JP2003299092A - Video encoder - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform effective encoding by detecting the redundant field contained in a telecine video image and deleting the image, when applying compression encoding to the telecine video image. <P>SOLUTION: A spatial filter 11 reduces noise in an inputted video signal and stores the signal in a field memory 12. An inter-field correlational value calculating section 13 calculates the inter-field correlational value for each partial region and for each composite partial region with respect to the image data of the memory 12. A redundant field judging section 14 judges whether or not a field is a redundant field by using the inter-field correlational value of the partial region or the composite partial region. A redundant field deleting section 15 deletes the field which is judged to be a redundant field from the memory 12. An encoder 16 applies compression encoding to the video signal formed by the field remaining without being deleted. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video encoding device for compressing telecine-converted video data with high efficiency.

[0002]

2. Description of the Related Art With the development of digital storage media, digital compression processing of video signals has become very important. In order to record a video signal having an enormous amount of information on a medium having a limited capacity for a long time and with high image quality, a highly efficient compression technique is required. Therefore, for a film image of a movie in which the video signal includes redundant fields, a method of removing the redundant fields and then compression-encoding is often used.

First, the redundancy, which is a characteristic of film images, will be described. A film image is generally a video signal with a frame frequency of 24 Hz, but when recorded on a VTR or the like, it is converted into an NTSC television signal with a frame frequency of 30 Hz. At this time 24Hz to 30Hz
Is converted into a video signal by repeatedly displaying a part of the fields constituting the video signal. This conversion is called telecine conversion, and the video signal subjected to telecine conversion is called telecine video. On the other hand, by removing the redundant field from the telecine image including the redundant field that is repeatedly displayed, 30 Hz to 24 Hz
The conversion to return to the video signal of is called inverse telecine conversion. Generally, when a telecine video is compression-encoded, a compression encoding process is performed using a video signal from which a redundant field has been removed by inverse telecine conversion for the purpose of improving the efficiency of the compression encoding.

Next, the telecine conversion and the inverse telecine conversion will be described. FIG. 3 shows an example of basic telecine conversion. In the conversion of FIG. 3, a 3 Hz telecine image is converted by repeatedly displaying field 3 and field 8 in a 24 Hz film image. In this case, the field 3 and the field 8 are redundant fields. On the other hand, FIG. 4 shows an example of basic inverse telecine conversion. In the conversion shown in FIG. 4, the field 3 and the field 8 which are repeatedly displayed in the telecine image are removed to convert the image to 24 Hz.

3 and 4 are examples of basic telecine conversion and inverse telecine conversion, but when irregular telecine conversion or editing is performed after telecine conversion, redundant fields included in the telecine image are displayed. The cycle is not always constant. Further, when an error or noise is mixed in the video signal in the process of telecine conversion or editing, even the fields displayed repeatedly do not always have the same image data. Therefore, in order to perform the inverse telecine conversion accurately, a method for detecting a redundant field included in the telecine image is required. As a method for performing the inverse telecine conversion, there is a method disclosed in Japanese Patent Laid-Open No. 2001-169252. JP 20
The method of No. 01-169252 is a method of detecting a redundant field by calculating a correlation between fields by statistically processing difference information between adjacent frames on a field-by-field basis.

[0006]

However, in the method of detecting the redundant field on a field-by-field basis as described above, in contrast to an image in which only a part of the screen is moving, a change in the image due to motion and an image due to noise are generated. However, there is a problem that it is impossible to discriminate the change of the image quality, the redundant field is erroneously detected, and the image quality is deteriorated in the video encoding.

[0007]

In order to solve such a problem, the invention according to claim 1 of the present application provides a video coding apparatus for inputting a video signal and performing compression coding processing on the video signal, 1 field that constitutes the video signal
In accordance with the inter-field correlation value calculating means for dividing the inter-field correlation value for each of the partial regions corresponding to adjacent frames, and the inter-field correlation value calculated for each of the partial regions A redundant field determining means for determining whether the field is a redundant field, and a redundant field deleting means for deleting the redundant field from the video signal based on a result determined by the redundant field determining means. It is a thing.

According to a second aspect of the present invention, in a video encoding device which receives a video signal as an input and compression-encodes the video signal, noise reducing means for reducing noise of the video signal, and A field forming a video signal is divided into one or more partial areas, and an inter-field correlation value calculating means for calculating an inter-field correlation value for each corresponding partial area between adjacent frames; Redundant field determination means for determining whether the field is a redundant field according to the inter-field correlation value, and redundant field deletion for deleting the redundant field from the video signal based on the result determined by the redundant field determination means. It is characterized by comprising means.

Further, in the invention according to claim 3 of the present application, the inter-field correlation value calculation means sets a composite partial area composed of one or more partial areas, and the partial area and the composite partial area are set. The feature is that the inter-field correlation value is calculated for each.

In the invention according to claim 4 of the present application, the redundant field determination means determines whether or not each of the composite partial areas is redundant, and when it is determined that all the composite partial areas are redundant. It is characterized in that the field is determined to be a redundant field.

In the invention according to claim 5 of the present application, the redundant field determination means is configured to perform the redundancy field determination means when the inter-field correlation values in all the partial areas forming the composite partial area are equal to or less than a predetermined threshold value. It is characterized in that the composite partial area is determined to be redundant.

Further, in the invention according to claim 6 of the present application, the redundant field determining means determines the threshold value by using the inter-field correlation value of the composite partial area calculated by the inter-field correlation value calculating means. The composite partial area is determined to be redundant when the inter-field correlation values in all the partial areas forming the composite partial area are equal to or less than the threshold value.

In the invention according to claim 7 of the present application, the inter-field correlation value calculating means calculates the inter-field correlation value by statistically processing inter-field difference information of image data for each of the partial areas. It is characterized by.

Further, in the invention according to claim 8 of the present application, the inter-field correlation value calculation means statistically processes the inter-field difference information of the image data for each of the partial region and the composite partial region, and the inter-field correlation value. It is characterized by calculating a value.

According to a ninth aspect of the present invention, the inter-field correlation value calculating means statistically processes the inter-field difference information of the image data for each of the partial areas to obtain a variance value and an average value, The dispersion value and the average value are used as the inter-field correlation value.

According to a tenth aspect of the present invention, the inter-field correlation value calculating means statistically processes the inter-field difference information of the image data for each of the partial area and the composite partial area to obtain a variance value and an average. A value is obtained, and the variance value and the average value are used as the inter-field correlation value.

The invention according to claim 11 of the present application is characterized by comprising a transmission means for transmitting compression-encoded video data. In addition, claim 12 of the present application
The described invention is characterized by comprising recording means for recording the compression-encoded video data.

The invention according to claim 13 of the present application is characterized by comprising a recording medium for recording the compression-encoded video data. In addition, claim 14 of the present application
The described invention is characterized by including a decoding means for decoding the compression-encoded video data. The invention according to claim 15 of the present application is characterized by comprising a decoding means for decoding the compression-encoded video data recorded on the recording medium.

With this configuration, it is determined whether the input video signal or the video signal with reduced noise is a redundant field according to the inter-field correlation value calculated for each partial area of the field. Then
When it is determined that the field is a redundant field, the field is reduced, and the video signal composed of the fields remaining without being deleted is compression-encoded.

[0020]

BEST MODE FOR CARRYING OUT THE INVENTION A video encoding apparatus according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 shows a block configuration of a video encoding device according to this embodiment. In FIG. 1, the input video signal is stored in the field memory 12 after being subjected to spatial filtering processing for reducing noise in the spatial filter 11.
The field memory 12 holds a plurality of consecutive image data for each of the first field and the second field. Inter-field correlation value calculator 1
Reference numeral 3 acquires field image data of temporally adjacent frames from the field memory 12, and calculates an inter-field correlation value for each corresponding partial area and corresponding composite partial area as shown in FIG. In FIG. 2, 1
Each area numbered from 1 to 9 corresponds to a partial area,
An area constituted by the first to ninth partial areas corresponds to a composite partial area. The redundant field determination unit 14 determines a threshold for determination using the inter-field correlation value of the composite partial area calculated by the inter-field correlation value calculation unit 13, and determines whether each composite partial area is redundant. To do. Here, when the inter-field correlation values in all the partial areas forming a certain composite partial area are equal to or less than the threshold value, it is determined that the corresponding composite partial area is locally redundant. As a result of determining whether all the composite partial areas forming the field are redundant, if all the composite partial areas are determined to be redundant, it is determined that the corresponding field is a redundant field. The redundant field deletion unit 15 deletes the field determined by the redundant field determination unit 14 as a redundant field from the field memory 12. The encoder 16 receives the video signal composed of the fields remaining without being deleted by the redundant field deleting section 15 as input, performs compression coding, and outputs a bit stream.

Hereinafter, a method of calculating the inter-field correlation value in the inter-field correlation value calculation section 13 will be described. With respect to the image data of the first field stored in the field memory 12, the difference data Ds1 (x, y) between the adjacent frames is obtained for each partial region by (Equation 1), and the average value As1 of the difference data Ds1 and Variance value Vs
1 is calculated by (Equation 2) and (Equation 3).

[Equation 1]

[Equation 2]

[Equation 3]

Here, Fs1 (t, x, y) is the pixel value at the time t and coordinates (x, y) in the partial area of the first field, and Ns is the number of pixels forming the partial area.
The absolute values | As1 | and Vs1 of As1 calculated by (Equation 2) and (Equation 3) are used as inter-field correlation values in the partial area of the first field.

Next, the difference data Dm1 (x, y) between adjacent frames is calculated for each composite partial area by the following (Equation 4),
The average value Am1 and the variance value Vm1 of the difference data Dm1 are calculated by (Equation 5) and (Equation 6).

[Equation 4]

[Equation 5]

[Equation 6]

Here, Fm1 (t, x, y) is the time t in the composite partial area of the first field and the coordinates (x, y).
, And Nm is the number of pixels forming the composite partial area. Am calculated by (Equation 5) and (Equation 6)
The absolute values | Am1 | and Vm1 of 1 are used as inter-field correlation values in the composite partial area of the first field.

Similarly, with respect to the image data of the second field stored in the field memory 12, the difference data Ds2 (x, y) between adjacent frames is obtained for each partial area by (Equation 7), and the difference data is obtained. Average value As2 of Ds2
And the variance value Vs2 is calculated by (Equation 8) and (Equation 9).

[Equation 7]

[Equation 8]

[Equation 9]

Here, Fs2 (t, x, y) is the pixel value at the time t and the coordinate (x, y) in the partial area of the second field. The absolute values | As2 | and Vs2 of As2 calculated by (Equation 8) and (Equation 9) are used as inter-field correlation values in the partial area of the second field.

Next, the difference data Dm2 (x, y) between adjacent frames is obtained by (Equation 10) for each composite partial area, and the average value Am2 and the variance value Vm of the difference data Dm2 are calculated.
2 is calculated by (Equation 11) and (Equation 12).

[Equation 10]

[Equation 11]

[Equation 12]

Here, Fm2 (t, x, y) is the time t and the coordinate (x, y) in the composite partial area of the second field.
Is the pixel value of. The absolute value | Am2 | and Vm2 of Am2 calculated by (Equation 11) and (Equation 12) is the second value.
It is used as the inter-field correlation value in the composite partial area of the field.

Hereinafter, a method of determining the threshold value for determination in the redundant field determination unit 14 and a method of determining whether the field is a redundant field will be described. First, the threshold value used to determine whether the composite partial area of the first field is locally redundant is calculated by using the inter-field correlation value in the composite partial area of the same position of the second field (Equation 1).
Calculated according to 3). Similarly, the threshold value used when determining whether the composite partial area of the second field is locally redundant is calculated by (Equation 14) using the inter-field correlation value in the composite partial area of the same position of the first field. To do.

[Equation 13]

[Equation 14]

Here, THm1_v and THm2_v
Is a threshold value for the variance value in the composite partial area of the first field and the second field, and THm1_
a and THm2_a are thresholds for the average values in the composite partial areas of the first field and the second field, respectively. Further, K_v, C_v, K_a, and C_a are constants, and in this embodiment, K_v = 0.75.
C_v = 3, K_a = 0, and C_a = 1. In (Equation 13) and (Equation 14), the inter-field correlation value of the second field is used when obtaining the threshold value of the first field, and the inter-field correlation value of the first field is used when obtaining the threshold value of the second field. Therefore, relative comparison determination is performed between the first field and the second field.

Next, using the threshold values determined in (Equation 13) and (Equation 14), it is determined whether or not each composite partial region is locally redundant by (Equation 15) and (Equation 16). . For each composite partial area of the first field, in all the partial areas constituting the composite partial area (Equation 15)
When the condition of is satisfied, it is determined that the corresponding composite partial area of the first field is locally redundant. Similarly, for each composite partial area of the second field, if the condition of (Equation 16) is satisfied in all the partial areas constituting the composite partial area, the corresponding composite partial area of the second field is locally redundant. It is determined that

[Equation 15]

[Equation 16]

Finally, when it is determined that all the composite partial areas of the first field are redundant, it is determined that the first field is a redundant field, and all the composite partial areas of the second field are redundant. If it is determined that the second field is a redundant field.

In this embodiment, the inter-field correlation value of the second field is used when obtaining the threshold value of the first field, and the inter-field correlation value of the first field is used when obtaining the threshold value of the second field. However, the inter-field correlation value of the first field may be used when obtaining the threshold value of the first field, and the inter-field correlation value of the second field may be used when obtaining the threshold value of the second field.

The field does not necessarily have to be composed of a plurality of composite partial areas, and may be composed of one composite partial area. The composite partial area is not necessarily composed of a plurality of partial areas. It is not necessary and may be composed of one partial area.

Further, in the present embodiment, a spatial filter for reducing noise in the input image is incorporated, but the spatial filter may be incorporated immediately before the inter-field correlation value calculating section, or the spatial filter may not be used. Good.

[0036]

As described above, according to the present invention, when a redundant field is detected, it is judged for each partial area of the field whether or not the field is redundant, so that only a part of the screen is moving. With respect to, also, an advantageous effect that an accurate inverse telecine conversion can be realized without erroneously detecting a redundant field is obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a video encoding device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a partial area and a composite partial area which are units for calculating an inter-field correlation value.

FIG. 3 is a diagram showing telecine conversion of a video signal.

FIG. 4 is a diagram showing inverse telecine conversion of a video signal.

[Explanation of symbols]

11 Spatial filter 12 field memory 13 Inter-field correlation value calculator 14 Redundant field determination unit 15 Redundant field deletion part 16 encoder

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Kichiichiro Kashiwagi             1006 Kadoma, Kadoma-shi, Osaka Matsushita Electric             Sangyo Co., Ltd. F-term (reference) 5C059 LA06 LB07 TA00 TB05 TC00                       TC42 TD03 TD04 TD11 UA02                 5C063 BA03 CA05 CA12

Claims (15)

[Claims] 1. A video encoding device which receives a video signal as an input and compression-encodes the video signal, divides a field constituting the video signal into one or more partial areas, and responds between adjacent frames. An inter-field correlation value calculating means for calculating an inter-field correlation value for each partial area,
Redundant field determining means for determining whether the field is a redundant field according to the inter-field correlation value calculated for each of the partial regions, and the video signal based on the result determined by the redundant field determining means. A video coding apparatus, comprising: redundant field deleting means for deleting a redundant field. 2. A video encoding device for inputting a video signal and compressing and encoding the video signal, wherein at least one noise reduction means for reducing noise of the video signal and at least one field constituting the video signal. Divided into sub-regions of
Inter-field correlation value calculation means for calculating an inter-field correlation value for each corresponding partial area between adjacent frames, and determining whether the field is a redundant field according to the inter-field correlation value calculated for each partial area And a redundant field deleting means for deleting the redundant field from the video signal based on the result of the judgment made by the redundant field judging means. 3. The inter-field correlation value calculation means is 1
2. A composite partial area composed of one or more partial areas is set, and an inter-field correlation value is calculated for each of the partial area and the composite partial area.
The video encoding device according to any one of 2 above. 4. The redundant field determining means determines whether or not each of the composite partial areas is redundant, and determines that the field is a redundant field when all the composite partial areas are determined to be redundant. The video encoding device according to claim 3, wherein 5. The redundant field determining means determines that the composite partial area is redundant when the inter-field correlation values in all the partial areas forming the composite partial area are equal to or less than a predetermined threshold value. The video coding apparatus according to claim 4, wherein the video coding apparatus is a video coding apparatus. 6. The redundant field determining means determines a threshold value by using the inter-field correlation value of the composite partial area calculated by the inter-field correlation value calculating means, and determines all thresholds of the composite partial area. The video encoding device according to claim 4, wherein the composite partial area is determined to be redundant when the inter-field correlation value in the partial area is equal to or less than the threshold value. 7. The inter-field correlation value calculating means calculates the inter-field correlation value by statistically processing the inter-field difference information of image data for each of the partial regions. The video encoding device according to any one of claims. 8. The inter-field correlation value calculating means calculates the inter-field correlation value by statistically processing inter-field difference information of image data for each of the partial region and the composite partial region. Items 3, 4, 5,
6. The video encoding device according to any one of 6 above. 9. The inter-field correlation value calculation means statistically processes inter-field difference information of image data for each of the partial areas to obtain a variance value and an average value, and calculates the variance value and the average value between the inter-field values. The video encoding device according to claim 7, wherein the video coding device is a correlation value. 10. The inter-field correlation value calculation means comprises:
It is characterized in that the inter-field difference information of the image data is statistically processed for each of the partial area and the composite partial area to obtain a variance value and an average value, and the variance value and the average value are used as the inter-field correlation value. The video encoding device according to claim 8. 11. A transmission means for transmitting compression-encoded video data is provided.
The video encoding device according to any one of 0. 12. A recording means for recording compression-encoded video data is provided.
1. The video encoding device according to any one of 1. 13. The video encoding apparatus according to claim 12, further comprising a recording medium for recording the compression-encoded video data. 14. The video encoding apparatus according to claim 1, further comprising a decoding unit that decodes the compression-encoded video data. 15. The video coding apparatus according to claim 14, further comprising a decoding unit that decodes the compression-coded video data recorded on the recording medium.