JP2004096223A - Video signal converting method and apparatus thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To convert interlace image signals into progressive image signals at a high precision by precisely detecting to what kind of image signal sequence the original interlace image signals are based on. <P>SOLUTION: In interpolating pixel signals using pixel signals and adjacent pixel signals thereto on horizontal lines in each field of an interlace image signal, a generator 16 obtains interpolated pixel signals at the same spatial positions as those of the pixel signals on the horizontal lines in the just previous field, and a detector 13 detects the difference between the interpolated pixel signal and that at the same spatial position in the just previous field. If the detected difference below a threshold is repeatedly detected at specified intervals, the interlace image signals are decided to be those based on the 2/2 sequence and converted to corresponding progressive image signals. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a video signal conversion method and apparatus for converting an interlaced video signal into a progressive video signal, and more particularly to a conversion method and apparatus suitable for converting a video signal based on a video created by a movie film or computer graphics.
[0002]
[Prior art]
In the interlaced video signal, a video signal of one frame is composed of a video signal of two fields, and a video signal of an odd field and a video signal of an even field among the video signals of two fields constituting one frame. Thus, the positions of the horizontal lines are alternately set at different positions. Therefore, the configuration is such that the signals of all the horizontal lines of the video signal of one frame are aligned with the video signal of two fields. Conventionally, all video signals for television broadcasting are interlaced video signals, and the television receiver processes and displays the interlaced video signals.
[0003]
In recent years, for such an interlaced video signal, a receiver has been developed which can generate a progressive video signal capable of displaying a higher-quality video and process and display the progressive video signal. I have. The progressive video signal is a video signal that is displayed without interlacing by setting the number of horizontal lines of the video signal of one field to twice that of the interlaced video signal, and the video signal displayed in each field period is 2 in the vertical direction. Double the density, contributing to higher image quality of the displayed video.
[0004]
[Problems to be solved by the invention]
By the way, when performing a conversion process for generating a progressive video signal from an interlaced video signal, it is necessary to double the number of horizontal lines of the video signal of each field. It is necessary to detect what kind of image signal the image signal is and to perform a conversion process for increasing the number of horizontal lines based on the detection.
[0005]
For example, assuming that the frame frequency of a video signal is 30 Hz, in the case of a video such as an animation created by computer graphics, 30 videos are created per second as an original video, and 1 In this case, an interlaced video signal is generated by allocating each video to one frame. The process of generating such an interlaced video signal is also called a 2.2 sequence.
[0006]
Therefore, in the case of such a video signal, the video signal of two fields constituting one frame is always a signal generated from the same video. When such a video signal is converted into a progressive video signal, a signal of a horizontal line to be increased in each field may be generated from a signal in one frame.
[0007]
On the other hand, in the case of a video recorded on a movie film, the number of frames on the film is 24 frames per second or the like, and does not match the frame frequency of the video signal, such as 30 Hz. A period in which a 2-frame video signal is generated from one frame of video and a period in which a 3-frame video signal is generated are combined so that a video of 24 frames per second is adjusted to a cycle of 30 frames. Therefore, in the case of such a video, the video in the odd field and the video in the even field may not be generated from the same video. The process of generating such an interlaced video signal is also called a 2.3 sequence.
[0008]
When such a video signal is converted to a progressive video signal, if the signal of the horizontal line to be increased in each field is generated only from the signal in one frame, the video of two different frames may be used in some cases. That is, a video signal of one field is generated, and the accuracy of the signal converted into the progressive video signal is deteriorated.
[0009]
From such a point, when converting an interlaced video signal into a progressive video signal, the sequence of the original interlaced video signal is detected as a video signal, and the conversion processing state is determined based on the detection. Needs to be changed. The details of the 2.2 sequence and the 2.3 sequence will be described in more detail in an embodiment described later.
[0010]
As a conventional process for detecting what kind of sequence the interlaced video signal is, for example, comparing the video signal of each field with the video signal of the immediately preceding field for each pixel, When the difference between the levels is lower than a predetermined threshold value and higher than the threshold value in one field cycle, the above-mentioned 2.2 sequence is detected. In other cases, when the period is lower and higher than the threshold value in other cycles, the sequence is detected as another sequence.
[0011]
This is a detection process using a point in which an odd field and an even field in one frame are basically composed of the same video in the case of a 2.2 sequence. However, even if the odd field and the even field are composed of the same video, because of the interlaced video signal, the positions of the adjacent pixels in the two fields are shifted by 1/2 line from each other. Also, even if the levels of adjacent pixels in two fields are compared, the levels are not always exactly the same.
[0012]
Therefore, the conventional 2.2 sequence detection processing cannot be said to have perfect detection accuracy, and there is a possibility of erroneous detection. If erroneous detection is performed, the interlaced video signal is converted to a progressive video signal. However, there has been a problem that the conversion processing is performed by an erroneous processing method and the resolution of the converted video signal is reduced.
[0013]
The present invention has been made in view of such a point, and when converting an interlaced video signal into a progressive video signal, performs accurate detection of what sequence the original interlaced video signal is based on. To perform high-precision conversion.
[0014]
[Means for Solving the Problems]
The present invention, when converting an interlaced video signal to a progressive video signal, the interpolation using the pixel signal of each horizontal line of each field of the interlaced video signal and the neighboring pixel signal of the pixel signal, the immediately preceding Obtain an interpolated pixel signal at the same spatial position as the pixel signal present on each horizontal line of the field, detect the difference between the interpolated pixel signal and the pixel signal at the same spatial position of the previous field, In the comparison between the detected difference and the threshold value, when it is detected that the difference becomes equal to or smaller than the threshold value, which is repeated at predetermined intervals, it is determined that the signal is an interlaced video signal by a 2.2 sequence, and the corresponding progressive signal is determined. The conversion into a video signal is performed.
[0015]
By doing so, when comparing the video signals of the two fields, the positions of the pixels to be compared are spatially the same, and an accurate pixel difference can be detected between the same positions. .
[0016]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.
[0017]
FIG. 1 is a diagram illustrating an example of a system configuration according to the present embodiment. In the present embodiment, the present invention is applied to an apparatus that performs a process of converting an interlaced video signal into a progressive video signal and displaying the converted progressive video signal on a built-in or connected video display unit. As shown in FIG. 1, an interlaced video signal is obtained at an interlaced video signal input terminal 1 by receiving a broadcast signal or reproducing a video tape, a video disk, or the like. As a signal system viewed from the number of lines of one frame of the video signal, there is a signal system of each line number such as 525i, 625i, and 1125i (i indicates that it is an interlace), but any of them may be used.
[0018]
The interlace video signal obtained at the input terminal 1 is supplied to a front video processing unit 2, and the interlace video signal processed by the front video processing unit 2 is sent to an interlace progressive conversion unit (hereinafter, referred to as an IP conversion unit) 3. The signal is supplied and converted into a progressive video signal in which the number of horizontal lines in each field period is doubled. The detailed configuration of the IP conversion unit 3 will be described later.
[0019]
The progressive video signal converted by the IP conversion unit 3 is supplied to a video display device driving unit 4 to be a video signal for driving the display device, and the driving unit 4 performs display driving of the video display unit 5, The video display unit 5 displays a video based on the progressive video signal. That is, non-interlace display for displaying all horizontal lines for each field period is performed. The image display unit 5 may be a display unit using any display means, such as a display unit using a cathode ray tube or a display unit using a liquid crystal display panel. The display may be of any type separate from the converter.
[0020]
FIG. 2 is a diagram illustrating a configuration example of the IP conversion unit 3 that converts an interlaced video signal into a progressive video signal. The interlaced video signal supplied from the front video processing unit 2 is supplied to a 2.2 conversion processing unit 11, an IP conversion processing unit 12, and a 2.2 sequence detection unit 13. The 2.2 conversion processing unit 11 is a circuit that performs a process of converting an interlaced video signal to a progressive video signal based on a 2.2 sequence. The IP conversion processing unit 12 is also a circuit that performs a process of converting an interlaced video signal into a progressive video signal. The IP conversion processing unit 12 is a circuit that performs a conversion process suitable for a sequence other than the 2.2 sequence. is there. The 2.2 sequence detector 13 is a circuit that detects that the supplied video signal is an interlaced video signal based on a 2.2 sequence.
[0021]
The interlaced video signal supplied from the front video processing unit 2 is also supplied to the writing processing unit 14, and the processing of the writing processing unit 14 stores the interlaced video signal in the field memory 18 connected to the IP conversion unit 3. , The input video signal is sequentially written. In FIG. 2, the field memory 18 is shown as an external circuit separate from the IP conversion processing unit 12, but may be a memory built in an integrated circuit in which the IP conversion processing unit 12 and the like are configured.
[0022]
The video signal written in the field memory 18 is read by the read processing unit 15 and supplied to the 2.2 conversion processing unit 11, the IP conversion processing unit 12, and the intra-field interpolation data generation unit 16. The intra-field interpolation data generating unit 16 generates pixel data of a horizontal line not in one field from the video signal of one field read from the memory 18 by interpolation, and generates the generated data (interpolated data). It is supplied to the IP conversion processing unit 12 and the 2.2 sequence detection unit 13.
[0023]
The progressive video signal converted by the 2.2 conversion processing unit 11 and the IP conversion processing unit 12 is supplied to the output circuit unit 17, and based on the detection result of the 2.2 sequence detection unit 13, one of the two is output. The video signal converted by the conversion processing unit 11 or 12 is supplied to the video display device driving unit 4 which is a subsequent circuit.
[0024]
The process of causing the intra-field interpolation data generation unit 16 in the IP conversion unit 3 to generate pixel data from signals in one field by interpolation is performed according to the principle shown in FIG. That is, since the video signal of each field is an interlaced video signal, the video signal is composed of half of a horizontal line included in the video of one frame, and is a signal decimated for each line. Therefore, for example, as shown in FIG. 4, there are real lines P and R in a certain field, and the line Q does not exist in that field. Line Q exists in the adjacent field. Circles a to o on each line indicate pixels existing on each line.
[0025]
Here, the intra-field interpolation data generation unit 16 generates a pixel signal of an interpolation line Q, which is a line between them, from each pixel signal of the real lines P and R in this field. For example, the center pixel h of the interpolation line Q shown in FIG. 4 is generated from the upper and lower pixels of the adjacent upper and lower lines P and R. Specifically, in order to generate a pixel h, a pair of upper and lower two pixels ao, bn, cm that pass through the pixel h in a straight line as shown by a line in FIG. , Dl, and ek, and selects the most probable data pair from the prepared pairs based on the state of the image at that time. Although a specific process of selecting the most probable data pair is omitted, a probable data pair may be selected based on, for example, determination of image movement at that time. Then, the data of the pixel h at the interpolation point is generated by the upper and lower value average value interpolation for taking the average of the selected upper and lower data pairs. For example, if the most probable data pair based on the state of the image at that time is (c−m), (data c + data m) / 2 is set as the data of the pixel h at the interpolation point.
[0026]
The data of the interpolation point calculated in this way is sent to the 2.2 sequence detecting unit 13 and the difference between the pixel on each line in one field and the pixel at the same spatial position in the next field is calculated. Is used to calculate That is, for example, it is assumed that a certain field F1 and a field F2 immediately before the field F1 exist as shown in FIG. 3, and lines L1 and L2 on the respective fields F1 and F2 exist. In the field F1, it is assumed that the odd lines L1, L3, L5} are actual lines, in the field F2, the even lines L2, L4, L6} are actual lines, and the remaining lines are interpolation lines.
[0027]
Here, for example, as for the pixel D13 on the line L3 of the field F1, the pixel at the same spatial position in the immediately preceding field F2 is the interpolation pixel D23 on the interpolation line L3, and this pixel D13 and the interpolation pixel D23 A difference is detected.
[0028]
In this way, for each pixel in one field, a difference from the interpolated pixel at the same spatial position in the immediately preceding field is detected, and the absolute value of the detected difference is determined by the 2.2 sequence detection. When a state lower than the threshold value detected by the unit 13 appears periodically every other field, it is determined that the video signal at that time is a video signal in a 2.2 sequence. Note that the 2.2 sequence detecting unit 13 determines the sum by adding the absolute values of the differences of all the pixels in one field. For example, when the number of lines in one frame is 525, 480 of them (that is, 1 (240 lines per field) are lines containing video information, and may be determined from the sum of differences between pixels on 240 horizontal lines in each field. However, the determination may be made from the sum of the absolute values of the pixels of some lines less than 240 lines.
[0029]
Here, an example of processing for determining a video signal of a 2.2 sequence will be described with reference to FIG. If the original video signal is a 2.2 sequence interlaced video signal, the original images A, B, and C change at one frame period as shown in FIG. In such a case, as shown in FIG. 5B, a video A ODD in a certain odd field period and a video A EVEN in an even field period are generated from the original image A as an interlaced video signal. Similarly, a video B ODD of the next odd field period and a video B EVEN of the even field period are generated from the original image B, and a video C ODD of the next odd field period and a video C of the even field period are generated from the original image C. EVEN is generated.
[0030]
In the case of a 2.2 sequence interlaced video signal, a video signal in which an original image changes in one frame unit (that is, in two field units) is obtained. For this reason, when the 2.2 sequence detector 13 detects the absolute value of the difference from each pixel in the immediately preceding field, unless the image is a still image, for example, as shown in FIG. The state below the threshold and the state above the threshold are repeated. However, if the original image is a still image that is the same image in succession, the state below the threshold is repeated. T1, t2, t3} in FIG. 6 are the absolute values of the differences calculated for each field.
[0031]
Here, in the case of the present example, the absolute value of the difference when the original image is the same and becomes lower than the threshold value is compared with the conventional case where actual pixels near adjacent fields are compared. Since the comparison is performed at the same position, the absolute value of the difference becomes low. Therefore, the threshold (the threshold indicated by the broken line in FIG. 6) for detecting the 2.2 sequence by the 2.2 sequence detecting unit 13 can be set to a lower value than in the related art, which is the 2.2 sequence. Can be detected with a higher probability and higher accuracy than before.
[0032]
Between adjacent fields lower than the threshold value, it is found that the fields are generated from the same original image, and the two fields are used to interpolate the pixels of the horizontal line that are mutually lacking to generate an interpolated pixel. By performing the processing to be performed by the 2/2 conversion processing unit 13, a progressive video signal in which the number of horizontal lines is doubled can be obtained for each field.
[0033]
When the 2.2 sequence detection unit 13 detects that the sequence is a 2.2 sequence, the output circuit unit 17 controls the selection of the output of the 2.2 conversion processing unit 13 so that the 2 · 2 sequence is output. The progressive video signal converted by the 2 conversion processing unit 13 is output. If the 2.2 sequence detecting unit 13 determines that the sequence is not a 2.2 sequence, the output circuit unit 17 controls to select the progressive video signal converted by the IP conversion processing unit 12. As for the IP conversion processing unit 12, the video signal supplied from the video processing unit 2 in the preceding stage, the video signal one field period before read from the field memory 18, and the interpolation signal generated by the intra-field interpolation data generation unit 16 Is a general conversion process of determining the motion / stillness of the image at that time and converting the interlaced signal into a progressive signal.
[0034]
Here, an example in which an interlaced video signal other than the 2.2 sequence is input will be described with reference to FIGS. For example, as described in the conventional example, in the case of a 2.3 sequence video signal in which a video signal of 30 frames per second is obtained from a video of 24 frames per second recorded on a movie film, the original image and The relationship between each field and the image is as shown in FIG. That is, when there are three original images A, B, and C as shown in FIG. 7A, one original image is composed of two fields as an interlaced video signal as shown in FIG. 7B. The case where it is used and the case where three fields are used periodically exist. For example, from the original image A, a video A ODD of a certain odd field period, a video A EVEN of the next even field period, and a video A ODD of the next odd field period are generated. Generates an image B EVEN in the next even field period and an image B ODD in the next odd field period. From the original image C, a video C EVEN of the next even field period, a video C ODD of the next odd field period, and a video C EVEN of the next even field period are generated.
[0035]
When the 2.2 sequence detector 13 detects the absolute value of the difference from each pixel in the immediately preceding field from the interlaced video signal of the 2.3 sequence, as shown in FIG. Results. That is, the state where the value is lower than the threshold value is continuous for two fields, then becomes higher than the threshold value for one field period, then is lower than the threshold value for one field period, and the value is lower than the threshold value for two consecutive field periods It becomes. When detection at such a cycle is repeated, it can be determined that the input video signal is an interlaced video signal of a 2.3 sequence, and is an input signal other than the 2.2 sequence, and is converted by the IP conversion processing unit 12. Control is performed to select a progressive video signal. Therefore, in such a case, the output of the 2.2 conversion processing unit 11 is not used.
[0036]
When a 2.3-sequence interlaced video signal based on the principle shown in FIG. 7 is input, a circuit for converting the interlaced video signal into a progressive video signal optimal for the 2.3-sequence is detected. Circuit may be provided so that the optimum conversion processing is performed even in the case of a 2.3 sequence.
[0037]
In the configuration shown in FIG. 2, the 2.2 conversion processing unit 11 and the IP conversion processing unit 12 are provided as separate circuits, and the conversion processing is executed by each circuit. And a state in which the conversion operation of the one conversion processing unit is performed based on the detection of the conversion sequence at that time, and the conversion operation for the 2.2 sequence is performed. May be switched. Further, a processing program for performing the conversion processing as in the present example may be installed in a video data processing device including a computer device or the like, and the same processing as the above-described device may be performed.
[0038]
【The invention's effect】
According to the present invention, when comparing the video signals of the two fields, the positions of the pixels to be compared are spatially the same, and accurate pixel differences can be detected between the same positions. Therefore, when an interlaced video signal is converted into a progressive video signal, detection of a 2.2 sequence or the like can be accurately performed, and accurate conversion processing can be performed based on the detection.
[0039]
In this case, when it is detected that the difference is equal to or less than the threshold value, which is a repetition at a cycle different from the predetermined cycle, the progressive video signal of the odd field and the progressive video signal of the even field are adjacent to each other. Since the creation from the interlaced video signal of the frame is periodically performed, for example, the conversion process in the case of a 2.3 sequence can be accurately performed.
[0040]
In addition, the interpolation pixel signal at each spatial position is an interpolation pixel signal selected from a plurality of candidates that can be generated from the pixel signals of two adjacent horizontal lines, thereby providing a more accurate interpolation pixel signal. Can be obtained, and a more accurate conversion sequence can be detected.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an example of the overall configuration according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration example of an IP conversion unit according to an embodiment of the present invention.
FIG. 3 is an explanatory diagram showing an example of generating an interpolation pixel according to an embodiment of the present invention;
FIG. 4 is an explanatory diagram showing an example of intra-field interpolation according to an embodiment of the present invention.
FIG. 5 is an explanatory diagram showing a video processing example of a 2.2 sequence according to the embodiment of the present invention;
FIG. 6 is an explanatory diagram showing an example of a calculation result of a difference absolute value in the example of FIG. 5;
FIG. 7 is an explanatory diagram illustrating an example of video processing of a 2.3 sequence according to an embodiment of the present invention.
FIG. 8 is an explanatory diagram showing an example of a calculation result of a difference absolute value in the example of FIG. 7;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Interlace video signal input terminal, 2 ... Front video processing part, 3 ... IP conversion part (interlace progressive conversion part), 4 ... Video display device drive part, 5 ... Video display part, 11 ... 2.2 conversion processing part , 12 ... IP conversion processing unit, 13 ... 2.2 sequence detection unit, 14 ... Write processing unit, 15 ... Read processing unit, 16 ... Inter-field interpolation data generation unit, 17 ... Output circuit unit, 18 ... Field memory

Claims (6)

In a video signal conversion method for converting an interlaced video signal into a progressive video signal,
Interpolation using the pixel signal of each horizontal line of each field of the interlaced video signal and the neighboring pixel signal of the pixel signal, the same spatial position as the pixel signal existing on each horizontal line of the immediately preceding field To obtain the interpolation pixel signal of
The difference between the interpolated pixel signal and the pixel signal at the same spatial position in the previous field is detected,
In the comparison between the detected difference and the threshold, when it is detected that the difference is equal to or smaller than the threshold, which is repeated every predetermined period,
A video signal conversion method for creating a two-field progressive video signal using an interlaced video signal of one frame composed of two fields. The video signal conversion method according to claim 1,
When the difference is equal to or smaller than the threshold, when it is detected that the repetition is performed at a different cycle from the predetermined cycle, the progressive video signal of the odd field and the progressive video signal of the even field are compared with the adjacent different frames. A video signal conversion method in which creation from an interlaced video signal is performed periodically. The video signal conversion method according to claim 1,
A video signal conversion method in which an interpolated pixel signal at each spatial position is an interpolated pixel signal selected from a plurality of candidates that can be generated from pixel signals of two adjacent horizontal lines. In a video signal conversion device that converts an interlaced video signal into a progressive video signal,
Interpolation using the pixel signal of each horizontal line of each field of the interlaced video signal and the neighboring pixel signal of the pixel signal, the same spatial position as the pixel signal existing on each horizontal line of the immediately preceding field Interpolation signal generation means for obtaining an interpolation pixel signal of
The difference between the interpolated pixel signal generated by the interpolated signal generating means and the pixel signal at the same spatial position in the immediately preceding field is detected, and the difference between the detected difference and the threshold value is equal to or smaller than the threshold value. That, a specific sequence detection means to determine whether to be repeated every predetermined period,
When the specific sequence detecting means detects that the difference becomes equal to or less than the threshold value is repeated at predetermined intervals, a progressive video signal of two continuous fields is generated using an interlaced video signal of one frame. A video signal conversion device comprising: The video signal conversion device according to claim 4,
The conversion unit, when the specific sequence detection unit detects that the difference is equal to or less than the threshold, which is a repetition at a period different from the predetermined period, the progressive video signal of the odd field and the even field A video signal conversion apparatus for periodically generating the progressive video signal from the interlaced video signals of adjacent different frames. The video signal conversion device according to claim 4,
The video signal conversion is such that the interpolation pixel signal at each spatial position generated by the interpolation signal generation means is an interpolation pixel signal selected from a plurality of candidates that can be generated from pixel signals of two adjacent horizontal lines. apparatus.

Images