JP5015089B2 - Frame rate conversion device, frame rate conversion method, television receiver, frame rate conversion program, and recording medium recording the program - Google Patents

Description

  The present invention relates to a processing apparatus for converting a video signal such as a television receiver, and more particularly to a frame rate conversion apparatus for converting a frame rate of a moving image.

  NTSC television receivers (hereinafter simply referred to as TVs or TVs) through filming of film sources such as movies taken at 24 frames per second through television broadcasts or package media such as videotapes or DVDs In order to display the video, it is necessary to convert it into a video image signal of 60 fields per second. This conversion is generally called telecine conversion by 3: 2 pull-down method.

  At this time, each frame image on the film is captured as a digital image. Assuming that each line in the captured digital image is numbered 1, 2, 3,... In order from the top, the image of the odd line is an odd field of the video image signal that is an interlaced scanning (interlace) method. The image of the even line is developed into the even field of the video image signal.

  Here, the details of the telecine conversion by the 3: 2 pull-down method will be specifically described with reference to FIG. FIG. 8 is a diagram schematically showing telecine conversion by the 3: 2 pull-down method. In FIG. 8, the frames of the film taken at a time interval of 1/24 seconds are arranged from left to right. In the 3: 2 pull-down method, for example, the leftmost frame among the film frames is allocated to two fields in the video image signal, and the next frame (second frame from the left) is allocated to three fields in the video image signal. . In the same manner, 24 frames / second film is converted into a video image signal of about 60 fields / second by repeatedly assigning each frame to two fields and three fields.

  By performing such a telecine conversion by the 3: 2 pull-down method, the television can handle a film source image in the same manner as an interlaced image by a normal NTSC method. At this time, by detecting that the video image has been telecine-converted by the 3: 2 pull-down method, higher-definition progressive scanning display is possible. Such detection and display are generally referred to as film mode or cinema mode.

  A method for detecting a video image that has been telecine-converted by the 3: 2 pull-down method and displaying a higher-definition progressively-scanned video image will be described below with reference to FIG.

  First, the telecine conversion device detects whether or not the video image signal is in a film mode by comparing the image contents of each field of the video image signal. An example of this detection method will be specifically described with reference to field C and field E in FIG. The video content in field E in FIG. 8 is always the same repeated field as the video content in field C. This is peculiar to a video image signal subjected to telecine conversion by the 3: 2 pull-down method. Therefore, for example, by calculating the sum of the absolute differences of the pixel values of the field E and the field C, it is detected whether or not the video image signal is in the film mode.

  In this way, when the telecine conversion device detects that the video signal displayed on the television is a video image signal that has been telecine converted by the 3: 2 pull-down method, the video constituting the frame of the original film An odd field and an even field of the video signal are specified, reconstructed, and sequentially displayed as a scanned image.

  More specifically, with reference to FIG. 8, by using the odd field A and the even field B, the sequentially scanned image I is reconstructed. Similarly, the sequentially scanned image II is reconstructed using the odd field A and the even field B. That is, the sequentially scanned images I and II are the same image. Further, by using the odd field C (= E) and the even field D, the sequentially scanned images III to V are reconstructed. Similar to the sequentially scanned images I and II, the sequentially scanned images III to V are the same image. The point to be noted here is that the car always moves from left to right at the position on the display screen in the original film frame, but the car is stopped and displayed between the sequential scanning displays I and II. The vehicle is stopped and displayed even during the sequential scanning displays III to V.

  This point will be described in detail below with reference to FIG. FIG. 9 is a diagram schematically showing the display of a video image signal subjected to telecine conversion by the 3: 2 pull-down method. In FIG. 9, the horizontal axis indicates the position in the display screen, and the vertical axis indicates time. That is, FIG. 9 shows a state in which an object is moving horizontally from left to right at a constant speed. The vertical time interval is 1 frame = 1/60 second interval. In a hold-type video display device such as a liquid crystal panel, the display state during this time does not change.

  As shown in FIG. 9, the image data is written and updated at intervals of 1/60 seconds as shown in FIG. 9, but in FIG. 8, the car is switched between the sequential scanning displays I and II and between the sequential scanning displays III to V. Just as the car is stopped between, the object is stopped and displayed between the first frame (frame i) and the second frame (frame ii) from the top, and the third frame ( The object stops and is displayed between the frames iii) to 5 (frame v).

  In this way, the movement that was the same time interval in the original film frame, but in the sequential scanning display in the film mode has become an irregular time interval movement followed by a pattern of 2 frames: 3 frames, It is recognized as an unnatural motion that is not continuous. Also, compared to the case of displaying a normal video image signal, the original film movement interval is 24 frames / second, so that there is a problem that a smooth and jerky movement (so-called judder) is recognized. Exists.

  A frame rate conversion technique exists as a technique for displaying the telecine-converted video image signal as a smooth motion at regular intervals in the same manner as a normal video image. Interpolation frame generation by frame rate conversion will be specifically described with reference to FIG. FIG. 10 is a diagram schematically illustrating how an interpolation frame is generated by frame rate conversion.

  First, for each block that is divided into a small section between the reference frame n and the reference frame n + 1, a corresponding portion where the texture in the block most closely matches is found. Then, the motion vector v of the corresponding portion between the reference frame n and the reference frame n + 1 is calculated. Next, the vector length of the motion vector v is reduced by multiplying the motion vector v calculated between the reference frames by a predetermined conversion ratio to calculate the interpolation vector v ′. Finally, the reference frame image is actually moved based on the interpolation vector v ′. A group of images moved in this way becomes an image of an interpolation frame. FIG. 10 is a diagram showing a case where an interpolated frame at an intermediate time (1.5 msec) between the reference frame n (1 msec) and the reference frame n + 1 (2 msec) is generated. An interpolation frame at an arbitrary time between the reference frame n + 1 can be generated.

FIG. 11 is a diagram schematically showing the display of the video signal when the film rate originally converted from 24 Hz is converted to a frame rate of 120 Hz which is 5 times. A frame i-0 and a frame iii-0 in FIG. 11 are reference frames existing in the original film image. That is, four interpolation frames (frames i-1 to i-4) are interpolated between the frame i-0 that is the reference frame n and the frame iii-0 that is the reference frame n + 1. Each interpolation frame is obtained by multiplying the vector length of a motion vector between the reference frame n and the reference frame n + 1 by a conversion ratio determined for each interpolation frame to be interpolated (so-called Interpolated vector). As shown in FIG. 11, when four interpolation frames are interpolated, the conversion ratios are 1/5 times, 2/5 times, 3/5 times, and 4/5 times, respectively. By generating the interpolation frame in this way, it is possible to display smooth motion at regular intervals. The same applies to the frames vi-0 and viii-0 shown in FIG.
JP 2005-269164 A (published September 29, 2005) JP 2005-354528 A (published on December 22, 2005)

  Since the amount of processing calculation for detecting a motion vector is enormous, in general, detection of a motion vector is performed by limiting a search range within a certain range centered on a stationary position. Therefore, when the actual corresponding location is outside the motion vector search block, such as when the camera is panned at high speed, the location with the most similar texture in the motion vector search block is detected as the corresponding location. Thus, an incorrect motion vector is detected.

  If the video is continuously interpolated with the interpolation frame created based on the wrong motion vector, the image quality of the display video is naturally deteriorated. This is particularly noticeable when the movement of an object occupies a large area in the entire frame or in the frame.

  For this reason, an apparatus for detecting a motion vector generally includes an error detection unit that evaluates the correctness of the detected motion vector. As an example of the error detection means, a value obtained by accumulating reliability information at the time of detecting a motion vector over the entire frame is detected, and when the detected accumulated value becomes smaller than a certain threshold, frame rate conversion is set to “OFF”. . That is, the frame rate conversion is stopped (see FIG. 12). As described above, when an incorrect motion vector is detected, the error detection unit stops the frame rate conversion, so that the apparatus can avoid deterioration of image quality due to an interpolation error. However, if the frame rate conversion is stopped at the moment when the motion vector cannot be detected, the amount of judder generated increases rapidly, resulting in an unnatural video display.

  An example is shown in FIG. FIG. 13 is a schematic diagram schematically showing video display when frame rate conversion is stopped halfway. In FIG. 13, from 0 seconds to 5/60 seconds from the start of video display, the motion vector between frames i and iii is detected, and images of four interpolation frames are generated based on the detected motion vectors. is doing. That is, during this period, video can be displayed at a frame rate of 120 Hz. However, since the frame rate conversion is stopped between 5/60 seconds and 6/60 seconds from the start of video display, video display is performed at a normal frame rate of 24 Hz after 5/60 seconds from the start of video display. . For this reason, the amount of judder generated rapidly increases between 5/60 seconds and 6/60 seconds from the start of video display, and the image quality of the video display deteriorates rapidly.

  Patent Document 1 discloses a technique that enables highly accurate vector detection while suppressing a calculation amount and a circuit scale by appropriately determining a motion vector search range based on an analysis of motion vector statistics already obtained. Is disclosed. Further, in Patent Document 2, instead of setting a search range centering on a stationary position, a global vector representing the motion between the current image and the reference image is obtained, and the search range in the reference image is set using the global vector. Thus, a technique is disclosed that enables detection of a motion vector even in a moving image with large motion.

  However, since the techniques disclosed in Patent Documents 1 and 2 are methods for expanding a motion vector search range on the premise of a predictive coding device, the frame rate conversion device suppresses deterioration in the quality of a displayed video image. Therefore, the search range cannot be expanded sufficiently. That is, even if the techniques described in Patent Documents 1 and 2 are used (even if a motion vector is detected using the entire frame as a search range), a motion vector may not be detected. As described above, the image quality of the display image is deteriorated. Therefore, in the frame rate conversion apparatus, it is necessary to solve the above problem from an approach different from Patent Documents 1 and 2.

  The present invention has been made in view of the above-mentioned problems, and its main purpose is to reduce the sudden increase in judder amount that occurs when the frame rate conversion process is stopped, and the image quality of the display video is drastically reduced. It is an object of the present invention to provide a frame rate conversion device that suppresses deterioration.

In the frame rate conversion device according to the present invention, in order to solve the above problems,
An interpolation frame is generated based on a motion vector starting from each block on the first frame and ending at any block on the second frame, and the generated interpolation frame is the first frame and the first frame. A frame rate conversion device for converting a frame rate of a moving image by interpolating between two frames,
Statistical analysis means for calculating the degree of dispersion of the motion vector and a representative value of the motion vector;
A vector obtained by multiplying the conversion ratio by which the motion vector is uniformly multiplied when the degree of dispersion is smaller than the first threshold and the representative value of the motion vector is larger than the second threshold. Conversion ratio setting means for setting so that all of the values are equal to or less than the second threshold value;
Interpolation vector calculation means for calculating an interpolation vector starting from each block on the first frame by multiplying the set conversion ratio by the motion vector;
Interpolation frame generation means for generating the interpolation frame based on the calculated interpolation vector;
It is characterized by having.

  In the frame rate conversion device according to the present invention, when the degree of dispersion of motion vectors is smaller than the first threshold value and the representative value of the motion vector is larger than the second threshold value, the conversion ratio for uniformly multiplying the motion vector is calculated. , The size of the vector after multiplying by the conversion ratio is set to be equal to or less than the second threshold value.

  As described above, in the frame rate conversion apparatus according to the present invention, if a motion vector between reference frames can be detected, an interpolation vector used to generate an interpolation frame is calculated based on the detected vector length of the motion vector. The conversion ratio can be set. In other words, if the frame rate conversion apparatus according to the present invention can detect a motion vector between reference frames, it can generate an interpolation frame based on the interpolation vector calculated from the set conversion ratio. Therefore, even if the judder amount generated by the frame rate conversion process is suddenly increased in the conventional device, the frame rate conversion device according to the present invention reduces the increase in the judder amount. be able to.

  As a result, the frame rate conversion apparatus according to the present invention has an effect that it is possible to suppress a rapid deterioration of the image quality of the displayed video due to a sudden increase in the amount of judder generated.

  Further, in the frame rate conversion device according to the present invention, when the motion vector search range is larger than the interpolation image generation range, even if the conventional device cannot interpolate the interpolation frame, Interpolated frames can be generated. As a result, the effect of suppressing the deterioration of the image quality of the displayed image can be achieved.

  In addition, as the representative value of the motion vector, an average value of the motion vector, a mode value of the motion vector, or the like can be used.

  In the frame rate conversion apparatus according to the present invention, it is preferable that the conversion ratio setting means sets the value of the conversion ratio according to the magnitude of the representative value of the motion vector.

  According to said structure, since the conversion ratio can be set according to the magnitude | size of the representative value of a motion vector, the magnitude | size of the produced judder amount can be enlarged gradually. That is, the image quality of the displayed video can be gradually deteriorated.

  As a result, the frame rate conversion device according to the present invention has an effect of further suppressing the rapid deterioration of the image quality of the displayed video due to the sudden increase in the amount of judder generated.

  In the frame rate conversion device according to the present invention, the conversion ratio setting means further includes a magnitude of the motion vector with respect to a difference value between the second threshold value and a critical value indicating a vector length at which the conversion ratio is zero. It is preferable to set the conversion ratio based on the ratio of the difference value between the representative value and the critical value.

  According to said structure, the conversion ratio set according to vector length can be set so that it may reduce linearly. That is, since the image quality of the display image can be gradually deteriorated, the frame rate conversion device according to the present invention can further suppress the rapid deterioration of the image quality of the display image due to the sudden increase in the amount of judder generated. There is an effect.

  In the frame rate conversion apparatus according to the present invention, it is preferable that the critical value is equal to a limit value of the vector length that can be detected.

  According to the above configuration, the conversion ratio can be set to “0” in the limit value of the vector length that can be detected.

  Therefore, in the frame rate conversion apparatus according to the present invention, it becomes impossible to detect a motion vector, and even if the frame rate conversion process is stopped, the amount of judder generated does not increase. Therefore, it is possible to prevent the image quality of the display image from being rapidly deteriorated due to a sudden increase in the amount of judder generated.

In the frame rate conversion method according to the present invention, in order to solve the above problem,
An interpolation frame is generated based on a motion vector starting from each block on the first frame and ending at any block on the second frame, and the generated interpolation frame is the first frame and the first frame. A frame rate conversion method for converting a frame rate of a moving image by interpolating between two frames,
A statistical analysis step of calculating a degree of dispersion of the motion vector and a representative value of the motion vector;
A vector obtained by multiplying the conversion ratio by which the motion vector is uniformly multiplied when the degree of dispersion is smaller than the first threshold and the representative value of the motion vector is larger than the second threshold. A conversion ratio setting step for setting all the values to be equal to or less than the second threshold value;
Interpolation vector calculation means for calculating an interpolation vector starting from each block on the first frame by multiplying the motion vector by the set second conversion ratio;
An interpolation frame generation step for generating the interpolation frame based on the calculated interpolation vector;
It is characterized by including.

  According to said structure, there exists an effect similar to the frame rate conversion apparatus based on this invention.

  A television receiver provided with the frame rate conversion device according to the present invention is also included in the scope of the present invention.

  Furthermore, a program for operating the frame rate conversion apparatus according to the present invention, which is characterized in that the computer is driven as each of the above-mentioned means and a computer-readable recording medium on which the program is recorded. Included in the category.

  As described above, the frame rate conversion apparatus according to the present invention is uniform to a motion vector when the degree of dispersion of the motion vector is smaller than the first threshold value and the representative value of the motion vector is larger than the second threshold value. Is set so that the magnitudes of the vectors after multiplying by the conversion ratio are all equal to or less than the second threshold value.

  Therefore, even if the amount of judder caused by the frame rate conversion process is suddenly increased in the conventional apparatus, the frame rate conversion process is canceled by the frame rate conversion apparatus according to the present invention. The increase in the amount of judder that occurs when the process is performed can be reduced. As a result, the frame rate conversion apparatus according to the present invention has an effect that it is possible to suppress a rapid deterioration of the image quality of the displayed video due to a sudden increase in the amount of judder generated.

  Further, in the frame rate conversion device according to the present invention, when the motion vector search range is larger than the interpolation image generation range, even if the interpolation frame cannot be interpolated by the conventional device, Since the interpolation frame can be generated, the effect of suppressing the deterioration of the image quality of the front video image can be achieved as compared with the conventional case.

  An embodiment of a frame rate conversion apparatus according to the present invention will be described below with reference to FIGS.

(Configuration of Frame Rate Conversion Device 1)
The configuration of the frame rate conversion apparatus according to the present invention will be described below with reference to FIG. FIG. 1 is a block diagram showing the configuration of the frame rate conversion apparatus 1.

  The frame rate conversion device 1 converts the frame rate of the input video signal and outputs it. For example, the input 24 Hz film source video is output 5 times 120 Hz. In this case, the frame rate conversion apparatus 1 interpolates four interpolation frames between the reference frame n and the reference frame n + 1. In the present embodiment, four interpolation frames are provided between a reference frame n, which is one of a plurality of frames constituting a moving image, and a reference frame n + 1, which is a frame having a motion next to the reference frame n. A case of interpolating will be described as an example.

  As shown in FIG. 1, the frame rate conversion apparatus 1 includes an image information extraction unit 10, a motion vector search unit 11, a motion vector statistical analysis unit 12, a film mode detection unit 13, an interpolation vector calculation unit 14, and an interpolation image generation unit 15. Frame memories 16 to 18 and line memories 19 to 21. Each member will be described below.

(Image information extraction unit 10)
The image information extraction unit 10 extracts image feature information necessary for detecting the motion of the object from the input video signal. Image feature information is the main information of an image. More specifically, it is a luminance signal divided into color difference signals, coefficient information obtained by frequency decomposition, or outline information of an object in an image. Which of these pieces of information is extracted as the image feature information may be appropriately selected according to the motion vector detection method in the frame rate conversion apparatus 1.

(Motion vector search unit 11)
The motion vector search unit 11 detects a pixel set constituting an arbitrary motion vector search block among the motion vector search blocks obtained by dividing the reference frame n within the search range of the reference frame n + 1 corresponding to the motion vector search block. And make associations. That is, a value indicating how much the block (pixel set) of the motion vector search block in the reference frame n has moved in the reference frame n + 1 is detected. In other words, the motion vector search unit 11 detects a motion vector between the reference frame n and the reference frame n + 1.

  The motion vector search unit 11 continuously shifts the target motion vector search block from left to right and continuously from top to bottom, that is, searches for all motion vectors in the reference frame n. Similarly, by detecting the motion vector in the block, the motion vector for one frame between the reference frame n and the reference frame n + 1 is detected.

(Motion vector statistical analysis unit 12)
The motion vector statistical analysis unit 12 statistically processes the motion vector in the reference frame n detected by the motion vector search unit 11. The motion vector statistical processing in the motion vector statistical analysis unit 12 will be specifically described below.

  The motion vector statistical analysis unit 12 calculates the variance of the detected motion vector and calculates a representative value of the motion vector.

  Note that “dispersion” in this specification and the like refers to an average value of squares of deviations of detected motion vectors. In other words, “dispersion” in the present specification and the like is a degree of dispersion representing how much the distribution of detected motion vectors varies. The representative value may be an average value of motion amounts, or a motion amount histogram (frequency of motion vectors having the same value) is taken, and the value of the motion vector having the highest frequency is set as the representative value. It may be.

  Here, the large variance of the motion vector indicates that the object shown in the image on the reference frame n is moving at various speeds. That is, the image on the reference frame n means that no object occupies the whole or most of the image. On the other hand, that the variance of the motion vector is small means that the object shown in the image on the reference frame n is moving at a substantially uniform speed. That is, it means that an object occupying the whole or most of the image exists in the image on the reference frame n. Note that a large motion vector means that an object is moving at high speed between frames. In the present specification and the like, the magnitude of the motion vector is hereinafter referred to as “motion amount”.

  In this embodiment, the distribution of the motion vectors (the degree of dispersion) is calculated by calculating the variance, but the present invention is not limited to the variance as long as the value can calculate the degree of dispersion of the motion vectors. For example, other values such as a standard deviation calculated by the square root of the variance may be used.

(Film mode detection unit 13)
The film mode detection unit 13 determines whether or not two consecutive frames are the same image by capturing and comparing pixel values of the same coordinates of the two consecutive frames for one frame. The film mode detection unit 13 determines a frame to be used as a reference frame when creating an interpolation frame based on the determination result. That is, a frame determined as a frame that is not the same image as the reference frame n is used as the reference frame n + 1.

  At the same time, the film mode detection unit 13 supplies the interpolation vector calculation unit 14 with the motion vector conversion ratio R1 used when calculating the interpolation vector used to create the interpolation frame from the motion vector detected by the interpolation vector calculation unit 14. Output.

  The conversion ratio R1 output from the film mode detection unit 13 is a value set in advance according to the number of interpolation frames to be interpolated. That is, when four interpolation frames are interpolated between the reference frame n and the reference frame n + 1, the motion vector conversion ratios for generating the respective interpolation frames are 1/5, 2/5, 3 / 5, 4/5.

  The determination in the film mode detection unit 13 is made every time the frame recorded in the frame memory 18 is changed.

(Interpolation vector calculation unit 14)
The interpolation vector calculation unit 14 calculates an interpolation vector for generating an interpolation frame from the motion vector detected by the motion vector search unit 11. The calculation of the interpolation vector in the interpolation vector calculation unit 14 will be specifically described.

  First, the interpolation vector calculation unit 14 converts the conversion ratio R1 into the conversion ratio R2 when the variance calculated by the motion vector statistical analysis unit 12 is smaller than the threshold value Th1 and the representative value is larger than the threshold value Th2. Note that details of the conversion from the conversion ratio R1 to the conversion ratio R2 will be described in detail below, and the description thereof is omitted here.

  Subsequently, the interpolation vector calculation unit 14 calculates an interpolation vector corresponding to each interpolation frame to be interpolated by multiplying the motion vector detected by the motion vector search unit 11 by the conversion ratio R1 or the conversion ratio R2.

  Here, the variance in this specification and the like being smaller than the threshold Th1 means that an object occupying the whole or most of the image in the image on the reference frame n exists. The value of “threshold Th1” is not particularly limited, and is an arbitrary value set in advance in the frame rate conversion apparatus 1. The “threshold value Th1” may be a value fixed in the frame rate conversion apparatus 1 or a value appropriately set by the user.

  Also, the “threshold Th2” is also an arbitrary value set in advance in the frame rate conversion apparatus 1 like the threshold Th1, and may be a fixed value in the frame rate conversion apparatus 1, or may be set appropriately by the user. It may be a value to be set. A large threshold Th2 means that the interpolation image generation range when generating the interpolation image is large. That is, as the threshold Th2 increases, it is necessary to record or hold a wider range of images centered on the block position to be processed in the line memory 21 in the frame rate conversion apparatus 1. Therefore, the value of the threshold Th2 is preferably set to a value according to the circuit scale of the line memory 21 of the frame rate conversion device 1. Note that the value of the threshold Th2 may be “0”.

(Interpolation frame generation unit 15)
The interpolation frame generation unit 15 creates an interpolation image from the image of the reference frame n held in the line memory 21 based on the interpolation vector calculated by the interpolation vector calculation unit 14, thereby interpolating between the reference frames. Generate a frame. The interpolated frame generation unit 15 continuously shifts the interpolated image generation block from the left to the right and continuously shifts from the top to the bottom, that is, in the same way for all the interpolated image generation blocks in the interpolated frame. By generating an image, an interpolated image for one frame is generated from the reference frame n, and an interpolated frame to be interpolated is generated.

  Further, the size of the motion vector search range and the size of the interpolated image generation range may be the same size or different sizes. However, if the size of the motion vector search range is different from the size of the interpolated image generation range, a motion vector is required to generate the interpolated image. It is preferable that the vector search range is larger.

(Frame memory 16-18)
The frame memory 16 records and holds image feature information for at least two frames in order to perform motion vector search. The frame memory 18 records and holds the input video signal as it is (or in a reproducible format). That is, all frames constituting the video are recorded. The frame memory 17 records and holds the detected motion vector for one frame.

(Line memory 19-21)
The line memory 19 stores image feature information of an image within a motion vector search range centered on a motion vector search block to be processed, which is required when the motion vector search unit 11 searches for a motion vector. 16 is read and held. The line memory 20 reads and holds a motion vector necessary for generating an image of an interpolation image generation block to be processed among motion vectors for one frame. Further, the line memory 21 reads out and holds, from the frame memory 17, an image within the interpolation image generation range centered on the interpolation image generation block to be processed, which is necessary when generating the interpolation image.

  When the motion vector search range is made larger in the frame rate conversion device 1 than in the conventional frame rate conversion device, the line memory 19 needs to have a larger circuit scale than the line memory mounted in the conventional device. However, the image feature information extracted by the image information extraction unit 10 is sufficiently smaller than the information of the input video signal, and the circuit scale of the line memory 19 may be increased depending on the search method in the motion vector search unit 11. It can also be made extremely small.

  Further, information handled in the line memory 21 is information that does not allow deterioration of image quality, and is generally a large amount of information of 8 to 12 bits each of FullHD (1920 × 1080) RGB or YPbPr. Therefore, the line memory 21 needs to have a larger circuit scale than the other line memories.

(Operation of Frame Rate Conversion Device 1)
Next, the operation of the frame rate conversion apparatus 1 will be described below with reference to FIG. FIG. 2 is a flowchart showing the operation of the frame rate conversion apparatus 1.

  When the video signal is input, the image information extraction unit 10 extracts the image feature information and records it in the frame memory 16 (step S1). The image information extraction unit 10 extracts image feature information for at least two frames (here, reference frame n and reference frame n + 1 as an example) and records them in the frame memory 16 (step S2).

  Subsequently, the frame rate conversion apparatus 1 reads the image feature information of the reference frame n and the reference frame n + 1 in an arbitrary motion vector search block from the extracted image feature information, and records it in the line memory 19 ( Step S3). The motion vector search unit 11 reads the image feature information from the line memory 19, and selects a pixel set that most closely matches the pixel set constituting the motion vector search block of the reference frame n from the motion vector search range of the reference frame n + 1. To detect. That is, the motion vector search unit 11 detects a motion vector in the read motion vector search block (step S4). Similarly, the motion vector search unit 11 detects the motion vectors of all the motion vector search blocks in the reference frame n. That is, the motion vector search unit 11 detects a motion vector for one frame between the reference frame n and the reference frame n + 1. The motion vector search unit 11 records the detected motion vector for one frame in the frame memory 17 (step S5).

  The motion vector statistical analysis unit 12 calculates a representative value of the motion vector per unit frame of the image on each motion vector search block based on the motion vector recorded in the frame memory 17 (step S6). That is, it is calculated how fast the object is moving between the reference frame n and the reference frame n + 1. Subsequently, the motion vector statistical analysis unit 12 calculates the variance of the detected motion vector (step S7).

  The interpolation vector calculation unit 14 determines whether or not the variance calculated in step S7 is less than the threshold Th1 (step S8). That is, the interpolation vector calculation unit 14 determines whether there is an object that occupies all or most of the image in the image on the reference frame n. When the variance is less than the threshold Th1 (Yes in Step S8), the interpolation vector calculation unit 14 determines whether or not the representative value of the motion vector is larger than the threshold Th2 (Step S9). That is, it is determined whether or not the speed of an object occupying all or most of the image is greater than a threshold value. When the representative value of the motion vector is larger than the threshold value Th2 (Yes in step S9), the interpolation vector calculation unit 14 calculates the amount of motion calculated from the interpolation vector from the predetermined conversion ratio R1 input from the film mode detection unit 13. Is set such that the conversion ratio R2 becomes smaller than the threshold Th2 (step S10).

  Subsequently, the interpolation vector calculation unit 14 calculates an interpolation vector by multiplying the motion vector acquired from the line memory 20 by the conversion ratio R2 set in step S10 (step S11). On the other hand, when the representative value is equal to or smaller than the threshold value Th2 (No in step S9), the interpolation vector calculation unit 14 performs interpolation by multiplying the motion vector acquired from the line memory 20 by the conversion ratio R1 acquired from the film mode detection unit 13. A vector is calculated (step S11). That is, normal frame rate conversion processing is executed.

  If the variance is equal to or greater than the threshold Th1 (No in step S8), the interpolation vector calculation unit 14 determines whether the representative value is greater than the threshold Th2 (step S13). When the representative value is equal to or less than the threshold Th2 (No in step S13), an interpolation vector is calculated by multiplying the motion vector acquired from the line memory 20 by the conversion ratio R1 acquired from the film mode detection unit 13 (step S11). . If the representative value is larger than the threshold value Th2 (No in step S13), the frame rate conversion apparatus 1 ends the process without generating an interpolation frame. In this case, the frame rate conversion process is turned off.

  Next, the interpolation frame generation unit 15 reads out from the reference frame n recorded in the frame memory 18 based on the calculated interpolation vector, and records the image (that is, the processing target) recorded in the line memory 21. An interpolation image of each interpolation image generation block is generated from an image within an interpolation image generation range centered on the interpolation image generation block. Then, an interpolation frame is generated by generating interpolation images of all the interpolation image generation blocks (step S12). Finally, the generated interpolation frame is interpolated between the reference frame n and the reference frame n + 1, and the video signal subjected to the frame rate conversion process is output from the frame rate conversion device 1.

  Note that the frame rate conversion apparatus 1 generates an interpolation frame based on the number of interpolation frames to be interpolated. That is, when four interpolation frames are interpolated between the reference frame n and the reference frame n + 1, four interpolation frames are generated and output in step S12.

(Increase in judder amount due to increase in movement)
FIG. 3 shows an increase in the judder amount accompanying an increase in the amount of motion in the frame rate conversion device 1, the conventional frame rate conversion device, and the case without the frame rate conversion device. FIG. 3 is a diagram illustrating an increase in judder amount with respect to an increase in motion amount. In FIG. 3, an increase in the judder amount with respect to the motion amount without the frame rate conversion device is indicated by a broken line, and an increase in the judder amount with respect to the motion amount in the conventional frame rate conversion device is indicated by a one-dot chain line. The increase in judder amount relative to the amount of movement is indicated by a solid line. Moreover, about the location where a mutual line overlaps, each line is shifted for convenience and displayed so that a mutual line may not overlap.

  As shown in FIG. 3, in the case of no frame rate conversion device, the amount of judder generated according to the amount of increase in motion increases. That is, as the moving speed of the object displayed on the frame increases, the amount of judder generated increases. Further, since the conventional frame rate conversion apparatus can interpolate interpolation frames up to the threshold Th2, almost no judder occurs (see FIG. 11). However, when the threshold value Th2 is exceeded, the interpolation frame interpolation process is stopped, and the amount of judder generated increases rapidly (see FIG. 13). Therefore, the user who is viewing the moving image feels that the image quality of the display image has deteriorated rapidly.

  On the other hand, unlike the conventional frame rate conversion apparatus, the frame rate conversion apparatus 1 does not stop the interpolation frame interpolation process even when the threshold Th2 is exceeded. When the representative value of the motion vector calculated based on the detected motion vector exceeds the threshold Th2, the frame rate conversion apparatus 1 determines that the amount of motion between the reference frame n and the interpolation frame to be interpolated is the threshold Th2. A conversion ratio R2 is set as follows. As a result, as shown in FIG. 3, even if the representative value exceeds the threshold value Th2, it is possible to suppress an abrupt increase in the amount of judder that occurs. That is, it is possible to suppress a rapid deterioration in the image quality of the display image.

  However, even in the frame rate conversion apparatus 1, the conversion ratio R2 cannot be set when the motion vector cannot be calculated. That is, when the limit value LM is exceeded, the interpolation frame interpolation process is stopped. Therefore, it is preferable that the frame rate conversion apparatus 1 interpolates the interpolation frame so that the judder amount gradually increases until the limit value LM is reached.

  Here, the value of “limit value LM” in this specification and the like means a limit value at which a motion vector can be detected. The limit value LM may be a fixed value in the frame rate conversion apparatus 1 or may be a value set as appropriate by the user, similarly to the threshold value Th2. The increase in the limit value LM means that even a larger motion vector can be detected. However, as the limit value LM increases, it is necessary to increase the scale of the line memory 19, the motion vector search unit 11, and the line memory 20 in the frame rate conversion apparatus 1. Therefore, it is preferable that the limit value LM is set according to the circuit scale of the frame rate conversion device 1 in the same manner as the threshold value Th2. Note that the value of the limit value LM is larger than the value of the threshold value Th2.

  In addition, “judder” in this specification and the like means motion blur in a moving image caused by an interpolation frame not being inserted at a predetermined position. That is, it is a so-called jerky movement of the object that is not smooth. The size of such judder is determined by the movement of the object in the moving image with respect to the interval between frames. That is, the larger the interval between frames and the faster the movement of the object, the larger the judder amount.

(Advantages of Frame Rate Conversion Device 1)
As described above, in the frame rate conversion apparatus 1, as long as the motion vector between the reference frame n and the reference frame n + 1 can be detected, the interpolation vector used for generating the interpolation frame is determined based on the detected motion vector. A conversion ratio R2 for calculation can be set. That is, the frame rate conversion apparatus 1 can generate an interpolation frame based on the interpolation vector calculated from the set conversion ratio R2 as long as it can detect the motion vector between the reference frame n and the reference frame n + 1. .

  Therefore, even if the amount of judder generated when the frame rate conversion process is stopped increases rapidly in the conventional apparatus, the frame rate conversion apparatus 1 stops the frame rate conversion process. The increase in the amount of judder occurring can be reduced. As a result, the frame rate conversion apparatus 1 can suppress the rapid deterioration of the image quality of the display image due to the sudden increase in the amount of judder generated.

  Further, when the motion vector search range is made larger than the interpolation image generation range in the frame rate conversion apparatus 1, an interpolation frame is generated even when the interpolation frame cannot be interpolated with the conventional apparatus. You can also.

(Details of processing in the film mode detection unit 13)
Next, details of processing in the film mode detection unit 13 will be described. Here, a description will be given with reference to FIG. The film mode detection unit 13 accumulates the difference between the pixel values of the same coordinates in two consecutive frames stored in the frame memory 18 over one frame. By comparing the difference value of the accumulated pixel values with a predetermined threshold value, it is determined whether or not the entire frame is an image having a sufficiently small difference, that is, whether or not two frames are the same image. That is, the film mode detection unit 13 determines that the frame I and the frame II are the same image, and that the frame II and the frame III are different images. It is determined that the frames III and IV and the frames IV and V are the same image. Therefore, the film mode detection unit 13 determines that the frame I is the reference frame n and the frame III is the reference frame n + 1.

  At the same time, the film mode detector 13 interpolates frames i-1 to i-4 as interpolated frames between the frame I (reference frame n) and the frame III (reference frame n + 1), which are original frames before interpolation. And the conversion ratio R1 for generating each of the frames i-1 to i-4 is set. That is, as shown in FIG. 4, when generating the frame i-1, the conversion ratio R1 is 1/5, when generating the frame i-2, the conversion ratio R1 is 2/5, and the frame i- When 3 is generated, 3/5 is set as the conversion ratio R1, and when the frame i-4 is generated, 4/5 is set as the conversion ratio R1.

  When the frame i−1 is generated by the interpolation frame generation unit 15, when the frame i−1 is generated by the interpolation vector calculation unit 14, the film mode detection unit 13 sends the frame i−1 to the interpolation vector calculation unit 14. Information indicating that the conversion ratio R1 at -1 is 1/5 is output. Accordingly, the interpolation vector calculation unit 14 can calculate the interpolation vector based on the conversion ratio R1 or the conversion ratio R2 calculated from the conversion ratio R1. The same applies to the interpolation frames other than the frame i-1 and the interpolation frames iii-1 to iii-4 between the frame III and the next reference frame after the frame III. Note that since frame i-0 is the same frame as frame I and frame iii-0 is the same frame as frame III, it is not strictly an interpolation frame, but in FIG. Yes.

  Here, FIG. 4 is a diagram showing the frame in telecine conversion, each interpolation frame, and the vector length conversion ratio R1 in each interpolation frame in association with each other. The vector length conversion ratio R1 shown in FIG. 4 is the conversion ratio for the motion vector between the frame I and the frame III in the frames i-0 to i-4, and the frame III in the frames iii-0 to iii-4. And the conversion ratio for the motion vector between the frame III and the next reference frame.

  In FIG. 4, the frames i-3 and i-4 have the conversion ratio R1 set to 3/5 and 4/5, respectively, based on the frame I (that is, the reference frame n). The conversion ratio R1 may be set to -2/5 and -1/5 with reference frame (n + 1) as a reference.

(Detailed description of conversion ratio R2)
Next, details of a method for calculating the conversion ratio R2 will be described below with reference to FIG. FIG. 5 is a diagram illustrating a change in the value of the conversion ratio accompanying an increase in the amount of motion in the frame rate conversion device 1 and the conventional frame rate conversion device.

  In FIG. 5, a change in the value of the conversion ratio with respect to the motion amount in the conventional frame rate conversion apparatus is indicated by a one-dot chain line, and a change in the value of the conversion ratio with respect to the motion amount in the frame rate conversion apparatus 1 is indicated by a solid line. Moreover, about the location where a mutual line overlaps, each line is shifted for convenience and displayed so that a mutual line may not overlap.

  In FIG. 5, the conversion ratio (that is, the y-axis) when the representative value of the motion vector is equal to or less than the threshold Th2 indicates the conversion ratio R1, and the representative value of the motion vector is greater than the threshold Th2 and equal to or less than the limit value LM. The conversion ratio (that is, the y-axis) in this case indicates the conversion ratio R2. FIG. 5 shows a case where the limit value LM and the critical value that is the amount of motion at which the conversion ratio is 0 (that is, the value on the x-axis where y = 0) are the same value. The critical value is a value set in advance in the frame rate conversion apparatus 1. The range of the critical value is not particularly limited as long as it is a value larger than the threshold value Th2 similarly to the limit value LM. The value may be a fixed value in the frame rate conversion apparatus 1 or may be a value appropriately set by the user.

  As shown in FIG. 5, in the conventional frame rate conversion apparatus, the representative value of the motion vector exceeds a threshold value Th2, and the conversion ratio becomes zero. That is, in the conventional frame rate conversion apparatus, the interpolation value of the interpolation frame is stopped when the representative value of the motion vector exceeds the threshold value Th2. Therefore, the user who is viewing the moving image feels that the image quality of the displayed video has deteriorated rapidly.

  On the other hand, in the frame rate conversion device 1, unlike the conventional frame rate conversion device, even if the representative value of the motion vector exceeds the threshold Th2, the conversion ratio gradually increases until the motion amount reaches the limit value LM. Decrease. As a result, even if the representative value of the motion vector exceeds the threshold Th2, the amount of judder generated gradually increases, so that rapid deterioration of the image quality of the display video can be suppressed.

  In addition, as shown in FIG. 5, it is preferable that the limit value LM and the critical value are the same value. That is, the conversion ratio is preferably 0 at the limit value LM. Thereby, even if the interpolation processing of the interpolation frame is stopped due to the motion amount exceeding the limit value LM, the judder amount hardly increases. Therefore, it is possible to almost completely prevent the rapid deterioration of the image quality of the displayed image.

  Of course, the limit value LM and the critical value may not be the same value, but when the interpolation process of the interpolation frame is stopped due to the movement amount exceeding the limit value LM, the limit value LM and the critical value are The increase in judder amount is larger than that of the same value. That is, there is a high possibility that the user who views the moving image will feel uncomfortable with the deterioration of the image quality of the displayed video.

(Details of calculation method of conversion ratio R2)
As an example of an expression for calculating the conversion ratio R2, the following expression (1) can be given.
y = −b / (x2−x1) × (x−x2) (1)
In equation (1), y represents the conversion ratio R2, x represents the calculated amount of motion, x1 represents the threshold Th2, x2 represents the critical value LM, and b represents the conversion ratio R1.

  As shown in the above formula (1), the conversion ratio R2 is calculated by multiplying the conversion ratio R1 by the ratio of the difference value between the motion amount and the critical value with respect to the difference value between the threshold Th2 and the critical value. That is, the value of the conversion ratio R2 decreases linearly from the conversion ratio R1 as the amount of motion increases.

  FIG. 6 shows a change in the length of the vector length of the interpolation vector accompanying an increase in the amount of motion in the frame rate conversion apparatus 1 and the conventional frame rate conversion apparatus. Note that the frame rate conversion apparatus 1 exemplifies a case where the conversion ratio R2 is calculated using the above formula (1). FIG. 6 shows a case where the limit value LM and the critical value are the same value. In FIG. 6, the vector length when the representative value of the motion vector is equal to or smaller than the threshold Th2 is the vector length of the interpolation vector calculated using the conversion ratio R1, and the vector length increases as the motion amount increases. .

  As shown in FIG. 6, in the conventional frame rate conversion apparatus, the representative value of the motion vector exceeds the threshold value Th2, and the vector length of the interpolation vector becomes zero. That is, in the conventional frame rate conversion apparatus, the conversion ratio becomes 0 at the same time when the representative value of the motion vector exceeds the threshold Th2, and the interpolation processing of the interpolation frame is stopped. Therefore, the user who is viewing the moving image feels that the image quality of the displayed video has deteriorated rapidly.

  On the other hand, in the frame rate conversion apparatus 1, the vector length of the interpolation vector does not become zero at the same time as the representative value of the motion vector exceeds the threshold Th2. In the frame rate conversion device 1, even if the representative value of the motion vector exceeds the threshold Th2, if the motion amount is equal to or less than the limit value LM, the frame rate conversion device 1 is based on the conversion ratio R2 calculated according to the above equation (1). To calculate an interpolation vector. That is, in FIG. 6, the vector length of the motion vector representative value larger than the threshold value Th2 and equal to or less than the limit value LM is the vector length of the interpolation vector calculated using the conversion ratio R2.

  Therefore, as shown in FIG. 6, the vector length of the interpolation vector whose representative value of the motion vector is larger than the threshold value Th2 and equal to or less than the limit value LM is the motion amount and the limit value LM with respect to the difference value between the threshold value Th2 and the limit value LM. And becomes “0” in the limit value LM. That is, the vector length value of the interpolation vector decreases in a quadratic curve so as to be “0” at the limit value LM.

The interpolation vector length can be calculated by multiplying the above equation (1) for calculating the conversion ratio R2 by the amount of motion. More specifically, the interpolation vector length is expressed by the following equation (2).
y = −b / (x2−x1) × (x−x2) × x (2)
Can be used to calculate.

The above formula (2) is
y = -b / (x2-x1) * {(x-x2 / 2) 2- (x2 / 2) 2}
When x = x2 / 2, it becomes a vertex. Further, the interpolation vector length needs to satisfy x2 / 2 <threshold Th2, since it is necessary to maximize the length at the threshold Th2. That is, the limit value LM needs to be set to not more than twice the threshold Th2.

  As a result, the frame rate conversion apparatus 1 can gradually shorten the vector length even when the representative value of the motion vector exceeds the threshold Th2. That is, in the frame rate conversion device 1, the amount of judder that is generated gradually increases, so that it is possible to suppress a rapid deterioration in the image quality of the display video due to a sudden increase in the amount of judder that occurs. Further, when the vector length decreases so that the vector length becomes “0” at the limit value LM, even if the motion amount exceeds the limit value LM and the frame rate conversion process is stopped, the judder amount There is no increase. That is, when the frame rate conversion process is stopped, it is possible to prevent the image quality of the display image from being rapidly deteriorated. Here, in this specification and the like, “prevention” is used in the sense that it can be almost completely suppressed.

  In the present embodiment, the case where the conversion ratio R2 to the interpolation vector is linearly decreased has been described as an example, but the present invention is not limited to this. The conversion ratio R2 is not limited to the above equation (1) as long as the vector length can be shortened in accordance with the increase amount of the motion amount.

  In the present embodiment, the case where the critical value is equal to the limit value LM is described as an example. However, the present invention is not limited to this, and the critical value may be a value larger than the threshold Th2. For example, the value may be larger than the limit value LM or may be a smaller value. Even if the critical value is different from the limit value LM, it is possible to suppress the image quality of the display image from being rapidly deteriorated as compared with the conventional frame rate conversion device.

(Interpolation position of interpolation frame)
In the frame rate conversion apparatus 1, the interpolation frame generated using the conversion ratio R2 is shifted to the reference frame n side from the predetermined interpolation position by the time corresponding to the difference between the conversion ratio R1 and the conversion ratio R2. Inserted.

  The interpolation position of the interpolation frame generated using the conversion ratio R2 will be described below with a specific example with reference to FIG. FIG. 7 is a diagram showing a state in which the interpolation frame generated using the interpolation vector calculated by the conversion ratio R2 is interpolated in the frame rate conversion apparatus 1. FIG. 7 shows a case where one interpolation frame is inserted so that the present invention can be easily understood. FIG. 7 illustrates a case where the reference frame n is 1 msec, the reference frame n + 1 is 2 msec, the conversion ratio R1 is 1/2, and the conversion ratio R2 is 3/10.

  As shown in FIG. 7, when the conversion ratio R2 is set to 3/10, the generated interpolated frame is interpolated at a time of 1.3 msec, not 1.5 msec, which is the originally interpolated position. That is, the position where the generated interpolation frame is interpolated is shifted to the reference frame n side by 2/10 msec (0.2 msec).

  Actually, the interpolation ratio is not shifted because the conversion ratio R2 is set, but the frame rate conversion apparatus 1 sets a second conversion ratio that generates an interpolation frame at the shifted interpolation position. To do. The setting of the conversion ratio R2 can be paraphrased as the setting of the interpolation position.

(Additional notes)
Note that the frame rate conversion apparatus 1 is preferably provided in a television receiver. The configuration other than the frame rate conversion device 1 in the television receiver provided with the frame rate conversion device 1 may be the same as that of the related art.

(Program and recording medium)
Finally, each block included in the frame rate conversion apparatus 1 may be configured by hardware logic. Alternatively, it may be realized by software using a CPU (Central Processing Unit) as follows.

  That is, the frame rate conversion apparatus 1 includes a CPU that executes instructions of a program that realizes each function, a ROM (Read Only Memory) that stores the program, a RAM (Random Access Memory) that expands the program into an executable format, Also, a storage device (recording medium) such as a memory for storing the program and various data is provided. With this configuration, the object of the present invention can be achieved by a predetermined recording medium.

  This recording medium only needs to record the program code (execution format program, intermediate code program, source program) of the program of the frame rate conversion apparatus 1 which is software for realizing the above-described functions so as to be readable by a computer. This recording medium is supplied to the frame rate conversion apparatus 1. Thus, the frame rate conversion device 1 (or CPU or MPU) as a computer may read and execute the program code recorded on the supplied recording medium.

  The recording medium that supplies the program code to the frame rate conversion apparatus 1 is not limited to a specific structure or type. That is, the recording medium includes, for example, a tape system such as a magnetic tape and a cassette tape, a magnetic disk such as a floppy (registered trademark) disk / hard disk, and an optical disk such as a CD-ROM / MO / MD / DVD / CD-R. System, a card system such as an IC card (including a memory card) / optical card, or a semiconductor memory system such as a mask ROM / EPROM / EEPROM / flash ROM.

  Further, even if the frame rate conversion device 1 is configured to be connectable to a communication network, the object of the present invention can be achieved. In this case, the program code is supplied to the frame rate conversion apparatus 1 via the communication network. The communication network is not limited to a specific type or form as long as it can supply the program code to the frame rate conversion apparatus 1. For example, it may be the Internet, intranet, extranet, LAN, ISDN, VAN, CATV communication network, virtual private network, telephone line network, mobile communication network, satellite communication network, and the like.

  The transmission medium constituting the communication network may be any medium that can transmit the program code, and is not limited to a specific configuration or type. For example, wired communication such as IEEE 1394, USB (Universal Serial Bus), power line carrier, cable TV line, telephone line, ADSL (Asymmetric Digital Subscriber Line) line, infrared rays such as IrDA and remote control, Bluetooth (registered trademark), 802.11 It can also be used by radio such as radio, HDR, mobile phone network, satellite line, terrestrial digital network. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.

  The present invention has been specifically described above based on the embodiments. However, the present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims and are different. Embodiments obtained by appropriately combining technical means disclosed in the respective embodiments are also included in the technical scope of the present invention.

  The frame rate conversion device according to the present invention can be suitably used in a video display device such as a television receiver.

It is a block diagram which shows the principal part structure of the frame rate conversion apparatus which concerns on this invention. It is a flowchart which shows operation | movement of the frame rate conversion apparatus based on this invention. It is a figure which shows the increase in the judder amount with respect to the motion amount in the frame rate conversion device according to the present invention, the conventional frame rate conversion device, and the case without frame rate conversion. It is the figure in which the frame in telecine conversion, each interpolation frame, and vector length conversion ratio R1 in each interpolation frame were associated with each other. It is a figure which shows the change of the value of the conversion ratio accompanying the increase in the amount of motion of the frame rate conversion apparatus which concerns on this invention, and the conventional frame rate conversion apparatus. It is a figure which shows the change of the value of the vector length of the interpolation vector accompanying the increase in the amount of motion of the frame rate conversion apparatus which concerns on this invention, and the conventional frame rate conversion apparatus. In the frame rate conversion apparatus according to the present invention, it is a diagram schematically showing a state in which an interpolation frame generated using an interpolation vector calculated by a conversion ratio R2 is interpolated. It is a figure which shows typically the telecine conversion by 3: 2 pull-down system. It is the figure which showed typically the display of the video image signal by which the telecine conversion by 3: 2 pulldown system was carried out. It is the figure which showed typically the mode of the production | generation of the interpolation frame by frame rate conversion. It is the figure which showed typically the display of the video signal at the time of carrying out the frame rate conversion process of the film image which was 24 Hz to 120 Hz. It is the figure which showed typically the mode of the production | generation of the interpolation frame when a motion vector cannot be detected. It is a schematic diagram which shows typically a video display when a frame rate conversion process is stopped on the way.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Frame rate conversion apparatus 10 Image information extraction part 11 Motion vector search part 12 Motion vector statistical analysis part (statistical analysis means)
13 Film Mode Detection Unit 14 Interpolation Vector Calculation Unit (Interpolation Vector Calculation Unit, Conversion Ratio Setting Unit)
15 Interpolation frame generation unit (interpolation frame generation means)
16, 17, 18 Frame memory 19, 20, 21 Line memory

Claims (8)

An interpolation frame is generated based on a motion vector starting from each block on the first frame and ending at any block on the second frame, and the generated interpolation frame is the first frame and the first frame. A frame rate conversion device for converting a frame rate of a moving image by interpolating between two frames,
Statistical analysis means for calculating the degree of dispersion of the motion vector and a representative value of the motion vector;
A vector obtained by multiplying the conversion ratio by which the motion vector is uniformly multiplied when the degree of dispersion is smaller than the first threshold and the representative value of the motion vector is larger than the second threshold. Conversion ratio setting means for setting so that all of the values are equal to or less than the second threshold value;
Interpolation vector calculation means for calculating an interpolation vector starting from each block on the first frame by multiplying the set conversion ratio by the motion vector;
Interpolation frame generation means for generating the interpolation frame based on the calculated interpolation vector;
A frame rate conversion apparatus comprising:   2. The frame rate conversion apparatus according to claim 1, wherein the conversion ratio setting means sets the value of the conversion ratio according to the magnitude of the representative value of the motion vector.   The conversion ratio setting means is based on a ratio of a difference value between the representative value of the motion vector and the critical value with respect to a difference value between the second threshold and a critical value indicating a vector length at which the conversion ratio is zero. The frame rate conversion apparatus according to claim 2, wherein the conversion ratio is set.   4. The frame rate conversion apparatus according to claim 3, wherein the critical value is equal to a limit value of a vector length that can be detected. An interpolation frame is generated based on a motion vector starting from each block on the first frame and ending at any block on the second frame, and the generated interpolation frame is the first frame and the first frame. A frame rate conversion method for converting a frame rate of a moving image by interpolating between two frames,
A statistical analysis step of calculating a degree of dispersion of the motion vector and a representative value of the motion vector;
A vector obtained by multiplying the conversion ratio by which the motion vector is uniformly multiplied when the degree of dispersion is smaller than the first threshold and the representative value of the motion vector is larger than the second threshold. A conversion ratio setting step for setting all the values to be equal to or less than the second threshold value;
Interpolation vector calculation means for calculating an interpolation vector starting from each block on the first frame by multiplying the set conversion ratio by the motion vector;
An interpolation frame generation step for generating the interpolation frame based on the calculated interpolation vector;
A frame rate conversion method comprising:   A television receiver comprising the frame rate conversion device according to any one of claims 1 to 4.   A program for operating the frame rate conversion apparatus according to any one of claims 1 to 4, wherein the frame rate conversion program causes a computer to function as each of the above means.   A computer-readable recording medium in which the program according to claim 7 is recorded.

Images