JP2014022899A - Video signal processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a video signal processing device which discriminates a 2:3 pull-down video signal included in an NTSC system video signal efficiently with high accuracy.SOLUTION: In the video signal processing device, an inter-frame difference calculation unit 210 calculates, for each field with respect to an input video signal, an inter-frame difference value which is a difference between the video signal of the field concerned and a video signal two fields before, a threshold calculation unit 220 determines the minimum value among the inter-frame difference values for five consecutive fields and calculates a threshold from the moving average of minimum inter-frame difference values up to the current point of time and the moving average of non-minimum inter-frame difference values up to the current point of time, and a 2:3 pull-down determination unit 240 determines a video signal whose inter-frame difference value becomes minimum in a 5-field cycle and is equal to or less than the threshold as a 2:3 pull-down video signal. In this case, an I/P conversion unit 250 performs, on a video signal 3 to 5 fields before, I/P conversion suitable for the configuration of the 2:3 pull-down video signal.

Description

  The present invention relates to a video signal processing apparatus for identifying a 2: 3 pull-down video signal included in an NTSC (National Television System Committee) video signal.

  A video signal reproduced from a movie film may be included in a standard television system video signal such as the NTSC system. Here, a video signal of 24 frames per second is reproduced from the movie film. On the other hand, the NTSC video signal is a video signal of 60 fields per second (30 frames per second). The NTSC video signal is an interlace video signal in which an odd field video signal corresponding to an odd horizontal scan line and an even field video signal corresponding to an even horizontal scan line appear alternately. It is. For this reason, when a video signal reproduced from a movie film is inserted into, for example, an NTSC video signal of a television program, first, a video signal of 24 frames per second reproduced from a movie film is converted into an NTSC video signal of 60 fields per second. Convert to The video signal obtained by this conversion is called a 2: 3 pull-down video signal. The 2: 3 pull-down video signal is inserted into the NTSC video signal of the television program.

  By the way, in a liquid crystal display, progressive display control is performed in which images on each horizontal scanning line in one screen are sequentially displayed without being skipped. For this reason, when displaying an NTSC video signal on a display device such as a liquid crystal display, interlace / progressive conversion (I / P conversion) for converting the NTSC video signal into a progressive video signal is performed in the display device. Is called.

  Here, various methods of I / P conversion for a normal NTSC video signal are provided. However, when this type of I / P conversion is performed on a 2: 3 pull-down video signal and displayed on a liquid crystal display or the like, there is a problem that an image with a deteriorated image quality is displayed compared to the original movie film image. .

  In order to avoid the occurrence of such a problem, the 2: 3 pull-down video signal included in the NTSC video signal is identified, and the 2: 3 pull-down video signal is configured for the 2: 3 pull-down video signal. It is necessary to perform I / P conversion suitable for the above.

  Various techniques for identifying this type of 2: 3 pull-down video signal have been proposed. For example, the technique disclosed in Patent Document 1 detects each correlation value between the video signal of the field of interest and the video signal of the previous and next field, and the two correlation values are detected. When the difference is large, it is determined that the video signal is a 2: 3 pull-down video signal. In Patent Document 1, since it is determined that the video signal is a 2: 3 pull-down video signal using only a total of three fields including the current and one field before and after the current field, the determination can be quickly made.

JP-A-11-261972

  However, since the technique disclosed in Patent Document 1 performs the determination using only three fields, the 2: 3 pull-down video signal with a lot of noise, the video content is fine (there are many high-frequency components), or the 2: 3 pull-down video signal or field. There is a problem that there is a possibility that a 2: 3 pull-down video signal with little correlation between the lines may not be distinguished from a 2: 3 pull-down video signal.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a video signal processing apparatus that efficiently and accurately identifies a 2: 3 pull-down video signal included in an NTSC video signal. It is said.

  The present invention relates to storage means for storing an input video signal, and for each input video signal, the video signal of the field and the video of the field two fields before the field stored in the storage means for each field. The difference calculation means for calculating the inter-frame difference value that is the difference from the signal, and the input video signal when the minimum value of the inter-frame difference value in five consecutive fields is equal to or less than the threshold value in the period of five fields. There is provided a video signal processing device characterized by comprising determination means for determining a 2: 3 pull-down video signal.

  According to the present invention, for each input video signal, an inter-frame difference value, which is a difference between the video signal of the field and the video signal of two fields before, is calculated for each field, and the inter-frame difference in the five consecutive fields is calculated. The minimum difference value is calculated. Then, when the minimum value of the inter-frame difference value in five consecutive fields becomes equal to or less than the threshold value in the period of five fields, it is determined that the input video signal is a 2: 3 pull-down video signal. Here, a 2: 3 pull-down video signal with a lot of noise, a 2: 3 pull-down video signal with a fine video content (a lot of high-frequency components) or a 2: 3 pull-down video signal with little correlation between lines in a field is a video signal processing device. However, even in such a case, if the minimum value of the inter-frame difference value is equal to or less than the threshold value with a period of 5 fields, the input video signal is 2: 3. There is a high possibility of being a pull-down video signal. Therefore, the 2: 3 pull-down video signal included in the input video signal can be identified efficiently and accurately.

It is a block diagram which shows the structural example of the video signal processing apparatus which is one Embodiment of this invention. It is a figure explaining a 2: 3 pull-down video signal. It is a block diagram explaining the video signal identification program 61 of the video signal processing apparatus which is one Embodiment of this invention. It is a figure explaining calculation of the threshold using the inter-frame difference value for 5 fields in the embodiment. It is a figure explaining determination of 2: 3 pull-down video signal in the embodiment.

Embodiments of the present invention will be described below with reference to the drawings.
<Embodiment>
FIG. 1 is a block diagram showing a configuration example of a video signal processing apparatus according to an embodiment of the present invention. The video signal processing apparatus according to the present invention is provided with a 2: 3 pull-down video signal included in an NTSC video signal in a display device capable of inputting an NTSC video signal, such as a liquid crystal television or an in-vehicle television. Means for identifying are provided. The means for identifying the 2: 3 pull-down video signal of the video signal processing apparatus according to the present invention can be realized by either software or hardware. In the present embodiment, a case will be described in which means for identifying a 2: 3 pull-down video signal is realized by software. However, as a matter of course, it may be realized as hardware by configuring the means for identifying the 2: 3 pull-down video signal by an electronic circuit having the same function as the software described below. Note that the video signal processing apparatus according to the present embodiment processes analog signals that do not include an additional signal indicating a 2: 3 pull-down video signal, such as NTSC video signals.

  As shown in FIG. 1, the video signal processing apparatus according to the present embodiment includes a CPU 10, a ROM 20, a RAM 30, a nonvolatile memory 60, a display unit 40, a receiving unit 70, and an interface unit 50. The CPU 10 is a control center that controls each part of the video signal processing apparatus. The ROM 20 is a read-only memory that stores a control program for controlling basic operations of the video signal processing apparatus such as a loader. The RAM 30 is a volatile memory used as a work area by the CPU 10.

  The display unit 40 is, for example, a liquid crystal display or the like, and is a display device whose scanning method is a progressive method.

  The receiving unit 70 includes, for example, an antenna and a circuit group that can receive an NTSC video signal from space.

  The interface unit 50 includes an interface capable of inputting and outputting video signals, such as a component terminal and a composite terminal, which are analog video terminals. A video playback device such as a DVD player is connected to the video signal processing device via a component terminal of the interface unit 50. An NTSC video signal is sent from the connected video reproduction device to the video signal processing device. The interface unit 50 may include a network interface for performing data communication with other devices via a network.

  The non-volatile memory 60 is a storage device for storing information such as programs, and for example, an HDD (Hard Disk Drive) is used. Among the contents stored in the nonvolatile memory 60, a video signal identification program 61 is unique to the present embodiment.

  Note that the program stored in the non-volatile memory 60 may be traded in a state installed in a liquid crystal television or the like, or may be traded separately in a state stored in a computer-readable storage medium. The program may be traded by downloading via a network.

  The video signal processing apparatus according to the present embodiment is characterized by the video signal identification program 61 stored in the nonvolatile memory 60. The video signal identification program 61 is a program for identifying a 2: 3 pull-down video signal included in an NTSC video signal.

  Prior to describing the video signal identification program 61, the 2: 3 pull-down video signal will be described with reference to FIG. The 2: 3 pull-down video signal is an interlace video signal of 60 fields per second generated from a video signal of 24 frames per second obtained from a movie film.

  FIG. 2 shows the relationship between each frame A, B, C, D of the movie film and the video signal of each field of the 2: 3 pull-down video signal generated therefrom. In the example shown in FIG. 2, the video signal corresponding to the odd-numbered horizontal scanning line is extracted from the frame A of the movie film to generate the video signal of the odd-numbered field Ao, and corresponds to the even-numbered horizontal scanning line pixels. The video signal of the even field Ae is generated by taking out the video signal. Also, by extracting the video signal corresponding to the pixels on the odd-numbered horizontal scanning lines from the frame B, the video signals of the two odd fields Bo having the same contents are generated, and the video corresponding to the pixels on the even-numbered horizontal scanning lines. The video signal of the even field Be is generated by extracting the signal. Further, the video signal corresponding to the even-numbered horizontal scanning line is extracted from the frame C to generate the video signal of the even-numbered field Ce, and the video signal corresponding to the odd-numbered horizontal scanning line is extracted to generate the odd-numbered video signal. A video signal of the field Co is generated. Further, by extracting the video signal corresponding to the pixels on the even-numbered horizontal scanning lines from the frame D, the video signals of the two even fields De having the same contents are generated, and the video corresponding to the pixels on the odd-numbered horizontal scanning lines is generated. The video signal of the odd field Do is generated by extracting the signal.

  In this way, processing for generating a video signal of two fields from one frame (for example, A) of two consecutive frames (for example, A and B) of a movie film and a video signal of three fields from the remaining one frame (for example, B) Is repeated to generate a 2: 3 pull-down video signal. In the present embodiment, an appropriate I / P conversion is performed on the 2: 3 pull-down video signal to generate a progressive video signal as illustrated. That is, the video signal of the even field Ae generated from the frame A is interpolated with the video signal of the odd field Ao generated from the frame A, and the frame A to the video signal of the even field Ae generated from the frame A. A video signal of a progressive system is generated by synthesizing the video signals of the odd field and even field generated from one frame, such as interpolating the video signal of the odd field Ao generated from. In order to perform this I / P conversion, it is necessary to identify whether or not the input video signal is a 2: 3 pull-down video signal. The video signal identification program 61 is a program for performing this identification.

  In the 2: 3 pull-down video signal, if attention is paid to the video signal of three fields generated from the same frame, the video signals of two fields adjacent to each other with one field in between are the same or similar contents (for example, The video signals of two odd fields Bo generated from the frame B and the video signals of two even fields De generated from the frame D). In addition, a 3-field video signal generated from the same frame in the 2: 3 pull-down video signal is generated in a 5-field cycle (for example, an odd field of field number 5 which is a third-field video signal generated from frame B). 5 fields after Bo is generated, the even field De of the field number 10 which is the video signal of the third field generated from the frame D is generated). Therefore, in the 2: 3 pull-down video signal, the difference between the video signals between the fields separated by 2 fields becomes zero or a small value close to zero in a period of 5 fields.

  Therefore, in the video signal identification program 61, for each input video signal, the difference between the video signal of the field and the video signal of two fields before is calculated for each field, and among the difference calculation results for five consecutive fields. The threshold value is calculated based on the difference calculation result that is the minimum value and the difference calculation result that is other than the minimum value, and the field in which the difference calculation result is the minimum value and less than or equal to the threshold value is repeated in a 5-field cycle. The video signal is determined to be a 2: 3 pull-down video signal. Details of the video signal identification program 61 will be described below.

  FIG. 3 is a block diagram for explaining the contents of the video signal identification program 61. The video signal identification program 61 includes software modules of an inter-frame difference calculation unit 210, a threshold calculation unit 220, a detection frame counter 230, a 2: 3 pull-down determination unit 240, and an I / P conversion unit 250. Further, the video signal identification program 61 uses a predetermined storage area in the RAM 30 as the one-field memories 110 to 150, the difference value registers 310 to 340, the minimum value register 400, and the non-minimum value register 410 to identify the video signal. Execute the process for

  The one-field memories 110 to 150 are means for storing NTSC video signals input via the receiving unit 70 or the interface unit 50, and each has a storage capacity capable of storing one field of video signals. Yes. Each one-field memory 110 to 150 is connected in series as shown in the figure, and functions as a FIFO (First-In First-Out) buffer.

  That is, when a video signal of a pixel in a certain field f (k) (f (k) indicates the k-th field) is written into the first-stage 1-field memory 110 via the receiving unit 70 or the interface unit 50, the field The video signal of the pixel in the field f (k−1) one field before f (k) is read from the one field memory 110 and written to the one field memory 120, and the field two fields before the field f (k). The video signal of the pixel of f (k-2) is read from the 1-field memory 120 and written to the 1-field memory 130, and the video of the pixel of the field f (k-3) three fields before the field f (k) The signal is read from the 1-field memory 130 and written to the 1-field memory 140, and the field f (k) The video signal of the pixel in the field f (k-4) before the field is read from the one field memory 140 and written in the one field memory 150, and the field f (k-5) five fields before the field f (k). The video signals of the pixels are read out from the one-field memory 150.

  The inter-frame difference calculation unit 210 receives the video signal of the current field f (k) input via the receiving unit 70 or the interface unit 50, and 2 fields before (ie, 1 frame before) read from the 1-field memory 120. Is a means for calculating a difference from the video signal of the field f (k−2). More specifically, the inter-frame difference calculation unit 210 calculates a difference between pixel values at the same pixel position in the video signal of the current field f (k) and the video signal of the field f (k−2) two fields before. An absolute value is calculated for all pixel positions in the field, and an operation of adding the absolute values of the differences calculated for all pixel positions in the field is executed. The calculation result of the inter-frame difference calculation unit 210 is referred to as an inter-frame difference value. The inter-frame difference calculation unit 210 calculates the inter-frame difference value d (k) calculated in the current field f (k) (that is, between the video signal in the field f (k) and the video signal in the field f (k−2). Difference value between frames) is output to the threshold value calculation unit 220 and the difference value register 310, respectively.

  The difference value registers 310 to 340 are means for temporarily storing the inter-frame difference value, which is the calculation result of the inter-frame difference calculation unit 210, and are connected in series as illustrated to form a four-stage shift register. Yes. In the present embodiment, the inter-frame difference calculation unit 210 outputs the inter-frame difference value at a period of one field, and the frame one field before output from the inter-frame difference calculation unit 210 until just before that. The difference value is written into the difference value register 310. At the same time, the inter-frame difference value is shifted from the difference value register 310 to the difference value register 320, the inter-frame difference value is shifted from the difference value register 320 to the difference value register 330, and the inter-frame from the difference value register 330 to the difference value register 340. The difference value is shifted.

  Therefore, when the interframe difference value d (k) between the field f (k) and the field f (k−2) is output from the interframe difference calculation unit 210 in the current field f (k), the difference value register 310 stores the difference value register 310. Stores the inter-frame difference value d (k-1) between the field f (k-1) and the field f (k-3) calculated one field before the inter-frame difference value d (k). The difference value register 320 stores the inter-frame difference value d (k−2) between the field f (k−2) and the field f (k−4) calculated two fields before. The inter-frame difference value d (k−3) between the field f (k−3) and the field f (k−5) calculated three fields before is calculated in the difference value register 340 four fields before. The interframe difference value d (k-4) between the field f (k-4) and the field f (k-6) is stored.

  The threshold calculation unit 220 receives the inter-frame difference value d (k) from the inter-frame difference calculation unit 210 and the inter-frame difference values d (k−1) to d (k−4) from the difference value registers 310 to 340, respectively. . The threshold calculation unit 220 calculates the moving average value of the inter-frame difference values d (k) to d (k−4) obtained for the five fields up to the present and the minimum value of the inter-frame difference values up to the present and the current frames. This is a means for calculating a threshold value using a moving average value that is not the minimum value of the difference values. This threshold value is used to detect an inter-frame difference value between two-field video signals generated from the same frame in a 2: 3 pull-down video signal that is likely to be an inter-frame difference value calculated sequentially. Is the threshold value.

  FIG. 4 is a diagram illustrating calculation of a threshold value using inter-frame difference values for five fields. In the example shown in FIG. 4, the inter-frame difference values d (k) to d (k-4) for five fields up to the present time are indicated by solid circles, and the inter-frame difference values for five fields up to one field before are shown. Is indicated by a dotted circle. Since the inter-frame difference value is a difference between video signals at intervals of two fields, if the video signal is a 2: 3 pull-down video signal, the inter-frame difference value is zero or a small value close to zero in a five-field period. For this reason, if the video signal is a 2: 3 pull-down video signal, the inter-frame difference values (d (k) to d (k-4) in FIG. 4) for five consecutive fields are included. A field having an inter-frame difference value (d (k) in FIG. 4) whose value is zero or a small value close to zero appears.

  Therefore, the threshold value calculation unit 220 first compares the inter-frame difference values (d (k) to d (k−4) in FIG. 4) for five fields up to the current time point, and compares these five fields. The minimum inter-frame difference value (d (k) in FIG. 4) and its field (f (k) in FIG. 4) are determined from among the inter-frame difference values.

  As a result of the determination, when the interframe difference value d (k) calculated in the current field is the minimum value among the interframe difference values for five fields up to the current time point, the threshold value calculation unit 220 sets the minimum value The moving average value of the minimum value of the inter-frame difference value in the value register 400 is updated. More specifically, first, the threshold calculation unit 220 reads the moving average value of the minimum value of the inter-frame difference value from the minimum value register 400. In the example shown in FIG. 4, before the field f (k), the minimum value d (k-5) of the inter-frame difference value is generated in the field f (k-5). The moving average value read from the minimum value register 400 is the moving average value of the minimum value of the inter-frame difference values generated up to this field f (k-5). Then, the threshold calculation unit 220 performs a weighted average of the moving average value read from the minimum value register 400 and the minimum value d (k) of the inter-frame difference value at the present time, thereby calculating the inter-frame difference value up to the present time. The moving average value of the minimum value is calculated and written in the minimum value register 400. In this way, the threshold value calculation unit 220 uses the minimum value in the minimum value register 400 every time the inter-frame difference value of the current field becomes the minimum value among the inter-frame difference values for five fields up to the present time. The moving average value of the minimum value of the inter-frame difference values is updated.

  On the other hand, as a result of comparing the inter-frame difference values for five fields up to the current time point to determine the minimum value, the inter-frame difference value d (k) calculated in the current field becomes five fields up to the current time point. When the difference value is not the minimum value among the inter-frame difference values, the threshold value calculation unit 220 updates the moving average value of the non-minimum value difference value in the non-minimum value register 410. More specifically, first, the threshold calculation unit 220 reads a moving average value of a value that is not the minimum value of the inter-frame difference value from the non-minimum value register 410. The moving average value read from the non-minimum value register 410 is a moving average value that is not the minimum value of the inter-frame difference values generated up to one field before (field f (k−1)). Then, the threshold calculation unit 220 performs a weighted average of the moving average value read from the non-minimum value register 410 and the inter-frame difference value d (k) that is not the minimum value at the present time, thereby obtaining the inter-frame difference value up to the present time. The moving average value of the values other than the minimum value is calculated and written in the non-minimum value register 410. In this way, the threshold value calculation unit 220 uses a value that is not the minimum value every time the inter-frame difference value of the current field becomes a value that is not the minimum value among the inter-frame difference values for five fields up to the present time. The moving average value of the non-minimum value of the inter-frame difference value in the non-minimum value register 410 is updated.

  When the threshold calculation unit 220 updates the moving average value of the minimum value of the inter-frame difference value in the minimum value register 400, the moving average value of the value that is not the minimum value of the inter-frame difference value from the non-minimum value register 410. , And the intermediate value between the read moving average value that is not the minimum value of the inter-frame difference value up to the present time and the moving average value of the minimum value of the inter-frame difference value written in the minimum value register 400 is the current value. Calculate as the threshold. Further, when the threshold calculation unit 220 updates the moving average value of the non-minimum value difference value in the non-minimum value register 410, the moving average value of the minimum value of the inter-frame difference value from the minimum value register 400 is updated. And the intermediate value between the read moving average value of the minimum value of the inter-frame difference value up to the present time and the moving average value of the non-minimum value register value written in the non-minimum value register 410 is the current time point. Is calculated as a threshold value.

  The threshold calculation unit 220 executes the above processing every time the field is switched, and updates the threshold based on the inter-frame difference value.

  In FIG. 3, the detection frame counter 230 of the video signal identification program 61 outputs an interframe difference value d (k) that has a minimum value within 5 fields and is equal to or less than a threshold value. Next, it is a means for counting the number of fields until the inter-frame difference value d (k) that is minimum within 5 fields and equal to or less than the threshold value is output.

  The 2: 3 pull-down determination unit 240 is a unit that determines from the count value of the detection frame counter 230 whether the NTSC video signal input to the video signal processing device is a 2: 3 pull-down video signal.

  The determination process performed by the 2: 3 pull-down video signal will be described with reference to FIG. In the example shown in FIG. 5, when the video signal of the field f (k-10) is input, the inter-frame difference value d (k-10) becomes the minimum value within the previous five fields and is equal to or less than the threshold value. Become. Therefore, the detection frame counter 230 counts the number of fields after the inter-frame difference value d (k−10) is generated.

  Then, video signals of fields f (k-9), f (k-8), f (k-7)... Are sequentially input, and a video signal of field f (k-5) is input. Sometimes, the inter-frame difference value d (k−5) becomes the minimum value within the previous five fields and is equal to or less than the threshold value. Here, the detection frame counter 230 is from when the previous minimum value d (k−10) of the inter-frame difference value is generated to when the minimum value d (k−5) of the current inter-frame difference value is generated. 5 fields are counted.

  Similarly, the detection frame counter 230 counts the number of fields after the inter-frame difference value d (k−5) is generated. When the video signal in the field f (k) is input, the inter-frame difference value d (k) becomes the minimum value within the previous five fields and is equal to or less than the threshold value. Here, the detection frame counter 230 is set to 5 during a period from when the previous minimum value d (k−5) of the inter-frame difference value is generated to when the minimum value d (k) of the current inter-frame difference value is generated. Count fields.

  In this way, if the input video signal is a 2: 3 pull-down video signal, the interframe difference value is equal to or less than the threshold value in a 5-field cycle. Therefore, the detection frame counter 230 determines whether the interframe difference value is equal to or less than the threshold value. The 2: 3 pull-down video signal can be identified by counting the period. Then, the 2: 3 pull-down determination unit 240 determines that the input NTSC video signal is the 2: 3 pull-down video signal when the detection frame counter 230 repeats the count of the period of five fields a predetermined number of times (a predetermined number of fields). It is determined that As described above, when the count of the period of 5 fields is repeated a predetermined number of times, it is determined that the input video signal is a 2: 3 pull-down video signal, so that the 2: 3 pull-down video signal can be accurately identified. When the input video signal is determined to be a 2: 3 pull-down video signal, the required number of repetitions of a 5-field period may be set by operating an operator provided in the display device.

  The 2: 3 pull-down determination unit 240 stores the inter-frame difference value and the threshold value in the minimum value register 400 in the latest five fields when the inter-frame difference value and the threshold value change greatly from the values calculated so far. The moving average value of the minimum value of the inter-frame difference value, the moving average value of the non-minimum value of the inter-frame difference value stored in the non-minimum value register 410, and the count value of the detection frame counter 230 are initialized. Specifically, when the inter-frame difference value and the threshold value greatly change with respect to the values calculated so far, the 2: 3 pull-down determination unit 240 then selects the inter-frame value that becomes the minimum value within the five consecutive fields. When a difference value occurs, the minimum value is written in the minimum value register 400 as a moving average value of the minimum value, and the average value of inter-frame difference values of non-minimum values in the five fields is moved to a value that is not the minimum value. The average value is written to the non-minimum value register 410, and the intermediate value between the minimum value written to the minimum value register 400 and the average value of the non-minimum value values written to the non-minimum value register 410 is used as a threshold value. The value is initialized to “0”, and the process of determining whether or not the input video signal is a 2: 3 pull-down video signal is resumed. In this way, even if the video signal changes greatly due to the influence of noise or the like and the threshold value or the like changes greatly from the value calculated so far, the 2: 3 pulldown is continuously performed by initializing the threshold value or the like. A video signal can be determined.

  Next, the I / P conversion unit 250 will be described. The I / P converter 250 is a means for converting an NTSC video signal (interlace video signal) into a progressive video signal based on the determination result of the 2: 3 pull-down determination unit 240. The I / P converter 250 receives the respective video signals (NTSC video signals) of the fields f (k-3) to f (k-5) from the 1-field memories 130 to 150. Further, the I / P conversion unit 250 receives the determination result of the 2: 3 pull-down determination unit 240.

  Upon receiving the determination result that the NTSC video signal is not a 2: 3 pull-down video signal, the I / P conversion unit 250 receives one of the fields f (k-3) to f (k-5). For the video signal (for example, field f (k-5)), I / P conversion is performed on a normal NTSC video signal. In this normal I / P conversion for NTSC video signals, different conversion methods are used for still images and moving images. In an I / P conversion performed on a still image, an even field including a video signal of each pixel on an even-numbered horizontal scanning line is used for an odd-numbered video signal including an image signal of each pixel on an odd-numbered horizontal scanning line. A progressive method video signal is generated by interpolating the video signal of the even-numbered field scan signal including the video signal of each pixel on the even-numbered horizontal scan line. A progressive format video signal is generated by interpolating an odd-field video signal including a pixel video signal. On the other hand, in the I / P conversion performed on the moving image, the video signal of each pixel on the odd-numbered horizontal scanning line adjacent to the upper and lower sides of the odd field including the video signal of each pixel on the odd-numbered horizontal scanning line is To generate a video signal of a progressive method by generating and interpolating a video signal of each pixel on the even-numbered horizontal scanning line, and to an even field including the video signal of each pixel on the even-numbered horizontal scanning line. On the other hand, a progressive format video signal is generated by interpolating by generating a video signal of each pixel on the odd-numbered horizontal scanning line between video signals of each pixel on the even-numbered horizontal scanning line adjacent to the top and bottom. Generate. Note that progressive video signals obtained only from odd-field video signals and progressive video signals obtained only from even-field video signals, such as I / P conversion performed on moving images, are based on the original video. Since it is not accurately reproduced, the image quality is deteriorated as compared with the original video signal (interlace video signal for one frame = 2 fields).

Upon receiving the determination result that the NTSC video signal is a 2: 3 pull-down video signal, the I / P converter 250 receives 2 for each video signal in the fields f (k−3) to f (k−5). : I / P conversion suitable for 3 pull-down video signals. Here, the I / P conversion suitable for the 2: 3 pull-down video signal will be described in more detail with reference to FIG. First, it is assumed that the video signals of the fields Ao, Ae, Bo, Be, Bo, Ce, Co, De, Do, De shown in FIG. 2 are sequentially input to the video signal processing apparatus according to the present embodiment. In this case, when the contents of the video signals in the fields f (k) to f (k-4) shown in FIG. 3 are as follows, the inter-frame difference calculation unit 210 has a minimum value within 5 fields. And an inter-frame difference value d (k) equal to or less than the threshold value is output.
f (k) = Bo
f (k-1) = Be
f (k-2) = Bo
f (k-3) = Ae
f (k-4) = Ao

Thereafter, when five fields have elapsed and the contents of the video signals in the fields f (k) to f (k-5) shown in FIG. 3 are as follows, the inter-frame difference calculation unit 210 is within five fields. Output an inter-frame difference value d (k) that is the minimum value and equal to or less than the threshold value.
f (k) = De
f (k-1) = Do
f (k-2) = De
f (k-3) = Co
f (k-4) = Ce
f (k-5) = Bo

  At this time, the detection frame counter 230 counts the number of fields “5” from the previous occurrence of the inter-frame difference value equal to or less than the threshold to the current occurrence. Therefore, at the shortest time, it is possible to determine that the input video signal is a 2: 3 pull-down video signal at this point. In this case, the I / P conversion unit 250 combines the video signal of the field f (k−3) = Co that is an odd field and the video signal of the field f (k−4) = Ce that is an even field. A progressive video signal representing the frame C of the movie film is generated.

Thereafter, when one field elapses, the contents of the video signals in the fields f (k-1) to f (k-5) shown in FIG. 3 are as follows.
f (k-1) = De
f (k-2) = Do
f (k-3) = De
f (k-4) = Co
f (k-5) = Ce

  In this case, the I / P converter 250 combines the video signal of the field f (k−4) = Co that is an odd field and the video signal of the field f (k−5) = Ce that is an even field. As in the previous field, a progressive video signal representing frame C of the movie film is generated.

Thereafter, when one field elapses, the contents of the video signals in the fields f (k-2) to f (k-5) shown in FIG. 3 are as follows.
f (k-2) = De
f (k-3) = Do
f (k-4) = De
f (k-5) = Co

  In this case, the I / P converter 250 synthesizes the video signal of the field f (k−3) = Do which is an odd field and the video signal of the field f (k−4) = De which is an even field. Then, a progressive video signal representing the frame D of the movie film is generated.

Thereafter, when one field elapses, the contents of the video signals in the fields f (k-3) to f (k-5) shown in FIG. 3 are as follows.
f (k-3) = De
f (k-4) = Do
f (k-5) = De

  In this case, the I / P converter 250 combines the video signal of the field f (k−4) = Do which is an odd field and the video signal of the field f (k−5) = De which is an even field. As in the previous field, a progressive video signal representing the frame D of the movie film is generated.

Thereafter, when one field elapses, the contents of the video signals in the fields f (k-4) and f (k-5) shown in FIG. 3 are as follows.
f (k-4) = De
f (k-5) = Do

  In this case, the I / P converter 250 combines the video signal of the field f (k−5) = Do which is an odd field and the video signal of the field f (k−4) = De which is an even field. As in the previous field, a progressive video signal representing the frame D of the movie film is generated.

  Hereinafter, the same applies, and the I / P converter 250 synthesizes the video signal of the field that was originally one frame in the video signal of each field of the 2: 3 pull-down video signal, as shown in FIG. To convert to a progressive video signal. As described above, in this embodiment, I / P conversion suitable for 2: 3 pull-down video signals is performed as compared with I / P conversion for normal NTSC video signals. The video can be reproduced.

  The video signal converted into the progressive format in this way by the I / P conversion unit 250 is displayed on the display unit 40. Note that the I / P converter 250 is not limited to the software module incorporated in the video signal identification program 61, and is based on the 2: 3 pull-down video signal identification result by the video signal identification program 61. You may make it function as another program which performs.

  When the video signal processing apparatus according to the present invention is realized by hardware, each software module (interframe difference calculation unit 210, FIG. 3) constituting the video signal identification program 61 is used instead of the video signal identification program 61. A video signal identification unit configured by an electronic circuit having the same functions as the threshold value calculation unit 220, the detection frame counter 230, the 2: 3 pull-down determination unit 240, and the I / P conversion unit 250). Is connected to the CPU 10 and functions according to the control of the CPU 10.

  As described above, according to the present embodiment, when the minimum value of the inter-frame difference value in five consecutive fields is repeated at a period of five fields, the input NTSC video signal is 2: It is determined that the signal is a 3 pull-down video signal, and at that time, the moving average value of the minimum value of the inter-frame difference value up to the present time and the moving average value of the value that is not the minimum value of the inter-frame difference value up to the present time Since the threshold value is updated based on this, even if the inter-frame difference value in the input video signal changes, the threshold value is updated following the change. For this reason, for example, a 2: 3 pull-down video signal with a lot of noise, a 2: 3 pull-down video signal with a fine video content (a lot of high-frequency components) or a 2: 3 pull-down video signal with little correlation between lines in a field is input. Even in this case, it can be efficiently and accurately identified that the input video signal is a 2: 3 pull-down video signal.

  In the video signal processing apparatus according to the present embodiment, the video signal is identified using the video signal of the current field f (k) and the video signal of the field f (k-2) two fields before, and the identification is performed. Based on the result, I / P conversion is performed on the video signals of fields f (k-3) to f (k-5) three to five fields before. As described above, since the identification of the video signal and the I / P conversion are performed in parallel, the identification of the video signal and the I / P conversion can be performed efficiently.

  Further, by changing the setting of the number of repetitions of the count value (5 fields) of the detection frame counter 230, for example, the image quality is improved by minimizing the determination error of the 2: 3 pull-down video signal, or 2: 3 pull-down. The degree of identification of the 2: 3 pull-down video signal can be freely changed, for example, by allowing a video signal determination error to improve the determination speed.

  In addition, the video signal processing apparatus according to the present embodiment does not require complicated calculations and can be easily increased in speed. In addition, the video signal processing apparatus according to the present embodiment requires a small circuit scale and a required memory amount, and can realize sufficient calculation accuracy.

<Other embodiments>
Although one embodiment of the present invention has been described above, other embodiments are conceivable for the present invention. For example:

(1) The video signal processing apparatus according to the above-described embodiment calculates, as a threshold value, an intermediate value between the moving average value of the minimum value of the interframe difference value and the moving average value of the value that is not the minimum value of the interframe difference value. . However, the moving average value of the minimum value of the inter-frame difference value and the average value of the inter-frame difference values that are not the minimum value among the inter-frame difference values for the five fields up to the present time (for example, the inter-frame difference value d (for five fields) When d (k) is the minimum value in k) to d (d-4), the average of d (k-1), d (k-2), d (k-3) and d (k-4) Value) may be calculated as a threshold value. Further, the minimum value itself of the latest predetermined number of inter-frame difference values before the current time is stored in the minimum value register 400, and the minimum value of the inter-frame difference values before the current time read from the minimum value register 400 and the current frame are stored. The average value of the inter-frame difference values is used as an intermediate value between the average value and the moving average value of the non-minimum inter-frame difference values, or the average value and the inter-frame difference value for 5 fields. An intermediate value with an average value of inter-frame difference values that is not the minimum value may be calculated as a threshold value. Also, an intermediate value between the current minimum value of the interframe difference value itself and the moving average value of the non-minimum interframe difference value or the current minimum value of the interframe difference value itself and the frames for five fields up to the present time An intermediate value with an average value of inter-frame difference values that is not the minimum value among the inter-difference values may be calculated as a threshold value.

  DESCRIPTION OF SYMBOLS 10 ... CPU, 20 ... ROM, 30 ... RAM, 40 ... Display part, 50 ... Interface part, 60 ... Non-volatile memory, 61 ... Video signal identification program, 70 ... Receiving part, 110,120,130,140,150 ... 1 field memory, 210 ... inter-frame difference calculation unit, 220 ... threshold calculation unit, 230 ... detection frame counter, 240 ... 2: 3 pull-down determination unit, 250 ... I / P conversion unit, 310, 320, 330, 340 ... difference Value register, 400 ... minimum value register, 410 ... non-minimum value register.

Claims (4)

Storage means for storing an input video signal;
Difference calculation means for calculating an inter-frame difference value, which is a difference between the video signal of the field and the video signal of the field two fields before the field stored in the storage means for each input video signal. When,
Determining means for determining that an input video signal is a 2: 3 pull-down video signal when a minimum value of inter-frame difference values in five consecutive fields is equal to or less than a threshold in a period of five fields;
A video signal processing apparatus comprising:   The minimum value of inter-frame difference values in five consecutive fields calculated by the difference calculating means is calculated, the moving average value of the minimum values of inter-frame difference values calculated up to the latest five fields, and the latest 2. The video signal processing according to claim 1, further comprising threshold value calculation means for calculating, as the threshold value, an intermediate value with a moving average value of non-minimum values of inter-frame difference values calculated up to five fields. apparatus.   When the determination means determines that the input video signal is a 2: 3 pull-down video signal, a progressive format video signal is obtained from the odd field video signal and the even field video signal stored in the storage means. 3. The video signal processing apparatus according to claim 1, further comprising interlace progressive conversion means for synthesizing.   When the inter-frame difference value or the threshold value changes beyond a predetermined limit, the determination unit calculates the moving average value of the minimum value of the inter-frame difference value calculated up to the latest 5 fields and the up to the latest 5 fields. 3. The video signal processing apparatus according to claim 2, wherein a moving average value that is not a minimum value of the inter-frame difference value is initialized.

Images