 2.0 as shown in FIG. 10 by the control signal SC, and the emphasis of the high frequency component is increased. The coring circuit 52 and the amplitude limiter circuit 53 of the nonlinear processing unit 375 are also controlled by the control signal SC. As shown in FIG. 10, the coring amount is “medium” and the limiter range is also “medium”.

  Further, in the case where there is no motion amount on the screen, that is, a still image, it is normal that 0 is detected as the motion vector, and high-frequency deterioration due to filtering processing by the interpolation frame generation unit 33 does not basically occur. Therefore, in this case, the control signal SC is controlled so as not to perform high-frequency component extraction processing by the vertical and horizontal high-pass filters 371 and 372, and the enhancement amount in the gain adjustment circuit 51 is controlled to 0 and the enhancement processing is not performed. . Further, the coring amount of the coring circuit 52 and the limiter range of the amplitude limiter circuit 53 are the same as when the amount of motion is small.

  The enhance signal subjected to the enhancer processing as described above is added to the non-enhancer processed signal delayed by the delay unit 343 in the adder 236. Here, the difference from the second embodiment is that the selector 347 is deleted and the signal added by the adder 236 is directly output. In the third embodiment, when there is no movement in the control signal SC, the vertical and horizontal high-pass filters 371 and 372 and the nonlinear processing unit 375 are controlled not to perform the enhancer process, so the selector 347 is unnecessary.

  As described above, according to the present embodiment, not only the presence / absence of the motion of the video, but also the motion of the video can be adaptively performed according to the amount of motion when there is motion. A high-frequency component emphasis effect suitable for the amount can be obtained.

  In the present embodiment, two types of characteristics of the high-pass filter are used. However, the present invention is not limited to this, and three or more types of characteristics may be prepared and switched. As for the high-pass filter, a differential operator such as Laplacian may be used so that the applied intensity can be varied according to the characteristics of the motion.

  In the above example, the same enhancer amount is used when the motion amount is small and when the motion amount is large. However, this may be different.

  In the above example, the limiter range of the amplitude limiter circuit 53 is controlled by the control signal SC. However, the limit level of the output signal described above may be controlled, or both the limiter range and the limit level may be controlled. It may be. When controlling the restriction level, for example, the restriction level is reduced when there is no or slow motion of the video, and the restriction level is increased when the movement is fast. When the movement is medium, the limit level is set to an intermediate value between when there is no movement or when it is slow and when it is fast. In the above example, the coring amount, the amplification degree, the limiter range, and / or the limit level are controlled. However, at least the amplification degree may be controlled, and the others may not be controlled. Only one of them may be controlled.

  Next, a third embodiment according to the present invention will be described with reference to FIG. The circuit configuration and the like are the same as in the second embodiment shown in FIG.

  FIG. 11 shows an example of enhancement processing control according to the motion of the video according to the third embodiment. In the second embodiment, the enhancer process is controlled according to the magnitude of the motion amount of the video. In the third embodiment, the type of motion vector in one frame (in addition to the magnitude of the motion amount) is used as the video feature. Direction) and the number of types are detected and the enhancement processing is controlled. In FIG. 11, the contents of the enhancement processing control are shown only for the filter characteristics and the enhancement amount.

  The video feature extraction unit 32 uses the vector histogram shown in FIG. 8 to exceed a count value (second predetermined count value) larger than the count value (40000) described in the first example of the second embodiment, for example. It is determined whether or not a motion vector MV having an appearance frequency exists for a certain period (for example, several tens of frames). Here, the second predetermined count value is set to, for example, about 80 to 90% of the total number of blocks in one frame. When the motion vector MV exceeding the second predetermined count value exists for several tens of frames, the frame is displayed in a certain direction (the coordinates of the motion vector MV having the appearance frequency exceeding the second predetermined count value). It is determined that the image is a so-called pan video that scrolls in the direction). Then, the video feature extraction unit 32 outputs a control signal SC indicating “whole screen scroll” as a video feature to the vertical and horizontal high-pass filters 371 and 372 and the gain adjustment circuit 51 of the adaptive enhancer unit 34, and the vertical and horizontal high-pass filter Control is performed so that the characteristics of 371 and 372 are the characteristics of FIG. 9B with a large number of taps, and the enhancer amount (amplification degree) of the gain adjustment circuit 51 is 2.0. As for the coring amount and the limiter range, the coordinates of the motion vector MV having the appearance frequency exceeding the second predetermined count value are determined and the magnitude of the motion amount is detected. Controlled by value.

  Further, a third count value (for example, 10 to 30% of the total number of blocks in one frame) smaller than the second predetermined count value is set, and only one motion vector MV exceeding this is set for a certain period (for example, several tens of times). If there is a frame period), it is determined that the frame is a video in which a part of the screen is scrolled in a certain direction (the direction of the coordinates of the motion vector MV having the appearance frequency exceeding the third predetermined count value). Then, the video feature extraction unit 32 outputs a control signal SC indicating “screen partial scroll” as a video feature to the vertical and horizontal high-pass filters 371 and 372 and the gain adjustment circuit 51 of the adaptive enhancer 34, and the vertical and horizontal high-pass. The characteristics of the filters 371 and 372 are controlled so as to have the characteristics shown in FIG. 9B with a large number of taps, and the enhancer amount (amplification degree) of the gain adjustment circuit 51 is set to 1.0. The control of the coring amount and the limiter range is the same as described above.

  When there are a plurality of motion vectors MV exceeding the third count value, it is determined that the frame has a plurality of motions on the screen. Then, the video feature extraction unit 32 outputs a control signal SC indicating “multiple motions” as video features to the vertical and horizontal high-pass filters 371 and 372 and the gain adjustment circuit 51 of the adaptive enhancer 34, and the vertical and horizontal high-pass filter 371. , 372 are controlled to have the characteristics shown in FIG. 9A with a small number of taps, and the enhancer amount (amplification degree) of the gain adjustment circuit 51 is set to 0.5. The control of the coring amount and the limiter range is the same as described above.

  Also, when the amount of motion of the video is large, for example, in the vector histogram, when the frequency of appearance of a motion vector having an absolute value of x of coordinate 5 exceeds a predetermined count value, the frame is determined as a video moving at high speed. . The video feature extraction unit 32 outputs a control signal SC indicating “high-speed motion” as a video feature to the vertical and horizontal high-pass filters 371 and 372 and the gain adjustment circuit 51 of the adaptive enhancer 34, and the motion amount “large” in FIG. In the same manner as the control in FIG. 9, the characteristics of the vertical and horizontal high-pass filters 371 and 372 are the characteristics shown in FIG. 9A with a small number of taps, and the enhancer amount (amplification degree) of the gain adjustment circuit 51 is 0.5. Control. The control of the coring amount and the limiter range is the same as the control of the movement amount “large” in FIG.

  As described above, according to the present embodiment, not only the amount of motion of the video but also when there is motion, the enhancer processing can be adaptively performed according to the type (direction) and the number of the types. A high-frequency component enhancement effect suitable for the direction of motion of the video can be obtained.

  Also in this embodiment, the high-pass filter has two types of characteristics, but the present invention is not limited to this, and three or more types of characteristics may be prepared and switched.

  Next, a fourth embodiment according to the present invention will be described with reference to FIGS.

  FIG. 12 is a block diagram showing an example of the FRC 3 which is the fourth embodiment of the present invention, and is denoted by reference numeral 3-1. In FIG. 12, the same components as those shown in FIG. 2 are denoted by the same reference numerals and description thereof is omitted.

  In the first to third embodiments, the video feature extraction unit 32 extracts the feature related to the motion of the video as the feature of the video, but this embodiment relates to the edge (contour) of the video frame in addition to the feature related to the motion of the video. A feature is also extracted, whereby the adaptive enhancer unit 34 is controlled.

  Hereinafter, the operation of the parts different from the first to third embodiments will be described. In this embodiment, a high-pass filter 121 and a counter 122 are newly added to the block diagram of the first embodiment shown in FIG.

  The high-pass filter 121 is configured to extract an edge component of the frame, that is, a high-frequency component, from the previous frame F2, and in the same way as the nonlinear processing unit 345 in the second to third embodiments, the edge in the vertical and horizontal directions. Two high-pass filters for extracting the components are included. Further, the filter characteristic operates so as to count pixels corresponding to the edge component extracted by the high-pass filter 121 to which one of FIGS. 10A and 10B is applied. By the operation of this counter, it is possible to detect how many edge components are included in the previous frame F2, that is, the features relating to the edges. For example, a grid-like image or an image including many characters has a high edge component, and the counter value of the counter 122 also increases. The counter value of the counter 122 is output to the video feature extraction unit 321.

  The video feature extraction unit 321 determines that the count value is equal to or higher than a predetermined threshold value, a high-edge video image that includes a large amount of edge components, and if the count value is less than the predetermined threshold value, the video feature extraction unit 321 determines that the video image is a low-edge video image. Are output to the adaptive enhancer 34 as a control signal SC together with information on the amount of motion extracted separately.

  The control content of the enhancer process corresponding to the control signal SC according to this embodiment is shown in FIG. 13, for example. That is, in the present embodiment, when the edge component is low, the enhancer processing is not performed as in the case where there is no motion of the video, and when the edge component is high, the enhancer processing according to the amount of motion is performed as in FIG. In this embodiment, the video with a low edge component has little deterioration (not so noticeable) of the high-frequency component (edge component) due to the filtering process at the time of interpolation frame creation. Therefore, when the edge component is low, the enhancer process is executed as described above. However, for example, a small enhancer process having an enhancer amount of about 0.1 to 0.3 may be performed.

  As described above, in this embodiment, the enhancer processing is controlled according to not only the feature related to the motion of the video but also the feature related to the edge component of the video. Appropriate enhancer processing can be performed.

  Next, a fifth embodiment according to the present invention will be described with reference to FIGS.

  FIG. 14 is a block diagram showing an example of the FRC 3 according to the fifth embodiment of the present invention, which is denoted by reference numeral 3-2. 14, the same components as those shown in FIG. 2 are denoted by the same reference numerals, and the description thereof is omitted.

  In the first to third embodiments, the feature relating to the motion of the video is extracted by the video feature extraction unit 32. However, in this embodiment, the feature relating to the frequency component of the video is also extracted in addition to the feature relating to the motion of the video. It is characterized in that the enhancer unit 34 is controlled.

  Hereinafter, the operation of the parts different from the first to third embodiments will be described. In this embodiment, a block processing unit 141 and an FFT (Fast Fourier Transform) calculation unit 142 for performing frequency analysis of video are newly added to the block diagram of the first example shown in FIG. It is.

  The blocking processing unit 141 performs N × N (N is an integer) blocking on the input previous frame F2 in the same manner as the motion vector detection unit 31, and the FFT operation unit 142 performs FFT on this block unit. Processing is performed, the frequency spectrum is calculated by obtaining the intensity for each frequency component, and the calculation result is output to the video feature extraction unit 322. The blocking process may be shared with the motion vector detection unit 31.

  The video feature extraction unit 322 extracts the intensity of a predetermined frequency component or higher (that is, a high frequency component) for each block from the frequency spectrum indicating the intensity of each frequency component calculated by the FFT calculation unit 142, and the video in the target block is extracted. It is determined whether or not the high frequency component is included at a predetermined threshold value or more. Hereinafter, a block including a high frequency component equal to or greater than a predetermined threshold is referred to as a “high frequency block”. Next, using the motion vector MV detected by the motion vector detection unit 31, it is determined where the high frequency block is located on the interpolation frame. The position determination of the high frequency block will be described with reference to FIG.

  Assume that the motion vector detection unit 31 detects a motion vector MV indicating that the block 151 in the previous frame F2 has moved to the block 153 in the current frame F1, as shown in FIG. Since this vector MV passes through the interpolation block 152 of the interpolation frame FI1, it is estimated that the block 151 of the previous frame F2 has moved to the interpolation block 152 of the interpolation frame FI1 according to the direction indicated by the detected motion vector MV. Therefore, if the block 151 of the previous frame F2 is a high frequency block, it can be determined that the high frequency block has moved to the position of the interpolation block 152 on the interpolation frame FI1.

  The video feature extraction unit 322 detects position information on the interpolation frame of the high frequency block by such processing. Based on the detected position information, the control signal SC is output to the adaptive enhancer processing unit 34 for the interpolation block 152 corresponding to the high frequency block of the previous frame F2, and the enhancer process is executed. On the other hand, the control signal SC for stopping the enhancer process is output to the adaptive enhancer processing unit 34 for the low frequency block whose high frequency component is equal to or less than the predetermined threshold. FIG. 16 shows the control mode of the enhancer processing according to this embodiment. As is apparent from FIG. 16, the high-frequency block is subjected to the same enhancer processing as in FIG. 10, that is, the enhancer processing according to the amount of motion of the video, and the low-frequency block is subjected to the filtering processing. Since there is little deterioration of the image, the enhancer processing is not performed as in the case of the image having no motion.

  As described above, according to the present embodiment, the enhancement process is controlled by detecting not only the motion amount information but also the frequency component of the image as the feature of the image. Therefore, an appropriate enhancer process according to the frequency spectrum of the image is performed. It can be performed. Further, in the present embodiment, since such enhancer processing is performed in units of blocks, it is possible to perform more precise and detailed enhancer control.

1 is a block diagram illustrating an example of a display device to which an embodiment of the present invention is applied. The block diagram which shows the example of 1 structure of the FRC part 3 which concerns on 1st Example of this invention. The figure explaining an example of the process which detects the motion vector by a block matching method, and produces | generates an interpolation frame. The figure which shows the structural example of the adaptive enhancer process part 34 which concerns on 1st Example of this invention. The figure which shows an example of the nonlinear process part 345 which concerns on 1st Example of this invention. The figure which shows an example of the input-output characteristic of the nonlinear processing part 345 which concerns on 1st Example of this invention. The figure which shows the example of 1 structure of the adaptive enhancer process part 34 concerning 2nd Example of this invention. The figure which shows an example of the vector histogram produced by the video feature extraction part 32. FIG. The figure which shows an example of the filter characteristic of a vertical and a horizontal high-pass filter. The figure which shows the control content of the enhancer process which concerns on 2nd Example of this invention. The figure which shows the control content of the enhancer process which concerns on 3rd Example of this invention. The block diagram which shows the example of 1 structure of the FRC part which concerns on 4th Example of this invention. The figure which shows the control content of the enhancer process which concerns on 4th Example of this invention. The block diagram which shows the example of 1 structure of the FRC part which concerns on 5th Example of this invention. The figure which shows the method for pinpointing the position on the interpolation frame of the high frequency block in a front frame. The figure which shows the control content of the enhancer process which concerns on 5th Example of this invention. The figure which shows the principle of the flicker generation | occurrence | production by a frame rate conversion process.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Input video signal, 2 ... Resolution conversion part, 3 ... Frame rate conversion process part, 4 ... Image memory, 5 ... Timing controller, 6 ... Display part, 7 ... FRC conversion mode signal, 31 ... Motion vector detection part, 32 ... Video feature extraction unit, 33 ... Interpolation frame generation unit, 34 ... Adaptive enhancer unit, 35 ... Memory I / F, 341 ... Vertical high-pass filter, 342 ... Horizontal high-pass filter, 344, 346 ... Adder, 345 ... Non-linear processing unit 347, selector, 51, coring circuit, 52, gain adjusting circuit, 53, amplitude limiter circuit, F1, current frame, F2, previous frame, FI1, interpolation frame, SC, control signal, MV, motion vector.

Claims (11)

Using the first frame included in the input video signal and the second frame that is temporally past with respect to the first frame, the motion vector of the video is detected in units of pixels or predetermined blocks, and the detected motion vector is And generating an interpolation frame by filtering the pixel information of the first frame and the second frame, and performing an enhancement process on the generated interpolation frame to increase a high-frequency component of the interpolation frame. A processing unit;
An output unit that outputs a combination of the interpolation frame generated by the video processing unit and the frame sequence of the input signal ;
In the enhancement processing of the video processing unit, the amount of enhancer can be controlled according to the characteristics of the video extracted from the input video signal,
The video processing unit extracts, based on the detected motion vector, the number of pixels or blocks in which the motion vector in one frame is detected, and the direction of the motion vector as features of the video,
The first enhancer amount is used when the number of pixels or blocks in which the motion vector pointing in the same direction in one frame is detected is greater than or equal to a predetermined number, and when the number of pixels or blocks is less than a predetermined value, the first A video signal processing apparatus characterized in that the second enhancer amount is smaller than the enhancer amount.   The video signal processing apparatus according to claim 1, wherein the number of pixels or blocks in which motion vectors in different directions are detected in one frame is greater than or equal to a predetermined value, respectively, based on the second enhancer amount. A video signal processing apparatus having a small third enhancer amount.   The video signal processing device according to claim 1 or 2,
  The video processing unit performs enhancement processing on the interpolation frame when the number of blocks or pixels in which the motion vector is detected is greater than a predetermined value in one frame, and the number of blocks or pixels in which the motion vector is detected. The video signal processing apparatus is characterized in that an enhancement process is not performed on the interpolated frame when is less than a predetermined value.   A video display device comprising a display unit for displaying a video whose frame rate has been converted by the video processing device according to claim 1.   Using the first frame included in the input video signal and the second frame that is temporally past with respect to the first frame, the motion vector of the video is detected in units of pixels or predetermined blocks, and the detected motion vector is And generating an interpolation frame by filtering the pixel information of the first frame and the second frame, and performing an enhancement process on the generated interpolation frame to increase a high-frequency component of the interpolation frame. A processing unit;
  An output unit that outputs a combination of the interpolation frame generated by the video processing unit and the frame sequence of the input signal;
  In the enhancement processing of the video processing unit, the amount of enhancer can be controlled according to the characteristics of the video extracted from the input video signal,
  The video processing unit
  The amount of motion of the image is extracted from the detected motion vector as the feature of the image, the edge component of the image is further detected, and the number of pixels including the edge component as the feature of the image is detected from the detected edge component. Extract and
  When the number of pixels including the edge component is a predetermined number or less, the enhancer amount is set to 0,
  When the number of pixels including the edge component is greater than the predetermined number,
  If the amount of motion of the video is less than or equal to a first predetermined value, the first enhancer amount,
  A second enhancer amount larger than the first enhancer amount when the amount of motion of the video is greater than the first predetermined value and less than or equal to a second predetermined value;
  When the motion amount of the video is larger than the second predetermined value, the third enhancer amount is smaller than the second enhancer amount.
  A video signal processing apparatus for controlling the enhancer amount.   The video signal processing apparatus according to claim 5,
  The video processing unit performs enhancement processing on the interpolation frame when the number of blocks or pixels in which the motion vector is detected is greater than a predetermined value in one frame, and the number of blocks or pixels in which the motion vector is detected. The video signal processing apparatus is characterized in that an enhancement process is not performed on the interpolated frame when is less than a predetermined value.   A video display device comprising a display unit for displaying a video whose frame rate has been converted by the video processing device according to claim 5.   Using the first frame included in the input video signal and the second frame that is temporally past with respect to the first frame, the motion vector of the video is detected in units of pixels or predetermined blocks, and the detected motion vector is And generating an interpolation frame by filtering the pixel information of the first frame and the second frame, and performing an enhancement process on the generated interpolation frame to increase a high-frequency component of the interpolation frame. A processing unit;
  An output unit that outputs a combination of the interpolation frame generated by the video processing unit and the frame sequence of the input signal;
  In the enhancement processing of the video processing unit, the amount of enhancer can be controlled according to the characteristics of the video extracted from the input video signal,
  The video processing unit
  From the detected motion vector, the amount of motion of the video is extracted as a feature of the video, and further, a frequency analysis is performed to intensify the frequency component of the video. Extract and
  When the intensity of the high frequency component is below a predetermined value, the enhancer amount is set to 0,
  When the intensity of the high frequency component is greater than the predetermined value,
  If the amount of motion of the video is less than or equal to a first predetermined value, the first enhancer amount,
  A second enhancer amount larger than the first enhancer amount when the amount of motion of the video is greater than the first predetermined value and less than or equal to a second predetermined value;
  When the motion amount of the video is larger than the second predetermined value, the third enhancer amount is smaller than the second enhancer amount.
  A video signal processing apparatus for controlling the enhancer amount.   9. The video signal processing device according to claim 8, wherein the video processing unit performs the frequency analysis for each video region and controls the enhancer amount for each region. .   The video signal processing device according to claim 8 or 9, wherein
  The video processing unit performs enhancement processing on the interpolation frame when the number of blocks or pixels in which the motion vector is detected is greater than a predetermined value in one frame, and the number of blocks or pixels in which the motion vector is detected. The video signal processing apparatus is characterized in that an enhancement process is not performed on the interpolated frame when is less than a predetermined value.   11. A video display device comprising a display unit for displaying a video whose frame rate has been converted by the video processing device according to claim 8.

Images