Patent Application
20110122312
The present disclosure relates to a video signal processing device and a video signal processing method that convert the frame frequency of a video signal that is input into a frame frequency higher than that of the input video signal and then display the video on a display part, such as a liquid crystal panel.
A liquid crystal display device that uses a liquid crystal panel has such a problem that image persistence occurs when a motion picture is displayed. As one of methods of reducing this problem, Patent Document 1 discloses an image display device which interpolates interpolation frames between frames of an input video signal, converts the frame rate of the input video signal into a higher frame rate, and displays the video on a liquid crystal panel. According to the image display device described in Patent Document 1, it is possible to convert a video signal of a frame frequency of 60 Hz into a video signal of a frame frequency of, for example, 120 Hz.
On the other hand, a movie film represents a motion picture by switching still images of 24 frames per second or 30 frames per second. When the video signal of such a movie film is converted into a video signal of a vertical frequency of 60 Hz, which is the standard television signal, 2-3 pull-down conversion or 2-2 pull-down conversion is performed. For example, when a video signal that is 2-3 pull-down converted is displayed, a pattern in which an identical frame continues twice and then the next frame continues three times is repeated. Hence, there occurs a problem that smoothness of motion of images is lost, which is called a motion judder.
Patent Document
When a video signal of a movie film is input to the image display device described in Patent Document 1, if the pattern of frames to be input successively is the same, the interpolation frames to be interpolated between frames have the same pattern as that of the frames before and after the interpolation frame. Hence, with the image display device described in Patent Document 1, it is not possible to improve the motion judder for the video signal of a movie film, and an image gives an uncomfortable feeling to a viewer as a result.
In recent years, video signal processing devices capable of converting a frame rate into a higher one become the mainstream. Therefore, a video signal processing device is demanded which is capable of converting a frame rate into the same one as before or a higher one as well as improving the motion judder even for the video signal of a movie film. Further, a video signal processing device is desired which reduces a processing rate and reduces a frame memory as well as being capable of converting a frame rate into a higher one.
The present subject matter was developed in view of such circumstances and a video signal processing device is provided, which is capable of converting a frame rate into one equal to or higher than that of an input video signal as well as improving the motion judder when the input video signal is a video signal of a movie film and which reduces a processing rate and reduces a frame memory as well as being capable of converting a frame rate into a higher one as before when the input video signal is the standard television signal.
In order to solve the conventional technical problems described above, the present relates to a video signal processing device comprising a frame memory delaying each frame of a first video signal that is input by one frame period to output a second video signal, a motion vector detecting unit configured to detect a motion vector between the frames of the first video signal and the second video signal, a pull-down determining unit configured to determine whether or not the first video signal is a pull-down converted video signal, a sequence generating unit configured to generate pull-down sequence information based on the determination result of the pull-down determining unit and the motion vector, an image shifting unit (N+1) variable delaying unit to delay the first or second video signal, the (N+1) variable delaying units configured to control delay amounts with a repetition of a predetermined plurality of frames as a cycle based on the pull-down sequence information and the motion vector control delay amounts with a repetition of a predetermined plurality of frames as a cycle to generate the (N+1) of motion-compensated video signals, a time axis emphasizing unit configured to emphasize N motion-compensated video signals among the (N+1) motion-compensated video signals, respectively, in a time axis direction based on at least one or more of the motion-compensated video signals including the neighboring motion-compensated video signals in the order of time series to output N emphasized video signals, and a time series conversion memory converting the frame frequencies of the N emphasized video signals into those N times the original ones and outputting them in the order of time series.
Further, the present disclosure relates to a video signal processing device comprising a frame memory delaying each frame of a first video signal that is input by one frame period to output a second video signal, a motion vector detecting unit configured to detect a motion vector between the frames of the first video signal and the second video signal, a pull-down determining unit configured to determine whether or not the first video signal is a pull-down converted video signal, a sequence generating unit configured to generate pull-down sequence information based on the determination result of the pull-down determining unit, an image shifting unit having (M+1) variable delaying units to delay the first or second video signal, the (M+1) variable delaying units configured to control delay amounts with a repetition of a predetermined plurality of frames as a cycle based on the pull-down sequence information when the pull-down determining unit determines that the first video signal is a pull-down converted video signal and the motion vector, where M is a factor of N that is a natural number not less than 2 and each delay amount by the (M+1) variable delaying units being different from another in one frame, the (M+1) variable delaying units configured to compensate for a motion of at least one of the first and second video signals to generate the (M+1) motion-compensated video signals, a time axis emphasizing unit configured to emphasize M motion-compensated video signals among the plurality of motion-compensated video signals, respectively, in the time axis direction based on at least one or more of the motion-compensated video signals including the neighboring motion-compensated video signals in the order of time series to output M emphasized video signals when the pull-down determining unit determines that the first video signal is a pull-down converted video signal, and a time series conversion memory converting the frame frequencies of the motion-compensated video signal and the emphasized video signal into those N times the original ones and outputting them in the order of time series as well as repeatedly outputting the combination of at least one emphasized video signal based on the motion-compensated video signal N/M times per output for the M motion-compensated video signals, respectively, when the pull-down determining unit determines that the first video signal is a pull-down converted video signal.
Further, the present disclosure relates to a video signal processing device comprising a frame memory delaying each frame of a first video signal that is input by one frame period to output a second video signal, a pull-down determining unit configured to determine whether or not the first video signal is a pull-down converted video signal and to determine the kind of pull-down conversion when the first video signal is a pull-down converted video signal, a selecting unit configured to select the first video signal delayed by a first delay amount or the second video signal delayed by a second delay amount based on pull-down sequence information generated when the pull-down determining unit determines that the first video signal is a 2-3 pull-down converted video signal_and to output the video signal as a selected video signal, a time axis emphasizing unit configured to emphasize first video signal delayed by the first delay amount in a time axis direction based on the selected video signal to output the first video signal and to emphasize the second video signal delayed by the second delay amount in the time axis direction based on the selected video signal to output the second video signal, and a time series conversion memory converting the frame frequency of the and second video signals output by the time axis emphasizing unit into these N times the frame frequency of the first video signal input to the frame memory and repeatedly outputting 5K video signals per output in the order of time series, where N is an even number and K is 1.2 of N.
Further, the present disclosure relates to a video signal processing device comprising a frame memory delaying each frame of a first video signal that is input by one frame period to output a second video signal, a pull-down determining unit configured to determine whether or not the first video signal is a pull-down converted video signal and to determine the kind of pull-down conversion when the first video signal is a pull-down converted video signal
Further, the present disclosure relates to a video signal processing method comprising the steps of delaying each frame of a first video signal that is input by one frame period to output a second video signal, detecting a motion vector between the frames of the first video signal and the second video signal, determining whether or not the first video signal is a pull-down converted video signal, generating pull-down sequence information based on the results of determining whether or not the first video signal is a pull-down converted video signal and of determining the kind of pull-down conversion, compensating for the motion of at least one of the first and second video signals by (N+1) variably delaying steps to generate (N+1) motion-compensated video signals different from one another, where N is a natural number not less than 2 and the (N+1) variably delaying steps are included in this step for delaying the first or second video signal, and controlling delay amounts by the (N+1) variably delaying steps with a repetition of a predetermined plurality of frames as a cycle based on the pull-down sequence information and the motion vector, the delay amounts by the (N+1) variably delaying steps in one frame being different from one another in one frame, emphasizing N motion-compensated video signals among the (N+1) motion-compensated video signals, respectively, in the time axis direction based on at least one or more of the motion-compensated video signals including the neighboring motion-compensated video signals in the order of time series to output N emphasized video signals, and converting the frame frequencies of the N emphasized video signals into those N times the original ones and outputting them in the order of time series.
Further, the present disclosure relates to a video signal processing method comprising the steps of delaying each frame of a first video signal that is input by one frame period to output a second video signal, detecting a motion vector between the frames of the first video signal and the second video signal, determining whether or not the first video signal is a pull-down converted video signal, generating pull-down sequence information based on the determination result in the step of determining whether or not the first video signal is a pull-down converted video signal, compensating for the motion of at least one of the first and second video signals by (M+1) variably delaying steps to generate (M+1) motion-compensated video signals different from one another, where M is a factor of N that is a natural number not less than 2 and the (M+1) variably delaying steps are included in this step for delaying the first or second video signal, and controlling delay amounts by the (M+1) variably delaying steps with a repetition of a predetermined plurality of frames as a cycle based on the pull-down sequence information when the first video signal is determined to be a pull-down converted video signal and the motion vector to generate a plurality of motion-compensated video signals, each delay amount by the (M+1) variably delaying steps in one frame being different from one another in one frame, emphasizing M video signals among the (M+1) motion-compensated video signals, respectively, in the time axis direction based on at least one or more of the motion-compensated video signals including the neighboring motion-compensated video signals in the order of time series to output M emphasized video signals when the step of determining whether or not the first video signal is a pull-down converted video signal determines that the first video signal is a pull-down converted video signal, and converting the frame frequencies of the motion-compensated video signal and the emphasized video signal into those N times the original ones and outputting them in the order of time series as well as repeatedly outputting the combination of at least one emphasized video signal based on a motion-compensated video signal and the motion-compensated video signal N/M times per output for the M motion-compensated video signals, respectively, when the step of determining whether or not the first video signal is a pull-down converted video signal determines that the first video signal is a pull-down converted video signal.
Further, the present disclosure relates to a video signal processing method comprising the steps of delaying each frame of a first video signal that is input by one frame period to output a second video signal, determining whether or not the first video signal is a pull-down converted video signal and determining the kind of pull-down conversion when the first video signal is a pull-down converted video signal, selecting the first video signal delayed by a first delay amount or the second video signal delayed by a second delay amount based on pull-down sequence information generated when the step of determining whether or not the first video signal is a pull-down converted video signal and determining the kind of pull-down conversion when the first video signal is a pull-down converted video signal determines that the first video signal is a 2-3 pull-down converted video signal and outputting the video signal as a selected video signal, emphasizing the first video signal delayed by the first delay amount in a time axis direction based on the selected video signal to output the first video signal and emphasizing the second video signal delayed by the second delay amount in the time axis direction based on the selected video signal to output the second video signal, and converting the frame frequency of the first and second video signals output by the step of emphasizing a video signal in the time axis direction into those N times the frame frequency of the first video signal input to the frame memory and repeatedly outputting 5K video signals per output in the order of time series, where N is an even number and K is ½ of N.
Further, the present disclosure relates to a video signal processing method comprising the steps of delaying each frame of a first video signal that is input by one frame period to output a second video signal, determining whether or not the first video signal is a pull-down converted video signal and determining the kind of pull-down conversion when the first video signal is a pull-down converted video signal, generating L−(L+1) pull-down sequence information (L is 2N+(N−1)/2 and N is an odd number not less than 3) based on the motion vector when the step of determining whether or not the first video signal is a pull-down converted video signal and determining the kind of pull-down conversion when the first video signal is a pull-down converted video signal determines that the first video signal is a 2-3 pull-down converted video signal, selecting the first video signal or the second video signal and repeatedly outputting L then (L+1) video signals having the same time series per output for each one frame cycle as selected video signals when the step of determining whether or not the first video signal is a pull-down converted video signal and determining the kind of pull-down conversion when the first video signal is a pull-down converted video signal determines that the first video signal is a 2-3 pull-down converted video signal, emphasizing the selected video signal after the time series switches by one frame cycle in a time axis direction based on the selected video signal in the previous frame cycle, and converting the frame frequency of the first video signal into one N times the original one and repeatedly outputting L then (L+1) selected video signals per output for each one frame cycle in the order of time series.
A video signal processing device and a video signal processing method in a first embodiment are explained below with reference to
The image memory 10 stores the first video signal F0 corresponding to one frame and delays it by one frame period to output a second video signal F1. That is, the second video signal F1 is a video signal one frame ahead of the first video signal F0. The first video signal F0 and the second video signal F1 are supplied to each of a motion vector detection circuit 20 having a motion vector detecting unit and a pull-down determination circuit 22 having a pull-down determining unit.
The motion vector detection circuit 20 detects a motion vector MV between the frames of the supplied first video signal F0 and the second video signal F1. It is also possible to use, for example, a matching method etc. to detect a motion vector. The detected motion vector MV is supplied to a motion vector conversion circuit 21.
The pull-down determination circuit 22 determines whether or not the first video signal F0 is a pull-down converted video signal from the first video signal F0 that is input and the second video signal F1 and supplies the determination result to a sequence generation circuit 23. For example, it is possible to determine whether or not the first video signal F0 is a pull-down converted video signal by finding a difference between frames of the first video signal F0 and the second video signal F1. In this case, it is also possible to determine whether the first video signal F0 is a 2-2 pull-down converted video signal or a 2-3 pull-down converted video signal by detecting a cycle in which the difference between frames becomes large.
When the first video signal F0 includes pull-down converted information, it is also possible to determine whether or not the first video signal F0 is a pull-down converted video signal using the information. It may also be possible to determine whether or not the first video signal F0 is a pull-down converted video signal and the kinds of pull-down conversion using other publicly-known techniques.
The sequence generation circuit 23 having a sequence generating unit generates a sequence signal (pull-down sequence information) to execute a pull-down sequence based on the determination result supplied from the pull-down determination circuit 22 and supplies it to the motion vector conversion circuit 21. The pull-down sequence means to perform control determined in advance in the motion vector conversion circuit 21 etc. depending on the kind of pull-down conversion of the first video signal F0. The kinds of pull-down conversion include a case where pull-down conversion is not performed. The motion vector conversion circuit 21 generates three kinds of delay control signals a1 to a3 to control the delay amount in each of variable delay circuits 30 to 32, to be described later, based on the motion vector MV and the sequence signal. The delay control signals a1 to a3 are supplied to the respective variable delay circuits 30 to 32 having an image shifting unit.
The variable delay circuit 30 delays the second video signal F1 by a predetermined amount based on the delay control signal a1 to generate a motion-compensated video signal b1 the motion of which is compensated for and supplies the motion-compensated video signal b1 to a time axis emphasis circuit 40. The variable delay circuit 31 delays the first video signal F0 by a predetermined amount based on the delay control signal a2 to generate a motion-compensated video signal b2 the motion of which is compensated for and supplies it to the time axis emphasis circuit 40 and a time axis emphasis circuit 41. The variable delay circuit 32 delays the first video signal F0 by a predetermined amount based on the delay control signal a3 to generate a motion-compensated video signal b3 the motion of which is compensated for and supplies it to the time axis emphasis circuit 41.
The variable delay circuits 30 to 32 are circuits capable of variably delaying video signals based on the respective delay control signals a1 to a3. The image shifting unit is not limited to the variable delay circuit shown in
On the other hand, when an image is displayed on a liquid crystal panel, the response speed is low, and therefore, there is such a problem that image persistence occurs when a motion picture is displayed. In order to reduce the image persistence, it is necessary to emphasize a video signal near the video signal when the video signals are put side by side in the order of time series in a time axis direction.
The time axis emphasis circuits 40, 41 each having a time axis emphasizing unit are filters that emphasize video signals in the time axis direction, and filters to prevent image persistence by raising the voltage of a desired video signal steeply.
fo=fa+c(fa−fb) (1)
Here, the gain coefficient c is intended to determine the degree of emphasis of the video signal fa and set in accordance with the response characteristic of the liquid crystal. For example, when the response speed of the liquid crystal material is comparatively high and the image persistence when a motion picture is displayed is slight, the gain coefficient c is set small and when the response speed is comparatively low and the image persistence when a motion picture is displayed is remarkable, the gain coefficient c is set large.
In the time axis emphasis circuit 40, the video signal fa is the motion-compensated video signal b2 and the video signal fb is the motion-compensated video signal b1. In the time axis emphasis circuit 41, the video signal fa is the motion-compensated video signal b3 and the video signal fb is the motion-compensated video signal b2. Then, the time axis emphasis circuit 40 emphasizes the motion-compensated video signal b2 in the time axis direction based on the motion-compensated video signal b1 to generate an emphasized video signal DF0 and supplies it to a time series conversion memory 50. The time axis emphasis circuit 41 emphasizes the motion-compensated video signal b3 in the time axis direction based on the motion-compensated video signal b2 to generate an emphasized video signal DF1 and supplies it to the time series conversion memory 50.
The time axis emphasis circuits 40, 41 are not limited to the configuration shown in
The time series conversion memory 50 temporarily stores the video signals DF0 and DF1 that are supplied and outputs the video signals DF0 and DF1 in this order as a video signal F0′ to a liquid crystal panel, not shown schematically. Further, the time series conversion memory 50 doubles the frame frequency to output the video signal F0′.
Next, the case is explained, where the video signal F0, which is a 2-3 pull-down converted video signal recorded by 24 frames/second, such as a movie film, is input to the image signal processing device 1. The 2-3 pull down conversion is conversion in which an odd-numbered frame is converted into two fields and an even-numbered frame into three fields. On the other hand, the 3-2 pull-down conversion, converse to the above, is conversion in which an odd-numbered frame is converted into three fields and an even-numbered frame into two fields. In the following explanation, it is assumed that the 2-3 pull-down conversion includes the 3-2 pull-down conversion. When the pull-down determination circuit 22 determines that the first video signal F0 is a 2-3 pull-down converted signal, the sequence generation circuit 23 generates a sequence signal to execute 2-3 pull-down sequence information and transmits it to the motion vector conversion circuit 21.
Frames S2, S3 and S4 are the same image pattern and frames S5, S6 are the same image pattern, and the same image pattern is input to the three successive frames, then the same image pattern is input to the two successive frames and this is repeated.
The motion vector detection circuit 20 detects a motion vector amount when frame S1 changes to S2 as the MV. When the first video signal F0 that is input is not subjected to any processing in the video signal processing device 1, there is produced a shift between the direction of visual tracking and the moving object as shown in
The motion vector conversion circuit 21 supplies the delay control signals a1 to a3 to the respective variable delay circuits 30 to 32 to control the delay amount so as to be as shown, for example, in Table 1 based on the motion vector amount MV supplied from the motion vector detection circuit 20 and the sequence signal. In Table 1, the delay amount is the same in a 5/60 second cycle and it is assumed that the delay amount is the same when the first video signal F0 (frame input currently) is S2 and S7. The delay amount shown in Table 1 represents a relative delay amount. Further, it is not possible to realize processing to delay in the negative direction in the variable delay circuit 30, and therefore, it is assumed that the delay amount in each delay circuit, to be explained in the following, is a relative one.
The variable delay circuits 31, 32 delay the first video signal F0 based on the delay control signals a2, a3 supplied from the motion vector conversion circuit 21. For example, the variable delay circuit 32 delays the first video signal F0 by a delay amount of −2·MV/5 to generate the motion-compensated video signal b3 when the first video signal F0 that is supplied is frame S2. The variable delay circuit 31 delays the first video signal F0 by a delay amount of −3·MV/5 to generate the motion-compensated video signal b2 when the first video signal F0 that is supplied is frame S2.
The variable delay circuit 30 delays the second video signal F1, which is the previous video signal, based on the delay control signal a1 supplied from the motion vector conversion circuit 21. For example, the variable delay circuit 30 delays the second video signal F1 by a delay amount of +MV/5 to generate the motion-compensated video signal b1 when the second video signal F1 that is supplied is frame S1.
The delay processing for the second video signal F1 in the variable delay circuit 30 is the same as the processing for the first video signal F0 in the previous frame of the variable delay circuit 32. Consequently, it may also be possible to generate the motion-compensated video signal b3 by delaying the output of the video signal of the variable delay circuit 32, instead of the variable delay circuit 30, by an amount corresponding to one frame. In such a case, a memory is necessary for the delay corresponding to one frame.
Next, the case is explained, where the first video signal F0, which is a 2-2 pull-down converted motion picture of 30 frames/sec, is input to the video signal processing device 1. When the pull-down determination circuit 22 determines that the first video signal F0 is a 2-2 pull-down converted signal, the sequence generation circuit 23 generates a sequence signal to execute the 2-2 pull-down sequence and transmits it to the motion vector conversion circuit 21.
The motion vector detection circuit 20 detects the motion vector amount when S1 changes to S2 as the MV as in the case of the 2-3 pull-down. When the first video signal F0 that is input is not subjected to any processing in the video signal processing device 1, there is produced a shift between the direction of visual tracking and the moving object as shown in
The motion vector conversion circuit 21 supplies the delay control signals a1 to a3 to the respective variable delay circuits 30 to 32 to control the delay amount so as to be as shown, for example, in Table 2 based on the motion vector amount MV supplied from the motion vector detection circuit 20 and the sequence signal. The delay amount shown in Table 2 is the same in a 2/60 second cycle and it is assumed that the delay amount is the same when the frames of the first video signal F0 are S2 and S4.
The variable delay circuits 31, 32 delay the first video signal F0 based on the delay control signals a2, a3 supplied from the motion vector conversion circuit 21. For example, the variable delay circuit 32 delays the first video signal F0 by a delay amount of −MV/4 to generate the motion-compensated video signal b3 when the first video signal F0 that is supplied is frame S2. The variable delay circuit 31 delays the first video signal F0 by a delay amount of −MV/2 to generate the motion-compensated video signal b2 when the first video signal F0 that is supplied is frame S2.
The variable delay circuit 30 delays the second video signal F1 based on the delay control signal a1 supplied from the motion vector conversion circuit 21. For example, the variable delay circuit 30 delays the second video signal F1 by a delay amount of +MV/4 to generate the motion-compensated video signal b1 when the second video signal F1 that is supplied is frame S1.
Next, the case is explained, where the video signal F0 having a vertical frequency of 60 Hz, which is the vertical frequency of a standard television, is input to the video signal processing device 1. When the pull-down determination circuit 22 determines that the first video signal F0 is not a pull-down converted signal, the sequence generation circuit 23 generates a sequence signal to perform control when pull-down conversion is not performed and transmits it to the motion vector conversion circuit 21.
The motion vector detection circuit 20 detects the motion vector amount when S1 changes to S2 as the MV. The motion vector conversion circuit 21 supplies the delay control signals a1 to a3 to the respective variable delay circuits 30 to 32 to control the delay amount so as to be as shown, for example, in Table 3 based on the motion vector amount MV supplied from the motion vector detection circuit 20 and the sequence signal. The delay amount shown in Table 3 is the same at all times regardless of the frame.
The variable delay circuit 32 sets the delay amount to zero by the delay control signal a3 supplied from the motion vector conversion circuit 21, and therefore, does not delay the first video signal F0 and uses it as the motion-compensated video signal b3 as it is. The variable delay circuit 31 delays the first video signal F0 by a delay amount of −MV/2 to generate the motion-compensated video signal b2 based on the delay control signal a2 supplied from the motion vector conversion circuit 21.
As explained above, according to the first embodiment, it is possible to compensate for the motion by varying the delay amount in the variable delay circuits 30 to 32 depending on whether or not a video signal is pull-down converted, and therefore, a high frame rate can be realized. Because of this, it is possible to provide a smooth image regardless of whether or not the video signal is pull-down converted.
Further, in the first embodiment, two emphasized video signals based on two motion-compensated video signals are converted so as to have a frame rate twice that of the first video signal F0, which is an input video signal, using the three variable delay circuits (image shifting units) that generate three motion-compensated video signals, however, this is not limited. For example, when the frame rate of an input video signal is converted into a frame rate N (N is a natural number not less than 2) times the original one, an image shifting unit (variable delay circuit) for generating a plurality, that is, N or more, of motion-compensated video signals is necessary. Then, it is only required to select N motion-compensated video signals from among the plurality of motion-compensated video signals to generate N emphasized video signals emphasized in the time axis direction.
Furthermore, the delay amounts included in the delay control signals a1 to a3 to be supplied to the variable delay circuits 30 to 32 may be those other than the delay amounts shown in Table 1 to Table 3. Such a configuration may be accepted, in which to the variable delay circuits 30 to 32, one of or both the first video signal F0 and the second video signal F1 delayed by an amount corresponding to one frame are supplied and thus the motion-compensated video signals b1 to b3 are generated. Further, as shown in
A second embodiment differs from the first embodiment in that when the first video signal F0 is determined to be a pull-down converted signal, the time series conversion memory outputs a plurality of video signals based on the same motion-compensated video signal. In the second embodiment, points that differ from those in the first embodiment are explained.
A selection circuit 60 selects the motion-compensated video signal b2 delayed by a predetermined delay amount in the variable delay circuit 31 when the pull-down determination circuit 22 determines that the first video signal F0 is a pull-down converted signal. On the other hand, the selection circuit 60 selects a video signal generated in the time axis emphasis circuit 41 when the pull-down determination circuit 22 determines that the first video signal F0 is not a pull-down converted signal.
The case is explained, where the pull-down determination circuit 22 determines that the first video signal F0 is a 2-3 pull-down converted video signal.
The motion vector detection circuit 20 detects the motion vector amount when frame S1 changes to S2 as the MV. When the pull-down determination circuit 22 determines that the first video signal F0 is a pull-down converted signal, the sequence generation circuit 23 generates a sequence signal to execute a pull-down sequence for a pull-down converted signal and transmits it to the motion vector conversion circuit 21.
The motion vector conversion circuit 21 supplies the delay control signals a1, a2 to the variable delay circuits 30, 31 to control the delay amount so as to be shown as, for example, in Table 4 based on the motion vector amount MV supplied from the motion vector detection circuit 20 and the sequence signal. The delay amount shown in Table 4 is the same in a 5/60 second cycle and it is assumed that the delay amount is the same when the frames of the first video signal F0 are S2 and S7.
The variable delay circuit 31 delays the first video signal F0 based on the delay control signal a2 supplied from the motion vector conversion circuit 21. For example, the variable delay circuit 31 delays the first video signal F0 by a delay amount of −2·MV/5 to generate the motion-compensated video signal b2 when the first video signal F0 that is supplied is frame S2.
The variable delay circuit 30 delays the second video signal F1 in the previous frame by a predetermined delay amount supplied from the motion vector conversion circuit 21. For example, the variable delay circuit 30 delays the second video signal F1 by a delay amount of +MV/5 when the second video signal F1 that is input is frame S1.
The time series conversion memory 50 doubles the frame frequency to output the video signals DF0, DF1 in this order as the video signal F0′ to a liquid crystal panel, not shown schematically. Consequently, as shown in
There is a case where it is preferable not to improve too much the motion judder or blurring of a motion picture depending on its content when the video signal is a pull-down converted still image of a movie film. That is, when the video signal processing device 1 in the first embodiment is used, there may be a case where an image causes a feeling of discomfort because its motion judder is improved too much. In such a case, it is recommended to output a video signal based on the same motion-compensated video signal b2 in two frames each to compensate for the motion moderately using the selection circuit 60 explained in the second embodiment. It is also possible to prevent the effect of correcting the blurring of a motion picture from becoming strong because the time axis emphasis circuit 41 is not used in a pull-down converted video signal.
Next, the case is explained, where a video signal having a vertical frequency of 60 Hz, which is the standard television signal, is input to the video signal processing device 2. When the pull-down determination circuit 22 determines that the first video signal F0 is not a pull-down converted signal, the selection circuit 60 selects the video signal as the video signal DF1, which is the motion-compensated video signal b3 supplied from the fixed delay circuit 33 and emphasized in the time axis direction by the time axis emphasis circuit 41.
In this case, the time series conversion memory 50 temporarily stores the video signal DF1 emphasized in the time axis direction by the time axis emphasis circuit 41 and the video signal DF0 emphasized in the time axis direction by the time axis emphasis circuit 40. The time series conversion memory 50 doubles the frame frequency to output the video signals DF0, DF1 in this order as the video signal F0 to a liquid crystal panel, not shown schematically. The delay amounts delayed by the variable delay circuits 30, 31 are the same as those in Table 3 in the first embodiment.
Consequently, as in
As explained above, according to the second embodiment, when the first video signal F0, which is an input video signal, is a pull-down converted video signal, it is possible to convert the frame rate into a higher one than that of the input video signal, however, the motion is compensated for more moderately than in the first embodiment. When the first video signal F0 is the standard television video signal, it is possible to compensate for the motion as in the first embodiment and realize a high frame rate.
Further, the time axis emphasis circuits 40, 41 emphasize the motion-compensated video signals b2, b3, respectively, in the time axis direction using only the previous motion-compensated video signal in the time series, however, it is not limited to the previous motion-compensated video signal. It is also possible to emphasize them in the time axis direction using a plurality of motion-compensated video signals.
Furthermore, in the second embodiment, in the case of a pull-down converted input video signal, one motion-compensated video signal and one emphasized video signal based on the motion-compensated video signal are converted so as to have a frame rate twice that of the first video signal F0, which is an input video signal, using two variable delay circuits (image shifting units) that generate two motion-compensated video signals, however, this is not limited.
For example, when the frame rate of an input video signal is converted into a frame rate N (N is a natural number not less than 2) times the original one, the video signal processing device 2 requires an image shifting unit (variable delay circuit) that generates a plurality of motion-compensated video signals. Then, the video signal processing device 2 selects M (M is a natural number and a factor of N) motion-compensated video signals from among the plurality of motion-compensated video signals and the emphasized video signals, which are the respective motion-compensated video signals emphasized in the time axis direction. Further, for each of the M selected video signals (motion-compensated video signals or emphasized video signals) having the same time series, the combination of the motion-compensated video signal and the emphasized video signal having the same time series based on the motion-compensated video signal is output N/M (M is a natural number and a factor of N) times per output. Consequently, the video signal processing device 2 is a device that generates N video signals in total by outputting each video signal having the same time series N/M times per output and converts the frame rate into a frame rate N times the original one. It is recommended to reduce image persistence by outputting at least one or more emphasized video signals together for each of the M selected signals.
In the second embodiment, explanation is given when N=2, M=1, however, M=1 when N=3 and the video signal based on the same motion-compensated video signal corresponding to three frames is output in 1/60 sec. The relationship is such that M=1 or 2 when N=4, M=1 when N=5, and M=1, 2, or 3 when N=6. As a result, two or more frames of the video signals based on the same motion-compensated video signal are output in the same number per output and thus the motion judder is improved.
Further, the delay amounts included in the delay control signals a1 to a3 to be supplied to the variable delay circuits 30, 31 may be those other than the delay amounts shown in Table 4. Such a configuration may also be accepted in which one of or both the first video signal F0 and the second video signal F1 delayed by an amount corresponding to one frame are supplied to the variable delay circuits 30, 31 and thus the motion-compensated video signals b1 to b3 are generated. Furthermore, as shown in
A third embodiment differs from the first and second embodiments in that the video signal processing device generates a sequence signal to execute a 5-5 pull-down sequence so that a plurality of, that is, five video signals based on the same motion-compensated video signal are output per output when the first video signal F0 is determined to be a 2-3 pull-down converted signal. In the third embodiment, points different from the first and second embodiments are explained.
A selection circuit 61 having a selecting unit selects either video signal b3 delayed by a predetermined delay amount in the fixed delay circuit 33 or video signal b1 delayed by a predetermined delay amount in the fixed delay circuit 34 based on the sequence signal to execute a pull-down sequence generated in the sequence generation circuit 23 when the pull-down determination circuit 22 determines that the first video signal F0 is a pull-down converted signal.
A selection circuit 62 selects the video signal selected in the selection circuit 61 as a selected video signal when the pull-down determination circuit 22 determines that the first video signal F0 is a pull-down converted signal. On the other hand, the selection circuit 61 selects the motion-compensated video signal b2 supplied from the variable delay circuit 31 as a selected video signal when the pull-down determination circuit 22 determines that the first video signal F0 is not a pull-down converted signal.
The case is explained, where the pull-down determination circuit 22 determines that the first video signal F0 is a 2-3 pull-down converted video signal. The sequence generation circuit 23 generates a sequence signal to execute a 5-5 pull-down sequence. The selection circuit 61 selects the video signal b1 or b3 as a selected video signal based on the sequence signal. The time series conversion memory 50 outputs the video signal F0′ based on the same video signal for five frames per output. That is, the video signal F0′ is switched in a 5/120 second cycle.
The time series conversion memory 50 temporarily stores the selected video signal including the motion-compensated video signal b1 or b3 having passed through the time axis emphasis circuit 40 as DF0 and the selected video signal including the motion-compensated video signal b1 or b3 having passed through the time axis emphasis circuit 41 as DF1. Then, the time series conversion memory 50 doubles the frame frequency to output the video signals DF0, DF1 in this order as the video signal F0′ to a liquid crystal panel, not shown schematically. As shown in
When the first video signal F0 is a 2-2 pull-down converted signal, the motion-compensated video signal b1 or b3 selected by the selection circuit 61 is used as a selected video signal so that the time series conversion memory 50 outputs the video signal F0′ based on the same motion-compensated video signal in four frames per output. When the first video signal F0 is the standard television signal, the motion-compensated video signal delayed in the variable delay circuit 31 is selected by the selection circuits 61 and 62 and the processing is the same as that in the first embodiment.
As explained above, according to the third embodiment, when the first video signal F0, which is an input video signal, is a 2-3 pull-down converted video signal, the motion is not compensated for by the video signal processing device 3. However, the first video signal F0 in which two frames and three frames are repeated becomes the video signal F0′ in which five frames are repeated at all times, and therefore, the motion judder is improved somewhat and a viewer views the image as an image that suppresses a feeling of discomfort to a certain degree.
There is a case where it is preferable not to compensate for the motion depending on its content when a video signal is a pull-down converted still image of a movie film. That is, in the second embodiment also, if the motion is compensated for too much and a feeling of discomfort is caused, it is recommended not to compensate for the motion of a pull-down converted video signal as explained in the third embodiment.
Further, in the second embodiment, in the case of a 2-3 pull-down converted input video signal, the configuration is shown in which five video signals based on the same selected video signal including the video signal emphasized in the time axis direction are output per output using the selection circuit and the frame rate is doubled, however, the configuration is not limited to this.
For example, using two variable delay circuits (image shifting units) that generate two motion-compensated video signals, one motion-compensated video signal and one emphasized video signal based on the motion-compensated video signal are converted so as to have a frame rate twice that of the first video signal F0, which is an input video signal, however, this is not limited.
For example, when the first video signal F0 that is 2-3 pull-down converted is input to the video signal processing device 3 that converts the frame rate of an input video signal into N (N is an even number not less than 2) times the original one, the sequence generation circuit 23 is only required to output a sequence signal to execute a 5K-5K (K is N/2) pull-down sequence. The 5K-5K pull-down sequence means to control so as to output 5K video signals per output based on the motion-compensated video signal having the same time series.
In the video signal processing device 3 in the third embodiment, explanation is given on the assumption that N=2, K=1, however, K=2 when N=4 and a 10-10 pull-down sequence is executed. Consequently, 10 video signals based on the same selected video signal are output in 5/120 sec. After that, 10 video signals based on the same selected video signal one frame cycle apart are output in 5/120 sec. As a result, 10 video signals each, 20 in total, based on the two kinds of selected video signal are output in 5/60 sec. Similarly, K=3 when N=6 and a 15-15 pull-down sequence is executed. Consequently, the video signal processing device 3 outputs 5K frames of the video signal based on the same selected video signal per output and thus the motion judder is improved somewhat.
When the first video signal F0 that is 2-3 pull-down converted is input to the video signal processing device 3 that converts the frame rate of an input video signal into N (N is an odd number not less than 3) times the original one, the sequence generation circuit 23 is only required to generate an L−(L+1) pull-down sequence. The L−(L+1) pull-down sequence means to control so that L video signals having the same time series are output and then (L+1) video signals having the same time series are output, and afterward, L video signals and (L+1) video signals are output repeatedly. Here, L is 2N+(N−1)/2. Further, like the 2-3 pull-down sequence, the L−(L+1) pull-down sequence includes an (L+1)-L pull-down sequence.
For example, L=7 when N=3 and a 7-8 pull-down sequence is executed. Consequently, seven video signals (selected video signals or emphasized video signals) having the same time series are output in 7/180 sec. After that, eight video signals (selected video signals or emphasized video signals) having the same time series one frame cycle apart are output in 8/180 sec. As a result, seven and eight video signals of two kinds having different time series are output in 5/60 sec. Similarly, L=9 when N=5 and a 12-13 pull-down sequence is executed. Consequently, the video signal processing device 3 outputs L frames of the video signal having the same time series then (L+1) video signals having the same time series repeatedly per output for any L and thus the motion judder is improved somewhat even when N is an odd number not less than 3.
It is desirable for a viewer to be able to select two or more embodiments of the first to third embodiments according to the viewer's preference. In such a case, only one circuit common to the first to third embodiments is required.
According to the present subject matter, when an input video signal is a signal of a movie film, it is possible to convert the frame rate into one equivalent to or higher than that of the input video signal as well as improving the motion judder, and it is also possible to similarly convert the frame rate into one higher than that of the input video signal when the input video signal is the standard television signal. Further, it is possible to achieve reduction in processing speed and reduction of a frame memory.
| Number | Date | Country | Kind |
|---|---|---|---|
| P2008-187739 | Jul 2008 | JP | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/JP2009/062859 | 7/16/2009 | WO | 00 | 1/17/2011 |
