JP2010206478A - Telecine video signal detection device, telecine video signal detection method, and computer program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To determine with high accuracy whether a signal is a telecine video signal. <P>SOLUTION: A pixel difference between a present field image F(t)11 and a field image F(t-2)13, which is 2 fields in front, is obtained by a pixel difference calculating circuit 20 and is divided into blocks BL obtained by dividing a field via an in-block difference extraction circuit 30, and then calculations are performed by a histogram circuit 41, an average value calculating circuit 42, and a maximum value calculating circuit 43 and a mode determination is carried out by a difference determining circuit 50. Thus, whether an input video signal is a telecine video signal can be determined in units of blocks. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for detecting that an input video signal is a telecine video signal generated by pull-down processing.

  In general, when a 24-frame progressive (sequential) video signal is converted to a 60-field interlace video signal, for example, telecine conversion called 32-pull-down processing is performed. According to this telecine conversion, a movie film of 24 frames per second can be converted into an interlace video signal (telecine video signal) having a speed of 60 fields per second.

  FIG. 11 is an explanatory diagram for explaining the 32 pull-down process. As shown in the figure, according to the 32 pull-down process, an interlaced video signal of 5 fields is formed from 2 frames of movie film. That is, from the first frame “A” of the movie film, an “Ao” field, an “Ae” field, and an “Ao” field of the television signal are generated, and the second frame of the movie film “ From “B”, a “Be” field and a “Bo” field of the television signal are generated.

  In the “A” frame, the “Ao” field is repeatedly inserted for speed adjustment. The subscript “o” indicates a field obtained by thinning out even-numbered scan lines (leaving odd-numbered scan lines) of a motion picture film frame, and the subscript “e” indicates a motion picture film frame. This is a field in which odd-numbered scan lines are thinned out (even-numbered scan lines are left). A telecine video signal is obtained by sequentially recording the fields Ao, Ae, Ao, Be, Bo converted in this way. Thereafter, since this conversion is repeated, the telecine video signal generated by the 32 pull-down process has a periodicity of 5 fields.

  Conventionally, the absolute value of the difference between one field (that is, the field two fields before) is cumulatively calculated for each field, and both fields are overlapped by comparing the sum obtained by the cumulative calculation with the threshold value. There has been proposed a telecine video signal device that detects a telecine video signal by determining whether it is a field and checking the periodicity of the appearance of the overlapping field (for example, Patent Document 1).

JP 10-65964 A

  However, in the conventional technique, it is determined which of the two modes “there is difference” where the sum of differences is larger than the threshold and “no difference” where the sum of the differences is smaller than the threshold. When the input video signal includes a large noise, the input video signal that should be detected as a telecine video signal may be erroneously determined as a non-telecine video signal. For this reason, there has been a problem that it cannot be determined with high accuracy whether or not it is a telecine video signal.

  Such a problem is not limited to 32 pull-down processing, but is common to all types of pull-down processing.

  The present invention has been made to solve the above-described conventional problems, and it is an object of the present invention to determine whether or not a telecine video signal is highly accurate.

  In order to solve at least a part of the above problems, the present invention can be realized as the following forms or application examples.

Application Example 1 A telecine video signal detection device that detects that an input video signal is a telecine video signal generated by pull-down processing,
A relationship between a first field included in the input video signal and a second field that is a field two fields before the first field is a first mode in which there is no difference, a second in which there is a difference. A difference determination unit that determines which of the mode and the third mode that has noise but no difference corresponds to;
A telecine video signal detection device comprising: a video signal determination unit that determines whether or not the input video signal is a telecine video signal based on a determination result determined by the difference determination unit.

  According to the telecine video signal detection device described in Application Example 1, the difference determination unit determines the relationship between the first field and the second field two fields before. This determination is to determine which of the first mode with no difference, the second mode with difference, and the third mode with noise but with no difference. For this reason, it is possible to expect a highly accurate determination as compared with the prior art that determines the two modes of “no difference” and “with difference”. In particular, since the third mode is a mode in which there is noise but it is determined that there is no difference, it is possible to continue to determine that the input video signal is a telecine video signal when the input video signal includes large noise. Thus, it is possible to determine with high accuracy whether or not the video signal is a telecine video signal.

Application Example 2 In the telecine video signal detection device according to Application Example 1, the difference determination unit subtracts a gradation value at the same pixel position between the first field and the second field. A telecine comprising: a pixel difference calculation unit for obtaining a pixel difference; and a mode determination unit for determining which of the first to third modes is applied based on the pixel difference detected by the pixel difference calculation unit Video signal detection device.

  According to the telecine video signal detection device described in the application example 2, the pixel difference between the first field and the second field is obtained, and any one of the first to third modes corresponds to the pixel difference. Judgment is made. Therefore, a difference between fields can be easily detected.

Application Example 3 In the telecine video signal detection device according to Application Example 1 or 2, the difference determination unit adds a fourth mode that is difficult to determine to the first to third modes. A telecine video signal detection device that is configured to determine which of the first to fourth modes is applicable.

  According to the telecine video signal detection apparatus described in the application example 3, determination in four modes is possible, and the determination accuracy as to whether or not the video signal is a telecine video signal can be further increased.

[Application Example 4] The telecine video signal detection device according to any one of Application Examples 1 to 3, further including a buffer for accumulating the result of the determination by the difference determination unit over six generations on a time axis, The video signal determination unit is a telecine video signal detection device configured to perform the determination based on the content stored in the buffer.

  In the telecine video signal detection device according to the application example 4, it is possible to easily determine that the input video signal is a telecine video signal by providing a buffer for accumulating the determination result by the difference determination unit over six generations. .

Application Example 5 In the telecine video signal detection device according to any one of Application Examples 1 to 3, the difference determination unit includes a calculation unit that obtains a numerical value corresponding to the sum of pixel differences for the one field. A buffer for storing the result of the determination by the difference determination unit and a numerical value corresponding to the total sum of the pixel differences, and accumulating the set for six generations going back on the time axis, and further, the difference determination unit Is a configuration for determining that the latest numerical value obtained by the arithmetic unit corresponds to the third mode when the numerical value for the past five generations stored in the buffer is smaller than any of the numerical values. A telecine video signal detection device.

  In the telecine video signal detection device according to Application Example 5, numerical values corresponding to the sum of pixel differences for one field are accumulated in the buffer over six generations, and the latest numerical values obtained by the calculation unit of the difference determination unit are When it is smaller than any of the past five generations of values stored in the buffer, it is determined that it corresponds to the third mode. When this determination is made, it is likely that the video signal has noise even if it is “no difference”, so it is appropriate to enter the third mode in which there is noise but no difference. . Therefore, the third mode can be reliably detected.

Application Example 6 A telecine video signal detection device that detects that an input video signal is a telecine video signal generated by pull-down processing,
An input unit for inputting a first field included in the input video signal and a second field that is a field two fields before the first field;
A block in which each field can be divided into a plurality of fields, and a first mode in which there is no difference and a second mode in which there is no difference between the blocks at the same position in the first and second fields. And a difference determination unit that determines, for each block, which corresponds to one of the third modes having noise but no difference,
A video signal detection device comprising: a video signal determination unit that determines whether or not the input video signal for each block is a telecine video signal based on the determination result for each block determined by the difference determination unit .

  According to the telecine video signal detection device described in the application example 6, the relationship between the first field and the second field two fields before is determined for each block that can divide the field into a plurality of blocks. Whether this determination corresponds to the first mode in which there is no difference, the second mode in which there is a difference, or the third mode in which noise is determined but there is no difference, as in Application Example 1. Is what you want. For this reason, as in the telecine video signal detection apparatus described in Application Example 1, when the input video signal includes large noise, it is possible to continue the determination that the input signal is a telecine video signal. It can be determined with high accuracy.

Application Example 7 In the telecine video signal detection device according to Application Example 6, the difference determination unit subtracts a gradation value at the same pixel position between the first field and the second field. A pixel difference calculation unit for obtaining a pixel difference, and a block that divides a group of pixel differences for one field detected by the pixel difference calculation unit into a plurality of the blocks, and sequentially extracts pixel differences included in each block for each block Telecine video signal comprising: an inner pixel difference extraction unit; and a mode determination unit that determines, for each block, which one of the first to third modes corresponds based on the extracted pixel difference for each block. Detection device.

  According to the telecine video signal detection device described in the application example 7, after obtaining the pixel difference between the first field and the second field, the pixel difference included in each block is extracted in order for each block. Therefore, the difference between fields for each block can be easily detected.

Application Example 8 In the telecine video signal detection device according to Application Example 6 or 7, the difference determination unit adds a fourth mode that is difficult to determine to the first to third modes. A telecine video signal detection device that is configured to determine which of the first to fourth modes is applicable.

[Application Example 9] The telecine video signal detection device according to any one of Application Examples 6 to 8, further comprising a buffer for accumulating the result of the determination by the difference determination unit over six generations in a time axis, The video signal determination unit is a telecine video signal detection device configured to perform the determination based on the content stored in the buffer.

Application Example 10 In the telecine video signal detection device according to any one of Application Examples 6 to 8, the difference determination unit includes a calculation unit that calculates a numerical value corresponding to the sum of pixel differences for the one block. A buffer for storing the result of the determination by the difference determination unit and a numerical value corresponding to the total sum of the pixel differences, and accumulating the set for six generations going back on the time axis, and further, the difference determination unit Is a configuration for determining that the latest numerical value obtained by the arithmetic unit corresponds to the third mode when the numerical value for the past five generations stored in the buffer is smaller than any of the numerical values. A telecine video signal detection device.

  The telecine video signal detection apparatus described in Application Examples 8 to 10 has the same effects as the telecine video signal detection apparatus described in Application Examples 3 to 5.

Application Example 11 A telecine video signal detection method for detecting that an input video signal is a telecine video signal generated by pull-down processing,
A relationship between a first field included in the input video signal and a second field that is a field two fields before the first field is a first mode in which there is no difference, a second in which there is a difference. Determine which mode, and which of the third mode has noise but no difference,
A telecine video signal detection method for determining whether or not the input video signal is a telecine video signal based on a result of the determination.

Application Example 12 A telecine video signal detection method for detecting that an input video signal is a telecine video signal generated by pull-down processing,
A first field included in the input video signal and a second field that is a field two fields before the first field;
A block in which each field can be divided into a plurality of fields, and a first mode in which there is no difference and a second mode in which there is no difference between the blocks at the same position in the first and second fields. , And a third mode in which noise is present but no difference is determined for each block,
The telecine video signal detection method of determining whether the said input video signal for every said block is a telecine video signal based on the determination result for every said block determined by the said difference determination part.

Application Example 13 A computer program for detecting that an input video signal is a telecine video signal generated by pull-down processing,
On the computer,
A relationship between a first field included in the input video signal and a second field that is a field two fields before the first field is a first mode in which there is no difference, a second in which there is a difference. A function and a function for determining which of the third mode that has noise but no difference is determined;
A computer program for realizing a function of determining whether or not the input video signal is a telecine video signal based on the result of the determination.

Application Example 14 A computer program for detecting that an input video signal is a telecine video signal generated by pull-down processing,
On the computer,
A function of inputting a first field included in the input video signal and a second field that is a field two fields before the first field;
A block in which each field can be divided into a plurality of fields, and a first mode in which there is no difference and a second mode in which there is no difference between the blocks at the same position in the first and second fields. And a function for determining for each block which corresponds to one of the third modes having noise but no difference,
A computer program for realizing a function of determining whether or not the input video signal for each block is a telecine video signal based on the determined determination result for each block.

  According to the telecine video signal detection method described in the application example 11 and the computer program described in the application example 13, whether or not the telecine video signal is a field unit, as in the telecine video signal detection apparatus described in the application example 1. And the determination can be performed with high accuracy. According to the telecine video signal detection method described in application example 12 and the computer program described in application example 14, whether or not the telecine video signal is in block units, as in the telecine video signal detection apparatus described in application example 6. And the determination can be performed with high accuracy.

  The present invention can be realized in various modes, and includes, for example, a video processing apparatus including the telecine video signal detection apparatus, or a method thereof, a recording medium on which the computer program is recorded, and the computer program. It can be realized in the form of a data signal embodied in a carrier wave.

It is a block diagram which shows the video processing apparatus 1000 provided with the telecine video signal detection apparatus 100 as 1st Example to which this invention is applied. It is explanatory drawing which shows an example which divides | segments the pixel difference D for 1 field into a some block. 5 is an explanatory diagram illustrating an example of a difference buffer 60. FIG. It is explanatory drawing for demonstrating IP conversion corresponding to 32 pull-down processing. 4 is a flowchart for explaining the overall processing of the telecine video signal detection apparatus 100. It is explanatory drawing which displays as a list the determination conditions of S, T, U, and D mode. It is explanatory drawing which shows an example of the histogram of the video signal in the case where there is no noise and when there is. It is explanatory drawing which shows the conditions which switch to the effect that the video signal identification data VD is a telecine video signal. It is explanatory drawing which shows the conditions which switch to the effect that video signal identification data VD is a non-telecine video signal. It is a block diagram which shows the video processing apparatus 2000 provided with the telecine video signal detection apparatus 300 as 2nd Example to which this invention is applied. It is explanatory drawing for demonstrating 32 pull-down process.

  Embodiments of the present invention will be described below based on examples with reference to the drawings.

1. First embodiment:
1-A. Constitution:
FIG. 1 is a block diagram showing a video processing apparatus 1000 including a telecine video signal detection apparatus 100 as a first embodiment to which the present invention is applied. As illustrated, the video processing apparatus 1000 includes a telecine video signal detection apparatus 100 and an IP conversion circuit 200.

  The telecine video signal detection apparatus 100 includes three field memories 11 to 13, a pixel difference calculation circuit 20, an in-block pixel difference extraction circuit 30, a histogram circuit 41, an average value calculation circuit 42, and a maximum value calculation circuit 43. A difference determination circuit 50, a difference buffer 60, and a 32 pull-down determination circuit 70.

  The three field memories 11 to 13 each store input images (F (t), F (t-1), and F (t-2): t is time) for three consecutive fields in time. An externally input video signal (input video signal) V1 is a signal indicating a 60-field interlaced video (telecine video) generated by 32 pull-down processing.

  The pixel difference calculation circuit 20 reads the current field image (F (t)) and the field image two fields before (F (t−2)) from the field memories 11 and 13, and calculates the gradation value at the same pixel position. Subtraction is performed to obtain a difference D (hereinafter referred to as “pixel difference”) D for each pixel. Specifically, the pixel difference calculation circuit 20 performs the calculation of the following expression (1) over all the pixels in the field.

Here, R _xy is the gradation value of the red component for the pixel at the position of horizontal x and horizontal y, and G _xy is the gradation value of the green component for the pixel at the position of horizontal x and vertical y. , B _xy are gradation values of the blue component for the pixels at the horizontal x and vertical y positions. F _t is for the current field image (F (t)), and F _t−2 (R _xy ) is for the field image (F (t−2)) two fields before. It shows that. Abs indicates an absolute value.

That is, according to the equation (1), the pixel differences D (R _xy , G _xy , B _xy ) are obtained by performing the following calculations (a) to (d).

(A) From the red component gradation value F _t-2 (R _xy ) of the pixel at the position (x, y) on the current field image (F (t)), the field image (F (t) -2)) The red component gradation value F _t-2 (R _xy ) of the pixel at the upper position (x, y) is subtracted to find the absolute value of the answer.
(B) From the green component gradation value F _t−2 (G _xy ) of the pixel at the position (x, y) on the current field image (F (t)), the field image (F (t) -2)) The green component gradation value F _t-2 (G _xy ) of the pixel at the upper position (x, y) is subtracted to obtain the absolute value of the answer.
(C) From the blue component gradation value F _t−2 (B _xy ) of the pixel at the position (x, y) on the current field image (F (t)), the field image (F (t -2)) The blue component gradation value F _t-2 (B _xy ) of the pixel at the upper position (x, y) is subtracted to find the absolute value of the answer.
(2) The average value of the absolute values for the respective color components obtained by the above (a) to (c) is defined as a pixel difference D (R _xy , G _xy , B _xy ).

The intra-block pixel difference extraction circuit 30 divides the pixel difference D (R _xy , G _xy , B _xy ) for one field detected by the pixel difference calculation circuit 20 into a plurality of blocks, and each pixel included in each block The difference D (R _xy , G _xy , B _xy ) is extracted sequentially for each block. Hereinafter, the pixel difference D (R _xy , G _xy , B _xy ) is simply referred to as “pixel difference D” as necessary.

  FIG. 2 is an explanatory diagram illustrating an example of dividing the pixel difference D for one field into a plurality of blocks. In the figure, SF is a set (pixel difference group) of pixel differences D for one field. If this set is called “difference field data”, the difference field data SF is divided into a plurality of blocks BL with the same area and the same shape. Specifically, when the input video signal V1 has a field screen size of, for example, 640 × 240 pixels, the size of the block BL is determined in advance so that the horizontal direction is 40 pixels and the vertical direction is 15 pixels. The difference field data SF is divided into a total of 256 pieces of 16 pieces in the horizontal direction and 16 pieces in the vertical direction. Here, “divide” does not physically separate data but means to divide into a plurality of blocks.

Note that the block BL divides the difference field data SF into a plurality of pieces, but it can also be said that each field from which the difference field data SF is generated can be divided into a plurality of pieces. The size of the block BL need not be limited to 40 × 15 pixels, but may be other sizes such as 320 × 120, 160 × 60, 80 × 30, and 8 × 3. Then, the intra-block pixel difference extraction circuit 30 sequentially extracts the pixel differences D (R _xy , G _xy , B _xy ) included in each block BL for each block BL.

  The histogram circuit 41, the average value calculation circuit 42, and the maximum value calculation circuit 43 perform the following processing using the pixel difference D for one block extracted by the intra-block pixel difference extraction circuit 30.

The histogram circuit 41 obtains a frequency distribution of the pixel difference D in each block. Specifically, whether D (R _xy , G _xy , B _xy ) obtained by the pixel difference calculation circuit 20 corresponds to any of the first to fifth distribution ranges Th0, Th1, Th2, Th3, Th4. judge. Thus, the size distribution of the pixel difference D in the block is obtained by counting the numbers corresponding to the distribution ranges Th0 to Th4 in one block.

  When the gradation value of the pixel is 8 bits, it takes a value of 0 to 255. In this case, the pixel difference D also takes a value of 0 to 255. In this case, the first to fifth distribution ranges Th0 to Th4 take the following values, for example.

0 ≦ Th0 <4
4 ≦ Th1 <8
8 ≦ Th2 <12
12 ≦ Th3 <16
16 ≦ Th4

  That is, the histogram circuit 41 outputs the number of pixel differences D corresponding to each of the first to fifth distribution ranges Th0 to Th4 as the frequency distribution His.

  The average value calculation circuit 42 obtains an average value Ave obtained by dividing the sum of the pixel differences D for all the pixels included in each block by the number of pixels in the block. The maximum value calculation circuit 43 obtains the maximum value Max of the pixel differences D of the pixels included in each block.

  The difference determination circuit 50 uses the calculation results of the histogram circuit 41, the average value calculation circuit 42, and the maximum value calculation circuit 43 to calculate the current field image (F (t)) and the previous field image (F (t)). The relationship between the blocks at the same position with respect to t-2)) is any of “S (Same)”, “T (Tentative Same)”, “U (Unknown)”, and “D (Different)”. It is determined whether the mode is applicable.

  Here, the S mode is a state in which the image change between the block in the current field image (F (t)) and the block two fields before is smaller than a predetermined level. That is, it is a state of “no difference” with almost no difference. In other words, the images in both blocks are in the same or almost the same state.

  In the T mode, the relationship between the block in the current field image (F (t)) and the block two fields before is determined as “difference” in the conventional difference determination, but there is noisy video. This is a state in which “no difference” is determined in consideration of The T mode corresponds to the “third mode” in the present invention.

  In the U mode, the relationship between the block in the current field image (F (t)) and the block two fields before is “difference exists” in the conventional difference determination (determination of whether it is larger or smaller than the threshold). "And" No difference "cannot be determined. In other words, the determination is difficult. The U mode corresponds to the “fourth mode” in the present invention.

  In the D mode, the image change between the block in the current field image (F (t)) and the block two fields before is larger than a predetermined level. That is, the state is “with difference”. In other words, the images in both blocks are different.

  It should be noted that the S, T, U, or D mode is determined by determining the frequency distribution His in the block detected by the histogram circuit 41, the average value Ave in the block detected by the average value calculating circuit 42, and This is performed by comprehensively determining the maximum value Max in the block detected by the maximum value calculation circuit, and a specific method thereof will be described later.

  The difference determination circuit 50 further determines a determination result as to which of the S, T, U, and D modes (hereinafter, this determination result is referred to as “difference mode”), the frequency distribution His, A process of outputting the average value Ave and the maximum value Max to the difference buffer 60 is performed.

  FIG. 3 is an explanatory diagram illustrating an example of the difference buffer 60. Differences are accumulated in the difference buffer 60 over six generations going back on the time axis including the present. The “difference” here is a difference mode, a frequency distribution His, an average value Ave, and a maximum value Max sent from the difference determination circuit 50. As shown in the figure, every time the latest difference is recorded, the oldest one is deleted in order. Although not shown, the difference buffer 60 has as many storage fields as the number of blocks, and each storage field corresponds to each block. As a result, the difference buffer 60 stores differences for each block (difference mode, His, Ave, Max) for 6 generations. Note that the example shown is for one block, and an array for six generations is actually stored for the number of blocks.

  In the present embodiment, the difference mode, the frequency distribution His, the average value Ave, and the maximum value Max are stored as differences, but instead of this, the difference except for the average value Ave and the maximum value Max is used. Only the mode and the average value Ave may be stored. Further, the average value Ave may be replaced with the sum of the pixel differences D for all the pixels included in each block.

  Based on the content stored in the difference buffer 60, the 32-pull-down determination circuit 70 determines for each block whether or not the input video signal is a telecine video signal generated by the 32-pull-down process. Details of this determination method will be described later. The determination result is output to the outside of the telecine video signal detection apparatus 100 for each block as video signal identification data VD that can identify a telecine video signal / non-telecine video signal.

  The intra-block pixel difference extraction circuit 30, the histogram circuit 41, the average value calculation circuit 42, the maximum value calculation circuit 43, and the difference determination circuit 50 correspond to the “difference determination unit” provided in the present invention. The 32 pull-down determination circuit 70 corresponds to the “video signal determination unit” provided in the present invention. With the configuration of the telecine video signal detection apparatus 100 described above, it is possible to detect that the input video signal V1 is a telecine video signal generated by 32 pull-down processing.

  The IP conversion circuit 200 receives the video signal identification data VD output from the telecine video signal detection device 100, appropriately selects a process for conversion based on the video signal identification data VD, and selects the input video signal V1 that is an interlace video signal. Convert to 60-field progressive video (IP conversion).

  FIG. 4 is an explanatory diagram for explaining IP conversion corresponding to 32 pull-down processing. The IP conversion shown corresponds to the 32 pull-down process shown in FIG. That is, when the 32 pull-down process shown in FIG. 11 is used, for example, when a cinema video is sent out from a broadcasting station as an NTSC TV signal, the IP conversion shown in FIG. 4 uses the telecine video signal sent from the broadcasting station. IP conversion is performed so that it can be displayed on a 60 frame progressive display device.

  Specifically, in view of the editing mode shown in FIG. 11 in the 32 pull-down process, 60 frames are obtained by combining the fields in which the odd-numbered scanning lines are left and the fields in which the even-numbered scanning lines are left in the telecine video signal. The progressive video signal is generated. As a result, a progressive video image obtained by repeating 2 frames and 3 frames every other frame of the original image is obtained.

  Returning to FIG. 1, in the IP conversion circuit 200 of this embodiment, instead of performing IP conversion in units of one field as described above, IP conversion is performed individually for each block set in one field. That is, the IP conversion circuit 200 receives the input video signal V1, divides one field of the input video signal with the same area and the same shape as the block division in the intra-block pixel difference extraction circuit 30, and each divided block is Whether the video signal is a telecine video signal is determined from the video signal identification data VD input from the telecine video signal detection device 100, and the video signal in the block determined to be a telecine video signal has the above-described mode. Perform IP conversion. On the other hand, the video signal in the block determined not to be a telecine video signal is not subjected to the IP conversion corresponding to the 32 pull-down process described above, and is subjected to, for example, a deinterlace process.

  As a result, the IP conversion circuit 200 converts the input video signal V1 into a 60-frame progressive video signal V2 and outputs it to a display device (not shown).

  Next, the entire process of the telecine video signal detection apparatus 100 will be described. FIG. 5 is a flowchart showing telecine video detection processing executed by the telecine video signal detection apparatus 100. This telecine video detection process is repeatedly executed every 1/60 seconds, which is the transmission interval of one field of the input video signal V1.

  As shown in FIG. 5, when the processing is started, the telecine video signal detection apparatus 100 first reads the input video signal V1 for one field by driving the pixel difference calculation circuit 20, and reads the read video signal V1. A pixel difference D between the field and the field two fields before when the telecine video detection process is executed twice before is calculated (step S110).

  Next, the telecine video signal detection apparatus 100 divides the pixel difference D for one field obtained in step S110 into a plurality of blocks BL by driving the intra-block pixel difference extraction circuit 30 (step S120). A process of selecting one block from among the blocks BL is performed (step S125). The size of the block BL is determined in advance. For example, as described above, the horizontal direction is 40 pixels and the vertical direction is 15 pixels. One block selected in step S125 is sequentially shifted every time the process in step S125 is repeatedly executed. Thereafter, the telecine video signal detection device 100 similarly drives the intra-block pixel difference extraction circuit 30 to extract each pixel difference D included in the selected block (step S130).

  Next, the telecine video signal detection apparatus 100 drives the histogram circuit 41, the average value calculation circuit 42, and the maximum value calculation circuit 43, thereby the frequency distribution of the pixel difference D in the block BL that is the extraction target in step S130. His, average value Ave, and maximum value Max are calculated (step S140).

  Next, the telecine video signal detection apparatus 100 drives the difference determination circuit 50 to perform a difference determination based on the frequency distribution His, the average value Ave, and the maximum value Max calculated in step S140 (step S150). In this difference determination, whether the image change in the block selected in step S125 corresponds to the above-described S, T, U, or D mode is determined according to the following determination condition.

  FIG. 6 is an explanatory diagram for displaying a list of determination conditions for the S, T, U, and D modes. The numerical values in the determination conditions shown in the figure are described as the case where the gradation value of the pixel takes a value from 0 to 255. As shown in the figure, the determination condition for entering the S mode is that when the average value Ave is 5 or less and both the distribution ranges Th3 and Th4 in the frequency distribution His both have the value 0 (that is, the distribution is on the larger side). When there is no pixel difference D to be performed).

  The determination conditions for entering the T mode are (i) the average value Ave is 6 or less, (ii) Th3 in the frequency distribution His is 5 or less, (iii) Th4 is 1 or less, (Iv) The average value Ave is satisfied when all of the average values for the past five generations of the corresponding block stored in the difference buffer 60 are satisfied. The average value for the five generations stored in the difference buffer 60 is the average value stored in the first position (order from the newest, the same applies hereinafter) to the fifth position of the difference buffer 60 at this time. is there. The physical meaning of the determination conditions (i) to (iv) will be described below.

  FIG. 7A is an example of a histogram of a video signal without noise. As shown in the figure, the frequency distribution H1 indicated by the solid line is for “S: no difference”, and the frequency distribution H2 indicated by the one-dot chain line is for “D: difference”. The frequency distribution H1 in the case of “S: no difference” is sufficiently biased toward the side where the pixel difference becomes 0, and the distribution is clearly divided between “with difference” and “without difference”. For this reason, it is easy to set the threshold value TH for determination, and it is possible to determine whether there is “difference” or “no difference” with high accuracy.

  FIG. 7B is an example of a histogram of a video signal with noise. As shown in the figure, the frequency distribution H1 in the case of “S: no difference” shows a value larger than 0 even when the pixel difference is relatively large. However, when there is noise, the noise is randomly superimposed on the entire screen up to a certain level. For this reason, it is difficult to clearly distinguish the distribution between “with difference” and “without difference”. Although it is possible to cope by changing the threshold value TH for determination, it is difficult to determine how much to change.

  Depending on the content, the noise may be more or less. Since there is a lot of noise and the influence on judgment is large, here we first considered reducing the influence. The above condition (iv) is for reducing the influence. As described above, the condition (iv) is “the average value Ave is smaller than any of the average values of the past five generations for the corresponding block stored in the difference buffer”, in other words, The sum Sum (0) of the pixel differences D for all the pixels included in the block is smaller than any of the sums Sum (−1) to Sum (−5) for the past five generations for the corresponding block. . If the sum Sum (0) of the current pixel difference D is smaller than any of the sums Sum (−1) to Sum (−5) for the past five generations, even if the video signal has noise, “ Since there is a high possibility of “no difference”, the T mode is set.

  The above conditions (i) to (iii) define the maximum allowable range of noise. The T mode was selected when the average value Ave was 6 or less, Th3 in the frequency distribution His was 5 or less, and Th4 was 1 or less. The content having noise that does not satisfy any one of the conditions (i) to (iii) is set to the D mode or the U mode, and the 32 pull-down process is not performed.

  The above conditions (i) to (iii) include the determination condition for the S mode, but when the determination condition for the S mode is satisfied, the determination condition for the T mode is not satisfied. S mode is assumed.

  Returning to FIG. 6, the determination condition for entering the U mode is that the average value Ave is 10 or less, Th3 in the frequency distribution His is 8 or less, and Th4 is 2 or less. It was time to meet. That is, even when the average value Ave is equal to or less than a certain value (= 10), a distribution exceeding the values (8, 2) in the large distribution range Th3, Th4 in the frequency distribution His is not found. Since neither the D mode nor the S mode can be determined, the U mode is set.

  The determination condition for entering the D mode is that the average value Ave is 10 or more, the maximum value Max is 80 or more, Th3 in the frequency distribution His is 5 or more, and Th4 is 2 or more. When any one of them was satisfied. That is, when the average value Ave is greater than or equal to a certain value (= 10), or when the maximum value Max is greater than or equal to a certain value (= 80), or the larger distribution range Th3, Th4 in the frequency distribution His. When a distribution greater than a certain value (5, 2) is observed, the D mode is set.

  The determination conditions are preferably performed in the order of S mode, T mode, U mode, and D mode. The determination result in step S150 is hereinafter referred to as “difference mode”.

  The above-described determination conditions for the S, T, U, and D modes are merely examples, and numerical values and mathematical expressions themselves can be changed to different conditions. In short, it is determined that the S mode has no difference, the T mode has noise but no difference, the U mode has difficulty in determination, and the D mode has difference. Any condition can be used as long as it is possible.

  Returning to FIG. 5, after the execution of step S150, the telecine video signal detection device 100 uses the difference mode obtained in step S150, the frequency distribution His, the average value Ave, and the maximum value Max obtained in step S140 for difference. Write to the buffer 60 (step S160). As described above with reference to FIG. 3, this writing is carried down to the oldest recorded one, and the oldest one is deleted every time the latest difference is recorded.

  After execution of step S160, the telecine video signal detection apparatus 100 determines that the input video signal is a telecine video signal based on the contents of the difference buffer 60 (specifically, the contents of the difference mode stored in the difference buffer 60). It is determined whether or not there is (step S170), and the determination result is output as video signal identification data VD (step S180). That is, when the determination result is a telecine video signal, the video signal identification data VD is output as, for example, “high level”, and when the determination result is a non-telecine video signal, the video signal identification data VD is, for example, “low level”. Is output.

  FIG. 8 is an explanatory diagram showing conditions for switching the video signal identification data VD to a telecine video signal. 60a in the figure is a field for storing the difference mode in the difference buffer 60 described above with reference to FIG. Similar to FIG. 3, the leftmost difference mode is the latest difference mode, and the generation is older as it is located to the right. As shown in FIG. 8A, when the latest differential mode is “S”, it is determined that the input video signal is a telecine video signal, and the video signal identification data VD is a telecine video signal. Switch to high level to indicate that. Further, as shown in FIG. 8B, when the latest differential mode is “T”, it is determined that the input video signal is a telecine video signal, and the video signal identification data VD is a telecine video. The signal is switched to a high level indicating that it is a signal. This means that the T mode functions as the S mode when switching from the non-telecine video signal to the telecine video signal.

  When the video signal identification data VD is at a low level and the latest differential mode is “U” or “D”, the video signal identification data VD is maintained at that level without being switched to a high level. This means that the U mode functions as a D mode when switching from a non-telecine video signal to a telecine video signal.

  FIG. 9 is an explanatory diagram showing conditions for switching the video signal identification data VD to a non-telecine video signal. FIG. 9D shows conditions for returning to the low level after the video signal identification data VD is switched to the high level. 9A to 9C, the high level is maintained without returning to the low level.

  As shown in FIGS. 9A to 9D, the contents of the field 60a of the difference buffer 60 are “D” for the second place (order from the newest, the same applies hereinafter), and “D” for the third place. The fourth place is “D”, the fifth place is “D”, and the sixth place is “S”. That is, it is assumed that “D” is input for four generations after “S” is input, and then the latest differential mode is input. Then, when the latest difference mode is “S” as shown in FIG. 9A, the five-cycle editing mode shown in FIG. 11 is shown, so that the input video signal is determined to be a telecine video signal. Thus, the video signal identification data VD maintains a high level indicating that it is a telecine video signal.

  Also, as shown in FIG. 9B, when the latest difference mode is “T”, it is determined that the input video signal is a telecine video signal, and the video signal identification data VD is a telecine video signal. Maintain a high level of effect. Further, as shown in FIG. 9C, when the latest differential mode is “U”, it is determined that the input video signal is a telecine video signal, and the video signal identification data VD is a telecine video signal. Maintain a high level of effect. This means that the T mode and the U mode function as the S mode when switching from the telecine video signal to the non-telecine video signal.

  On the other hand, when the latest differential mode is “D” as shown in FIG. 9D, it is determined that the input video signal is switched to the non-telecine video signal, and the video signal identification data VD is a non-telecine video signal. Switch to low level to indicate that there is.

  In the description of FIG. 9, the contents from the second position to the fifth position are “DDDD”, but it is not necessarily limited to this arrangement. For example, “SSSS” is another arrangement. Even in such a case, it may be configured to continue the telecine video signal. This is because “SSSS” is a still image, and in this case as well, it can be a telecine video signal.

  Upon receiving the high-level video signal identification data VD, the IP conversion circuit 200 performs IP conversion corresponding to the 32 pull-down process on the input video signal in the block indicated by the video signal identification data VD.

  Returning to FIG. 5, after execution of step S180, the block for which it is determined whether or not it is a telecine video signal is the last block among the plurality of blocks included in one field read in step S110. (Step S190), and if it is determined that it is not the last block, the process returns to step S125, and the determination target is shifted to the next block. On the other hand, if it is determined in step S190 that the block is the last block, the process returns to “RETURN” and the telecine video signal detection process is temporarily terminated.

1-B. effect:
According to the telecine video signal detection apparatus 100 of the first embodiment configured as described above, the current field image (F (t)) and the field image two fields before (F (t-2)) are between. Is determined for each block BL obtained by dividing the field into a plurality of blocks. For this reason, it can be determined whether or not the input video signal is a telecine video signal for each block. For this reason, for example, in the case of a video signal in which content that has not been pulled down 32 is inserted in one field such as a character telop, it is determined that only the block on which the content is superimposed is a non-telecine video signal. Can do.

  Furthermore, according to the telecine video signal detection device 100 of the first embodiment, the above difference relationship includes the S mode with no difference, the D mode with a difference, the T mode with noise but no difference, and the determination. Because it is determined to be one of the U modes that are difficult to perform, expecting a more accurate determination than the conventional technology that determines the two modes of “no difference” and “with difference” Can do. In particular, since the T mode is a mode for determining that there is noise but no difference, it is possible to continue to determine that the input video signal is a telecine video signal when the input video signal includes a large amount of noise. It is possible to determine with high accuracy whether the video signal is a telecine video signal.

1-C. Variation:
In the above embodiment, the difference determination result is divided into four modes, S, T, U, and D. When switching from a telecine video signal to a non-telecine video signal, the function of the U mode is changed to the D mode. The T mode function is switched to the S mode side. However, the mode of this switching may be appropriately different from the above embodiment. For example, the U mode may be configured to function as a D mode when switching from a non-telecine video signal to a telecine video signal and when switching from a telecine video signal to a non-telecine video signal. it can. Also, the T mode can be appropriately different from the present embodiment. For example, when the video input signal V1 passed through the noise removal filter is input to the telecine video signal detection device 100, when the difference mode is determined to be “T”, the non-telecine video signal is not switched to the telecine video signal. May be.

2. Second embodiment:
FIG. 10 is a block diagram showing a video processing apparatus 2000 including a telecine video signal detection apparatus 300 as a second embodiment to which the present invention is applied. Compared with the telecine video signal detection device 100 of the first embodiment, the telecine video signal detection device 300 of this embodiment uses the pixel difference D for one field detected by the pixel difference calculation circuit 20 as it is as the histogram circuit 41, The difference is that it is sent to the average value calculation circuit 42 and the maximum value calculation circuit 43, and the other configurations are the same. The same parts as those in the first embodiment are denoted by the same reference numerals.

  That is, in the telecine video signal detection apparatus 100 of the first embodiment, a plurality of blocks are transmitted by sending the pixel difference D for one field detected by the pixel difference calculation circuit 20 to the intra-block pixel difference extraction circuit 30 (FIG. 1). In this embodiment, each pixel difference D is sent to the histogram circuit 41, the average value calculation circuit 42, and the maximum value calculation circuit 43 for each block. The pixel difference D is sent to the histogram circuit 41, the average value calculation circuit 42, and the maximum value calculation circuit 43. Each configuration of the difference determination circuit 50, the difference buffer 60, and the 32 pull-down determination circuit 70 performs the same processing as in the first embodiment, with the processing target for each field.

  According to the telecine video signal detection device 300 of the second embodiment having the above-described configuration, it is possible to detect that it is a telecine video signal in units of one field of the input video signal. Furthermore, since the difference between fields is determined in four modes of S, D, T, and U, it is possible to determine whether or not it is a telecine video signal with high accuracy as in the first embodiment. Can do.

3. Other embodiments:
The present invention is not limited to the above-described first and second embodiments and modifications thereof, and can be implemented in various modes without departing from the gist thereof. For example, the following modifications are possible. Is also possible.

(1) In the first and second embodiments, the difference determination is performed by determining which of the four modes S, D, T, and U corresponds. It is good also as a structure which eliminates a mode and determines which of three modes, S, D, and T, corresponds.

(2) In the first and second embodiments, the difference buffer 60 stores the difference mode as the difference determination result, the frequency distribution His, the average value Ave, and the maximum value Max in the difference buffer 60. However, it is not always necessary to store the data in one buffer, and the data may be stored separately in another memory.

(3) In the first and second embodiments, (iv) of the determination conditions for entering the T mode is “the past five generations of the corresponding block in which the average value Ave is stored in the difference buffer 60. However, instead of this, the average value Ave is first written in the difference buffer 60 and stored in the difference buffer 60. (Iv) may be determined by determining whether or not the average value (that is, the content stored in the first place) is minimum. Furthermore, instead of the average value, the sum Sum of pixel differences D is stored for the past five generations, and it is determined that the latest total is smaller than any of the average values for the past five generations. Also good.

(4) In the first embodiment, the pixel difference is first obtained between the current field image and the field image two fields before, and the difference field data, which is a set of pixel differences for one field, is converted into a block. Instead of this, the current field image and the field image two fields before are divided into a plurality of blocks, and the current field image block and the field image side two fields before are divided. It is good also as a structure which calculates | requires a pixel difference between these blocks. Even with this configuration, the same effects as in the first embodiment can be obtained.

(5) In the first and second embodiments described above, the gradation value of the pixel of the input video signal has been described as 8 bits (256 gradations). However, the input video signal having the other number of bits (gradation value) However, it can be implemented and has the same effect.

(6) The video processing apparatus 1000 according to the first and second embodiments can be applied to devices capable of displaying progressive video such as information devices such as personal computers and display devices such as liquid crystal televisions.

(7) In the first and second embodiments, the telecine video signal generated by the 32 pulldown process is detected. However, the present invention is not limited to the telecine video signal generated by the 32 pulldown process. It may be configured to detect a telecine video signal generated by pull-down processing of another method such as 23 pull-down processing or 22 pull-down processing.

(8) In the first and second embodiments, a part of the configuration realized by hardware may be replaced with software. Conversely, a part of the configuration realized by software is changed. It may be replaced with hardware. That is, it is good also as an Example using the computer program for making a computer implement | achieve the function of each circuit 20-50, 70, 110 in the said 1st and 2nd Example.

DESCRIPTION OF SYMBOLS 11 ... Field memory 20 ... Pixel difference calculation circuit 30 ... In-block pixel difference extraction circuit 41 ... Histogram circuit 42 ... Average value calculation circuit 43 ... Maximum value calculation circuit 50 ... Difference determination circuit 60 ... Difference buffer 70 ... 32 Pull-down determination circuit 100, 300 ... Telecine video signal detection device 200 ... IP conversion circuit 1000, 2000 ... Video processing device D ... Pixel Difference V1 ... Input video signal V2 ... Video signal VD ... Video signal identification data SF ... Difference field data BL ... Block Th0-Th4 ... First to fifth distribution ranges Max. .. Maximum value Ave ... Average value His ... Frequency distribution

Claims (14)

A telecine video signal detection device for detecting that an input video signal is a telecine video signal generated by pull-down processing,
A relationship between a first field included in the input video signal and a second field that is a field two fields before the first field is a first mode in which there is no difference, a second in which there is a difference. A difference determination unit that determines which of the mode and the third mode that has noise but no difference corresponds to;
A telecine video signal detection device comprising: a video signal determination unit that determines whether or not the input video signal is a telecine video signal based on a determination result determined by the difference determination unit. The telecine video signal detection device according to claim 1,
The difference determination unit
A pixel difference calculation unit that obtains a pixel difference by subtracting a gradation value at the same pixel position between the first field and the second field;
A telecine video signal detection device comprising: a mode determination unit that determines which one of the first to third modes corresponds to the pixel difference detected by the pixel difference calculation unit. The telecine video signal detection device according to claim 1 or 2,
The difference determination unit
A telecine video signal detection device having a configuration in which a fourth mode that is difficult to determine is added to the first to third modes to determine which of the first to fourth modes is applicable. . The telecine video signal detection device according to any one of claims 1 to 3,
A buffer for accumulating over 6 generations the result of the determination by the difference determination unit going back in time axis;
The video signal determination unit
A telecine video signal detection device configured to perform the determination based on the contents stored in the buffer. The telecine video signal detection device according to any one of claims 1 to 3,
The difference determination unit includes a calculation unit for obtaining a numerical value corresponding to the sum of pixel differences for the one field,
A buffer that accumulates the results of the determination by the difference determination unit and a numerical value corresponding to the total sum of the pixel differences and accumulates the set over six generations over a time axis,
Furthermore, the difference determination unit
When the latest numerical value obtained by the arithmetic unit is smaller than any of the past five generations of numerical values stored in the buffer, the configuration is determined to correspond to the third mode. Telecine video signal detection device. A telecine video signal detection device for detecting that an input video signal is a telecine video signal generated by pull-down processing,
An input unit for inputting a first field included in the input video signal and a second field that is a field two fields before the first field;
A block in which each field can be divided into a plurality of fields, and a first mode in which there is no difference and a second mode in which there is no difference between the blocks at the same position in the first and second fields. And a difference determination unit that determines, for each block, which corresponds to one of the third modes having noise but no difference,
A video signal detection device comprising: a video signal determination unit that determines whether or not the input video signal for each block is a telecine video signal based on the determination result for each block determined by the difference determination unit . The telecine video signal detection device according to claim 6,
The difference determination unit
A pixel difference calculation unit that obtains a pixel difference by subtracting a gradation value at the same pixel position between the first field and the second field;
An intra-block pixel difference extraction unit that divides a group of pixel differences for one field detected by the pixel difference calculation unit into a plurality of the blocks, and sequentially extracts pixel differences included in each block for each block;
A telecine video signal detection device comprising: a mode determination unit that determines, for each block, which one of the first to third modes corresponds based on the extracted pixel difference for each block. The telecine video signal detection device according to claim 6 or 7,
The difference determination unit
A telecine video signal detection device having a configuration in which a fourth mode that is difficult to determine is added to the first to third modes to determine which of the first to fourth modes is applicable. . The telecine video signal detection device according to any one of claims 6 to 8,
A buffer for accumulating over 6 generations the result of the determination by the difference determination unit going back in time axis;
The video signal determination unit
A telecine video signal detection device configured to perform the determination based on the contents stored in the buffer. The telecine video signal detection device according to any one of claims 6 to 8,
The difference determination unit includes a calculation unit for obtaining a numerical value corresponding to the sum of pixel differences for the one block,
A buffer that accumulates the results of the determination by the difference determination unit and a numerical value corresponding to the total sum of the pixel differences and accumulates the set over six generations over a time axis,
Furthermore, the difference determination unit
When the latest numerical value obtained by the arithmetic unit is smaller than any of the past five generations of numerical values stored in the buffer, the configuration is determined to correspond to the third mode. Telecine video signal detection device. A telecine video signal detection method for detecting that an input video signal is a telecine video signal generated by pull-down processing,
A relationship between a first field included in the input video signal and a second field that is a field two fields before the first field is a first mode in which there is no difference, a second in which there is a difference. Determine which mode, and which of the third mode has noise but no difference,
A telecine video signal detection method for determining whether or not the input video signal is a telecine video signal based on a result of the determination. A telecine video signal detection method for detecting that an input video signal is a telecine video signal generated by pull-down processing,
A first field included in the input video signal and a second field that is a field two fields before the first field;
A block in which each field can be divided into a plurality of fields, and a first mode in which there is no difference and a second mode in which there is no difference between the blocks at the same position in the first and second fields. , And a third mode in which noise is present but no difference is determined for each block,
The telecine video signal detection method of determining whether the said input video signal for every said block is a telecine video signal based on the determination result for every said block determined by the said difference determination part. A computer program for detecting that an input video signal is a telecine video signal generated by pull-down processing,
On the computer,
A relationship between a first field included in the input video signal and a second field that is a field two fields before the first field is a first mode in which there is no difference, a second in which there is a difference. A function and a function for determining which of the third mode that has noise but no difference is determined;
A computer program for realizing a function of determining whether or not the input video signal is a telecine video signal based on the result of the determination. A computer program for detecting that an input video signal is a telecine video signal generated by pull-down processing,
On the computer,
A function of inputting a first field included in the input video signal and a second field that is a field two fields before the first field;
A block in which each field can be divided into a plurality of fields, and a first mode in which there is no difference and a second mode in which there is no difference between the blocks at the same position in the first and second fields. And a function for determining for each block which corresponds to one of the third modes having noise but no difference,
A computer program for realizing a function of determining whether or not the input video signal for each block is a telecine video signal based on the determined determination result for each block.

Images