JP2002300586A - Method and apparatus for inserting electronic watermark information for image signal for coding - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent deterioration of quality in a compression coded signal caused by inserting electronic watermark information into an image signal. SOLUTION: An inverse telecine conversion section 13, an electronic watermark information inserting section 13, and an image compression coding section are arranged in this order. The specific unit of the moving image signal is deleted for generating an image signal, for compression coding to the moving image signal that is inputted continuously. After that, electronic watermark information is inserted to the image signal for compression coding. Since the same image signal is deleted in the moving image signal, before inserting the electronic watermark information into the moving image signal, malfunctions in compression coding due to insertion of the electronic watermark information can be prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for inserting digital watermark information into a moving image signal and compressing and encoding the digital watermark information, and to an image compression method and apparatus. In particular, the present invention is effective when compression-encoding a moving image signal obtained by converting a film material into a video signal.

[0002]

2. Description of the Related Art At present, there are the following types (A), (B) and (C) for processing moving image signals.

(A) As a method for compressing a moving image signal,
International Standardized MPEG (Moving Picture Coding Expe
rt Group) -1 and MPEG-2. These are the so-called M
This is called the PEG method. In the MPEG system, an I picture (Intra-coded picture) that performs compression encoding using only information in a field of an image signal, and a P picture (Predictive) that performs compression encoding with reference to information of a field of a past image signal. -coded picture) and a B picture (Bidirectionally predictive-code) that performs compression encoding using information of past and future image signal fields.
d picture).

(B) From the standpoint of copyright protection, there is a process of inserting a special signal (herein, referred to as digital watermark information) into a moving image signal in order to perform copy protection and copy generation management. .

(C) Further, movie images (24 per second)
Frame) to a television signal (60 fields per second)
There is a form in which a so-called 2-3 pull-down process is performed in order to convert the data into. Conversely, there is a so-called inverse-telecine conversion process of returning such a television signal to a 24-frame image signal in order to perform compression encoding.

[0006]

By the way, when a moving image signal is compression-encoded, there is no conventional system which comprehensively considers the above-mentioned processes (A), (B) and (C). The inventor has noticed that the quality of a compression-encoded signal deteriorates if the procedure of each of the processes (A), (B), and (C) is randomly determined.

In particular, when digital watermark information is inserted, the inverse telecine operation becomes unstable, the generated code amount control suitable for the picture type becomes unstable, and the image quality deteriorates. In the variable bit rate coding, the generated code amount is There is a problem that control is performed irrespective of the characteristics of the signal.

Accordingly, the present invention has been made to solve the above-mentioned problem, and stabilizes inverse telecine operation, stabilizes generated code amount control suitable for a picture type, and realizes appropriate compression encoding. It is an object of the present invention to provide a method and apparatus for inserting digital watermark information and a method and apparatus for compressing an image.

[0009]

According to the present invention, in order to achieve the above object, a predetermined unit of a moving image signal is deleted from a continuously input moving image signal to perform compression encoding. , And thereafter, digital watermark information is inserted into the image signal for compression encoding.

By this means, prior to inserting the digital watermark information into the moving picture signal, the moving picture signal is compressed and encoded by inserting the digital watermark information because the same picture signal is deleted. Malfunction can be prevented.

[0011]

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

FIG. 1 is a block diagram showing an embodiment of the present invention. An image signal is input to the input terminal 11. This image signal is a movie image (24 frames per second)
Is converted into a television signal (60 fields per second). This image signal is first input to the inverse telecine conversion unit 12. The inverse telecine conversion unit 12 converts the field frequency 1/60 sec into the original 24
This is a part for converting into a frame image signal. This is to perform efficient compression and appropriate compression when performing compression encoding.

The operation of the inverse telecine conversion unit 12 will now be described with reference to FIG. First, FIG.
Means a movie image (film image signal). A movie image has 24 frames per second, and there is a period in which one film image signal is repeatedly used twice in order to convert it into an image signal of 1/60 seconds.

FIG. 2B shows a state where the film image is converted into an image signal having a field frequency of 1/60 second. In FIG. 2B, the upper part shows the top field, and the lower part shows the bottom field. Film image signal 0 shows a state where it is used twice in the top field and once in the bottom field. Further, the film image signal 1 is used once in the top field and once in the bottom field. Film image signal 2
Is used once in the top field and twice in the bottom field. The film image signal 3 is
Used once in the top field and once in the bottom field. The film image signal 4 is used twice consecutively in the top field and once in the bottom field.

FIG. 2C shows that the image signal shown in FIG.
The state after inverse telecine conversion is shown. No.
An image signal (field signal) selected from the top field is arranged in one field, and an image signal (field signal) selected from the bottom field is arranged in the second field.

The inverse-telecine conversion unit 12 determines whether the same field signal has been input repeatedly or a field signal without repetition has been input, and obtains a field signal having an arrangement as shown in FIG. 2C. This is, for example,
The number of coincidence detections of the pixel at the corresponding position between the previous field and the subsequent field is counted, and this field is repeated according to the percentage (85% or 90%) of the number of pixels of the entire field. Is determined.

The output of the inverse telecine conversion unit 12 is input to a digital watermark information insertion unit 13. The digital watermark information insertion unit 13 multiplexes, for example, watermark information having a specific meaning to a part of image information. There are various methods for inserting digital watermark information, and there is no particular limitation here. As a method of inserting digital watermark information, for example, there is a method of scattering a meaningful code over the entire field, a method of inserting a code at a specific position of the field, and the like. In any case, the code is inserted to the extent that the quality of the image is not degraded. There are various methods of using the digital watermark information. For example, there is a method of extracting digital watermark information and setting it as copy control information, or a method of judging whether or not the digital watermark information has been illegally copied based on the presence or absence of the digital watermark information.

The image signal into which the digital watermark information is inserted is
The image is input to a motion detector 14 for detecting the motion of an image and an image rearranger 15. The image rearranging unit 15 is required to perform the following image processing. That is, in the encoding and compression processing, an I-picture (Intra-cod) for performing compression and encoding using only information in a field of an image signal is used.
ed picture) and a P picture (Predictive-co
ded picture) and B picture (Bidirectionally predictive-coded picture) that performs compression encoding using past and future image signal field information.
There are two types of coded pictures. This is because it is necessary to change the image order of the input image signal.

In the coding of the I picture, the image signal from the image rearranging unit 15 is directly converted into a DCT signal through a subtractor 16.
(Discrete cosine) is supplied to the conversion unit 17. Here, the image signal is converted into a spatial frequency component. This DCT converter 17
Is input to the quantizer 18. At the time of encoding the I picture, a signal of zero is supplied to the other end of the subtractor 16. The quantizer 18 quantizes the spatial frequency component to reduce the code amount.

The signal quantized by the quantizer 18 is input to a grammar generator 23 and an inverse quantizer 23. The inverse quantizer 18 reverses the quantized output of the quantizer 18, and the output is further input to an inverse DCT transform unit 20. From the inverse DCT converter 20, a signal in a state before being input to the DCT converter 17 is obtained.

At the time of encoding an I picture, the output of the inverse DCT unit 20 returns to the original image signal, and is stored in the motion compensation image memory 22 via the adder 21 as a reference image signal.

Next, encoding of a B picture or a P picture will be described.

The output of the digital watermark information insertion unit 13 is also input to the motion detection unit 14. In the motion detection unit 14,
A motion part of the image is detected to generate motion vector information.
This motion vector information is supplied to the motion compensation image memory 22. The type information from the type determination unit 24 is also input to the motion compensation image memory 22. Therefore,
The motion-compensated image memory 22 generates a motion-compensated reference image signal using motion vector information according to the type of I-picture, B-picture, or P-picture processing. The motion-compensated reference image signal is input to the subtractor 16.

Thus, the subtracter 16 performs a subtraction process between the image signal from the image rearranging section 15 and the reference image signal from the motion compensation image memory 22. Subtractor 1
6, a difference signal (prediction error signal) is obtained. Thereafter, DCT processing and quantization processing are performed on the prediction error signal.

In the coding of P pictures and B pictures, compression coding is performed only on the prediction error signal. Therefore, the amount of code generated in the grammar generation unit 23 is significantly reduced as compared with the time of I-picture processing. The grammar generating section 23 is a section for adding additional information including type information and motion vector information to the encoded information.

As described above, the amount of code generated at the output of the quantizer 18 varies greatly depending on the coding type of the image signal. For this purpose, an encoded data buffer memory 25 is provided at a stage subsequent to the grammar generating section 23. This coded data buffer memory 25 is a part for smoothing the generated code amount. Thus, the fixed-rate data is output to the output terminal 26. Although not described here, a similar encoded data buffer memory is also provided at the input stage of the MPEG decoder that decodes and expands the compressed encoded signal. This data buffer memory absorbs fluctuations in the amount of consumed data generated in the decoding and decompression processing.

Further, when no image motion is detected by the motion detecting section 14, the amount of codes generated in the P picture processing and the B picture processing increases. For this reason, the coded data buffer memory 25 alone may not be able to sufficiently smooth variations in the generated code amount. For this reason, the output of the grammar generation unit 23 is also input to the code amount control unit 27.
The code amount control unit 27 monitors the code amount of the information output from the grammar generation unit 23, and adjusts the degree of quantization of the quantizer 18 according to the result. As a result, the generated code amount is adjusted so that the encoded data buffer memory 25 does not overflow. Adjustment of the degree of quantization is performed by switching the quantization table.

The above-described characteristic configuration of the present invention is characterized in that, first, an inverse video
In the telecine conversion unit 12, a predetermined unit of the moving image signal is deleted to generate an image signal for compression encoding. Next, electronic watermark information is inserted into the image signal for compression encoding. The point is that the procedure is adopted.

The inverse telecine conversion unit 12 detects whether a field having the same content has repeatedly arrived twice consecutively, and if a field of the same content has arrived twice repeatedly, one of them is selected. To adopt. In this detection processing, a coincidence for each pixel is determined with respect to a continuously arriving odd field or a continuously arriving even field, and if this coincidence number exceeds a predetermined value, it is determined that the field is a repetitive field. Has been determined.

If digital watermark information is inserted before the inverse telecine conversion, it may not be possible to detect a repeated field. Therefore, unlike the apparatus of the present invention, the inverse telecine conversion process is first performed on the input image signal, and then the digital watermark information is inserted, so that no problem occurs in the inverse telecine conversion process. .

Then, high-quality compression encoding can be performed on the image signal into which the digital watermark information is inserted.

The present invention is not limited to the above embodiment.

In the above example, 60 fields / second is set to 24
It was converted to a frame image (24 fields / second). The present invention is not limited to this. The present invention provides a field thinning means 30 for further converting a field converted into an image of 24 frames into an image signal of 12 frames (12 fields / second).
It may be provided after the telecine conversion unit 12.

FIG. 3 shows the embodiment. Other parts are the same as those in the embodiment of FIG.

FIG. 4 shows the progress of field conversion according to this embodiment. FIG. 4A indicates a movie image (film image signal). Movie images are 2 per second
It is 4 frames, and to convert it to a 1/60 second image signal,
There is a period in which one film image signal is used twice.

FIG. 4B shows how the film image is converted into an image signal having a field frequency of 1/60 second. In FIG. 4B, the upper part shows the top field, and the lower part shows the bottom field. Film image signal 0 shows a state where it is used twice in the top field and once in the bottom field. Further, the film image signal 1 is used once in the top field and once in the bottom field. Film image signal 2
Is used once in the top field and twice in the bottom field. The film image signal 3 is
Used once in the top field and once in the bottom field. The film image signal 4 is used twice consecutively in the top field and once in the bottom field.

FIG. 4C shows that the image signal shown in FIG.
The state after inverse telecine conversion is shown. No.
An image signal (field signal) selected from the top field is arranged in one field, and an image signal (field signal) selected from the bottom field is arranged in the second field.

The inverse telecine conversion unit 12 determines whether the same field signal has been input repeatedly or a field signal without repetition has been input, and obtains the field signals arranged as shown in FIG. 4C. This is, for example,
The number of coincidence detections of the pixel at the corresponding position between the previous field and the subsequent field is counted, and this field is repeated according to the percentage (85% or 90%) of the number of pixels of the entire field. Is determined.

Further, by the field thinning means 30,
A field image signal is selected once in two fields, as shown in FIG. This signal is supplied to the image compression unit. That is, the image signal of 24 frames / sec is further converted to an image signal of 12 frames / sec.

This is effective for reducing the bit rate by reducing the number of images to be compression-coded, for example, in an application in which the bit rate at the time of coding is restricted. Digital watermark information is inserted into the digital watermark information insertion unit 13 into the image signal of 12 frames / second.

Here, the image signal of 24 frames / sec is further converted to an image signal of 12 frames / sec. However, for example, the image signal is further converted to an image signal of 8 frames / sec and 6 frames / sec. Is still within the scope of the present invention.

[0042]

As described above, according to the present invention, the inverse telecine operation can be stabilized, the generated code amount control suitable for the picture type can be stabilized, and appropriate compression coding can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram showing one embodiment of the present invention.

FIG. 2 is an operation explanatory diagram illustrating an operation of an inverse telecine conversion unit;

FIG. 3 is a block diagram showing another embodiment of the present invention.

FIG. 4 is an operation explanatory diagram shown to explain an operation of a main part of the embodiment of FIG. 3;

[Explanation of symbols]

6: Digital watermark information superimposing device, 12: Inverse telecine conversion unit, 14: Motion detection unit, 15: Image rearrangement unit, 16: Subtractor, 17: DCT conversion unit, 18: Quantizer, 19: Inverse quantum Transform unit, 20: inverse DCT transform unit, 21: adder, 22: motion compensated image memory, 23: grammar generating unit,
24: Type determination unit, 25: Encoded data buffer memory, 27: Code amount control unit

 ──────────────────────────────────────────────────続 き Continued on front page F term (reference) BC08 BC16 BD01

Claims (4)

[Claims] 1. A means for generating an image signal for compression encoding by removing a predetermined unit of a moving image signal from a continuously input moving image signal; Means for inserting digital watermark information into an image signal. A digital watermark information inserting device for an encoding image signal. 2. The moving image signal which is continuously inputted,
Means for converting a film image signal photographed at a rate of 24 frames / sec or less into 60 fields / sec, wherein the means for generating an image signal for compression-encoding is based on the input image signal; An apparatus for inserting digital watermark information for an image signal for encoding, characterized in that a field signal of a part of the image signal of 60 fields / sec is removed to generate a film image signal. 3. The means for removing a field signal of a part of the 60-field / second image signal compares image signals of continuous odd-numbered fields or continuous even-numbered fields of the input moving image signal. 3. The apparatus for inserting digital watermark information for an image signal for encoding according to claim 2, wherein removal of the field signal is determined according to a result. 4. An input image signal obtained by converting a film image signal photographed at 24 frames / second or less into 60 fields / second is supplied, and a continuous odd field or a continuous even number of the input moving image signal is supplied. The input image signal of the field is compared, the removal of the field signal is determined in accordance with the comparison result, the image signal is converted into a restored image signal having a frequency of 24 sheets / sec or less, and the restored image signal is A digital watermark information insertion method for encoding image signals, characterized in that the digital watermark information is inserted into the compression encoding unit.