JP2001224029A - Image compression coder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image compression coder that can obtain a compression coded image of a film source with high quality by optimizing a data quantity remaining in an input buffer memory of a decoder just before decoding an I picture in the case of coding in compliance with the MPEG-2 video standards after applying inverse telecine processing to a telecine image signal to eliminate repeating fields. SOLUTION: An inverse telecine conversion circuit 101 gives information denoting the presence/absence of repeating fields to a type decision circuit 112. The type decision circuit 112 controls a coded picture type on the basis of this information so that a B picture before an I picture is an image signal with repeating fields. Thus, a display time of the B picture is a time equivalent to 3 fields, data remaining in an input buffer memory at a decoder just before decoding the I picture are increased so as to increase a generated code quantity of a noted I picture.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an ISO / IEC
MPEG-2 (moving pic
The present invention relates to an image compression encoding apparatus that compresses and encodes a moving image signal in accordance with the Video standards of the ture experts group 2).
More particularly, the present invention relates to a device for compressing and encoding a telecine image signal obtained by converting a moving image captured by a film into a signal with high quality.

[0002]

2. Description of the Related Art Recently, as a method for compressing moving images, MP
A method of compressing and encoding a moving image signal in accordance with the EG-2 Video standard has been developed. In the following description, a compression encoding device according to the above method is referred to as an MPEG encoder, and a device for returning an output signal from the MPEG encoder to an original moving image signal is referred to as an MPEG decoder.

It has been considered that a telecine image signal obtained by converting a film image of a movie or the like into a signal using the above-mentioned MPEG encoder is compression-coded with high image quality and used as a material to be provided for digital television broadcasting or the like. Telecine image signals are usually shot at 24 frames per second, so when connecting the signals picked up from each frame of the film, the same image signal is repeated every other frame for 60 fields per second. (3: 2 telecine signal). As described above, since the telecine image signal includes an overlapped field, the MP
In the EG encoder, to increase its efficiency,
At the first stage, duplicate fields are removed from the telecine image signal to return to the original 24 frames per second video signal. This processing is called inverse telecine processing.

[0004] In the inverse telecine processing, repeated fields are removed. Therefore, in the MPEG decoder, it is necessary to insert the repetition of the field, which is the reverse processing. Therefore, in the MPEG encoder,
Whether the coded signal has a repetition of fields,
In addition, a signal indicating whether the compression-encoded image starts in the top field or the bottom field is added. In the MPEG decoder, a decoded image signal similar to the original telecine image signal can be obtained by inserting a field based on the additional signal.

[0005] In the MPEG-2 Video standard,
An I picture (Intra-coded picture) for which compression encoding is performed using only information in an image, and a P picture (Predictive-code pi) for which compression encoding is performed by referring to information of a past image signal.
c), and a B picture (Bi-directionallypr) for which compression encoding is performed using information of past and future image signals.
There are three types of coded pictures (edictive-coded pictures).

[0006] In order to improve the quality of a compressed image, it is necessary to allocate a large amount of code to an I picture to be compression-coded using only information in the image. On the other hand, an encoded data buffer memory is inserted in the input stage of the MPEG decoder, and it is necessary to control the generated code amount of each image signal so that the buffer memory does not underflow. Therefore, it is effective to increase the amount of data remaining in the coded data buffer memory arranged at the input stage of the MPEG decoder as much as possible immediately before decoding the I picture.

However, in the image signal subjected to the inverse telecine processing, since the repeated fields are removed, the encoding intervals of each field are not equal. In the conventional MPEG encoder, since there is no relation between the field interval after the inverse telecine processing and the determination of the type of the I picture, the P picture, and the B picture, the input of the MPEG decoder immediately before the decoding of the I picture is performed. It has been difficult to perform an operation for maximizing the amount of data remaining in the encoded data buffer memory arranged in a row.

[0008]

As described above, in the conventional image coding and compression apparatus based on the MPEG-2 system, when coding and compressing a telecine image signal, inverse telecine processing is performed to repeat a field. Despite the encoding process after removal, the inverse
There is no relation between the field interval after the telecine processing and the determination of the type of the I picture, the P picture, and the B picture. For this reason, it is difficult to perform an operation to increase the amount of data remaining in the encoded data buffer memory arranged at the input stage of the MPEG decoder as much as possible immediately before the decoding of the I-picture. The compression efficiency was not sufficient.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-described problems. Even when encoding is performed after performing inverse telecine processing on a telecine image signal and removing repetitive fields, decoding of an I picture is also possible. Immediately before encoding, the amount of data remaining in the encoded data buffer memory arranged at the input stage of the MPEG decoder can be optimized, and a high-quality compressed and encoded film material image can be obtained. It is intended to provide a device.

[0010]

To achieve the above object, an image signal compression encoding apparatus according to the present invention comprises:
/ IEC13818-2 (MPEG-Video) standard, in which a telecine image signal in which fields are repeatedly inserted according to a predetermined pattern is compression-encoded, and the repetition of removing a repetitive field in the telecine image signal is performed. A field removing unit, and an I picture (Intra-coded picture), a P picture (Predictive-code picture), B-picture (Bi-dire
coded picture type determining means for deciding which type of coded picture is to be a coded picture (ctionally predictive-coded picture), and for each field output from the field removing means based on the type determined by this means. Compression encoding means for compressing and encoding the image signal in accordance with the standard, wherein the encoded picture type determining means converts the image signal having the repetition field into an encoded picture immediately before the I picture. It is characterized by being a type.

[0011]

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

(First Embodiment) FIG. 1 shows an image compression encoding apparatus according to a first embodiment of the present invention as
FIG. 2 is a block diagram illustrating a configuration in a case where a telecine image signal is compression-coded according to the 2 video standard. Here, a terminal 100 receives a telecine image signal.
The telecine image signal is pre-processed by the inverse telecine conversion circuit 101 and then input to the image rearrangement circuit 103.

As described above, MPEG-2 video
In the standard, there are three types of coded pictures: I picture, P picture, and B picture. These encoding types are indicated by the type determination circuit 112. Since the B picture needs information of a future image signal, the image rearrangement circuit 103 changes the image order of the input signal.

Here, compression coding of an I picture will be described. In this case, the image signal of the I picture output from the image rearrangement circuit 103 is supplied to a DCT (discrete cosine) conversion circuit 105 as it is via a subtractor 104 and is converted into a spatial frequency component. 106
Supplied to At this time, a zero signal is input to the other end of the subtractor 104. The quantizer 106 reduces the code amount by quantizing the signal changed only by the spatial frequency component, and the output is supplied to a grammar generating circuit 113, where the coding is performed according to the MPEG standard.

On the other hand, the quantized signal output from the quantizer 106 is inversely quantized by an inverse quantizer 107 in order to obtain a reference image signal when a B or P picture is created.
The inverse DCT transform is performed by the inverse DCT
It is returned to the data before CT conversion. In the compression encoding of the I picture, the output of the inverse DCT transformer 108 returns to the original image signal.
10 is stored as a reference image.

Next, compression encoding of a B or P picture will be described. In this case, the output of the inverse telecine conversion circuit 101 is also input to the motion detection circuit 102. The motion detection circuit 102 searches for a motion vector from the input image signal. The motion vector detected here is supplied to the motion compensation image memory 110.

The motion-compensated image memory 110 receives the type indication signal output from the type determination circuit 114 and generates a motion-compensated reference image signal using a motion vector corresponding to the type of the coded picture. . The motion-compensated reference image signal is supplied to a subtractor 104, and the subtractor 104 generates an image rearrangement circuit 103.
Is subtracted from the input image signal to obtain a difference signal (prediction error signal). After that, DCT processing and quantization processing are performed on the prediction error signal in the same manner as in the P picture.

In general, in the coding of a P or B picture, since only the prediction error signal is subjected to compression coding, the code amount generated by the grammar generating circuit 113 is significantly reduced as compared with the I picture. Will be done. As described above, the amount of code generated depending on the coding type of the image signal greatly varies. For this reason, an encoded data buffer memory 116 is provided at the subsequent stage of the grammar generation circuit 113. This buffer memory 116 smoothes the generated code amount and makes the output data a fixed rate.
The output is output from the output terminal 117 as encoded data at a fixed rate.

The input stage of the MPEG decoder for decoding and decompressing the compressed and encoded signal generated by the above processing is also provided in the same encoded data buffer memory as the encoded data buffer memory provided in the output stage of the MPEG encoder. A memory is provided to absorb fluctuations in the amount of data consumed in the decoding and decompression processing.

When a motion vector is not detected in the motion detection circuit 102, P or B
The amount of code generated in a picture also increases, and the generated code amount may not be sufficiently smoothed by the coded data buffer memory 116 alone. For this purpose, the output of the grammar generation circuit 113 is also input to the code amount control circuit 115, and the code amount control circuit 115 monitors the code amount generated by the grammar generation circuit 113 and monitors the degree of quantization in the quantizer 106. Is adjusted so that the encoded data buffer memory 116 does not overflow.

Here, the inverse telecine conversion circuit 10
Operation 1 will be described with reference to FIG.

First, a film is usually photographed at 24 frames per second, and an image signal picked up from the film is obtained at the timing shown in FIG. An image signal obtained by telecine processing this image signal is 1
In order to obtain an image signal of 60 fields per second, FIG.
As shown in (b), in each of the top field and the bottom field, a signal in which a field is repeated and a signal without repetition are alternately arranged (here, this is referred to as a 3: 2 telecine signal). The inverse telecine conversion circuit 101 detects this repeated field, converts it into the original 24 frame image signal, and outputs it to the image rearrangement circuit 103 and the motion detection circuit 102. This processing is called inverse telecine processing.

The image signal from which the repetitive field has been removed by the inverse telecine processing is subjected to image signal rearrangement by the image rearrangement circuit 103 in accordance with the instruction of the type determination 112, and compression encoding is performed by the subtractor 104 and thereafter. Is done. The output is shown in FIG. At this time, since the repetition field has been removed by the inverse telecine processing, it is necessary to insert the repetition of the field, which is the reverse processing described above, in the MPEG decoder. Therefore, a signal indicating whether or not the field has been repeated and whether the compression-encoded image has started in the top field or in the bottom field are added to the encoded signal. M
The PEG decoder can obtain a decoded image signal similar to the original telecine image signal by inserting a field based on the information of the additional signal.

Further, by removing the repetitive fields by the inverse telecine processing, the interval between the compression encodings is two fields or three fields as shown in FIG. 2D. In FIG. 2D, P
0, B2, B4, P6, and B8 are images having a repeated field, and the other images are images at an interval of two fields. The image from which the repetition field has been removed is compression-encoded and, via the encoded data buffer memory 116,
The data is input to an encoded data buffer memory of an MPEG decoder (not shown). In the decoder, decoding is performed in the order of compression encoding.

In the example of FIG. 2, as shown in (d) and (e), encoding is performed in the order of P3 → B1 → B2 → P6 → B4 → B5 → I8. f)
The encoded data is read from the encoded data buffer memory in the order and at the time shown in (1), and the decoding process is performed. In this decoding processing, the reverse of the image rearrangement performed on the encoder side is performed, and the display order of the image signals is changed from the original P0 to the original.
B1 → B2 → P3 → B4 → B5 → P6 ...

Here, in the decoder, the presence or absence of the repetition field and the additional signal indicating whether the compression-encoded image starts in the top field or the bottom field are determined for each image. Insert a repeating field. As a result, the decoded image has the same repetition field as the telecine image signal input to the encoder.

Next, a method of adjusting the code amount in the MPEG encoder will be described in detail. The purpose of this code amount adjustment is to prevent the encoded data buffer memory on the decoder side from underflow (there is no data to be decoded). As shown in FIG. 2 (h), writing to the coded data buffer memory in the decoder is performed at a fixed rate, and all data related to the decoding start time of each image is instantaneously read from this buffer memory.

Referring to the example of FIG. 2 (h), the decoding start time is equal to the image display start time.
At p6, that is, at the display start time of the image P3, all data of the image P6 corresponding to a is read from the decoder coded data buffer memory, and thereafter, data is written at a fixed rate during the display period of the image P6. Increase with a slope of. The reading time of the next image B4 is determined by the display image P3. In this case, since P3 is an image without a repetition field, the data of the image B4 becomes the decoding start time two fields later. Since the image P0 has a repeated field, the reading time of the image B1 comes after three fields.

As described above, the decoding start intervals of the compressed and coded images subjected to the inverse telecine processing are not uniform. In controlling the amount of generated codes in the encoder, it is necessary to control the occurrence of underflow in consideration of the fluctuation of the decoding start time and the free space of the encoded data buffer memory at the decoding start time.

On the other hand, paying attention to the quality of the compression-encoded image, for example, in order to obtain an image of the same quality, the I-picture requires the most codes, and the P-picture has the largest code amount. There are few B pictures. Further, since the I picture is used as a reference picture for the subsequent P or B picture, the amount of codes required for the P or B picture is reduced by allocating more codes, and the quality of the entire decoded picture is reduced. Also improve.

In order to increase the amount of codes generated in an I picture, the coded data buffer memory becomes as full as possible at the decoding time of the I picture on the decoder side, and the code amount of the I picture is reduced by the capacity of the buffer memory. It is effective to control so as to be close to Here, considering that the amount of data written to the coded data buffer memory of the decoder is a fixed rate, the picture before the I picture in the coding order is controlled so as to be an image having a repetitive field. What is necessary is to perform control so that the amount of data remaining in the coded data buffer memory on the decoder side at the picture decoding time is increased.

In this embodiment, taking the above into consideration, the inverse telecine conversion circuit 101 converts the type determination circuit 1
A signal indicating the presence or absence of a repeated field is output to 12;
The repetition field information of the inverse telecine output is given to the type determination circuit 112. That is, as shown in FIGS. 3A to 3H, the type determination circuit 112 repeats the B picture (B5) preceding the I picture (I9) on the basis of the repeated field information to become an image signal having a field. In this way, the coded picture type is controlled. As a result, the display time of the B picture becomes a time corresponding to three fields, the data remaining in the decoder-side coded data buffer memory at the decoding time of the I picture becomes larger than that in FIG. It is possible to increase the generated code amount. For example, at time ti8 in FIG. 2 (h), the encoded data buffer memory on the decoder side does not reach the full capacity, but in the example of FIG. 3, data is written to the full capacity as shown in (h). Therefore, more code amount can be allocated to the I picture.

As described above, according to the configuration of the first embodiment, the coding type is controlled so that the display interval of the coded picture located before the I picture requiring a large amount of code is increased. By doing so, it is possible to increase the data remaining in the coded data buffer memory on the decoder side, thereby making it possible to increase the code amount of the I picture, and to provide a high-quality compressed coded image. Can be obtained.

In the 3: 2 telecine signal shown in FIG. 3, repetitive fields occur alternately when viewed from the picture to be coded. On the other hand, in a widely used moving image compression encoding apparatus based on MPEG, the interval between I or P pictures is every three pictures,
It has a BIBBPBBP structure, and from the viewpoint of compression efficiency, the period of an I picture is generally five or more coded pictures. In addition, since the decoding cannot be performed unless the picture is an I picture, this interval cannot be so long, and generally 30 pictures or less.

Therefore, as shown in FIGS. 4A to 4C, the I picture interval is set to 6, 12, 18 (or 2).
When 4) is selected, the B picture before the I picture becomes an image signal having a repetition field each time, and the present invention is realized only by once determining the coded picture type, thereby simplifying the control. There are advantages that can be done. When controlling the I-picture interval in this way, an I-picture interval indicating signal for instructing the I-picture interval to be 6, 12, or 18 is input from the terminal 118 and supplied to the type determination circuit 112. In this case, the type may be determined according to the instruction.

(Second Embodiment) FIG. 5 is a block diagram showing a configuration of a second embodiment of an image compression encoding apparatus according to the present invention, in which a telecine phase indication signal synchronized with a telecine image signal is supplied. FIG. In FIG. 5, the same portions as those in FIG. 1 are denoted by the same reference numerals, and the description thereof will not be repeated.

In this embodiment, a telecine phase instruction signal synchronized with a telecine image signal is input from a terminal 119 and supplied to an inverse telecine conversion circuit 120. This telecine phase indication signal is a signal indicating the presence or absence of a repetitive field in the input telecine image signal and whether the image starts with a top field or a bottom field. The inverse telecine conversion circuit 120 removes repeated fields from the input telecine image signal based on the telecine phase indication signal, and outputs the signal to the image rearrangement circuit 103 and the motion detection circuit 102. As in the embodiment of FIG. 1, the type determination circuit 112 determines whether the B picture before the I picture is based on the repeated field information.
The coded picture type is controlled so that the picture becomes a picture signal having a repeated field.

According to this configuration, the inverse telecine conversion circuit 120 does not need to determine the presence or absence of a repeated field, so that the configuration can be simplified.

(Third Embodiment) FIG. 6 is a block diagram showing the configuration of a third embodiment of an image compression encoding apparatus according to the present invention when an edit point signal indicating an edit location of a telecine image signal is provided. It is a circuit diagram. In FIG. 6,
The same parts as those in FIG. 1 are denoted by the same reference numerals, and redundant description is omitted here.

In the present embodiment, the time code of the telecine image signal and the telecine phase signal are transmitted to terminals 123 and 124.
, And is supplied to the type determination circuit 122. Further, an image signal editing point signal is input from the terminal 125 and supplied to the type determination circuit 122. The image signal editing point is a point at which the telecine pattern becomes discontinuous because the film signal is telecine-converted, recorded on an image recording medium such as a VTR, and then edited. Consecutive points are indicated. If the telecine phase of an image signal sandwiched between two image signal editing points is known, except for the vicinity of the image signal editing point,
The telecine pattern of the image signal between these edit points is shown in FIG.
As shown in FIG.

Therefore, the type determination circuit 122 outputs a signal for removing a repetitive field from the input telecine image signal to the inverse telecine conversion circuit 121 based on the input time code, and simultaneously sets the encoding type to the image. By instructing the rearrangement circuit 103 and the like, the B picture before the I picture is controlled so as to be an image signal having a repeated field. Here, since the image signal is edited, as is clear from the example of FIG. 7, the coded picture is changed so that the B picture before the I picture becomes an image signal having a repeated field except for the vicinity of the edit point.・ Type can be controlled. The same can be applied to a case where there are a plurality of image editing points.

(Fourth Embodiment) FIG. 8 shows an image compression encoding apparatus according to a fourth embodiment of the present invention, in which 3: 2:
FIG. 3 is a block circuit diagram showing a configuration when processing a 2: 3 telecine image signal. 8, the same parts as those in FIG. 1 are denoted by the same reference numerals, and redundant description will be omitted.

In the first to third embodiments, an input image signal in which a repetition field is generated for each coded picture has been described. However, another method for converting a film signal into a 60-field image signal is described. There is also a method.
This is a method in which the repetition field continues twice consecutively, and thereafter there are no repetition fields in two video frames (hereinafter referred to as 3: 2: 2: 3 telecine).

With respect to the 3: 2: 2: 3 telecine image signal, by setting the I picture interval to 15 pictures, it is possible to control the B picture before the I picture to be an image signal having a repeated field. It becomes possible. Also, for a 3: 2: 2: 3 telecine image signal, for example, by inputting a field removal signal at the time of a repeated field in synchronization with the image signal,
A B picture before an I picture can be an image signal having a repeated field.

In FIG. 8, 3: 2:
A 2: 3 telecine image signal is input, and a field removal signal is input to a terminal 126 in synchronization with the 3: 2: 2: 3 telecine image signal at the time of a repeated field.
The inverse telecine conversion circuit 127 to which these signals are input removes the repetition field based on the repetition field removal signal, and outputs the removal information to the type determination circuit 128 at the same time. The type determination circuit 128 determines the coded picture type based on the removal information so that the B picture preceding the I picture becomes a repeated field.

According to this configuration, even with a 3: 2: 2: 3 telecine image signal, the same compression encoding as that of a 3: 2 telecine image signal can be realized.

(Fifth Embodiment) FIG. 9 shows a fifth embodiment of the image compression / encoding apparatus according to the present invention, in which a 3: 2: 2: 3 telecine image signal having an editing point is processed. It is a block circuit diagram showing. In FIG. 9, FIG.
The same parts as those described above are denoted by the same reference numerals, and redundant description will be omitted.

In FIG. 9, a 3: 2: 2: 3 telecine image signal in which a repetition field is repeated twice is supplied to a terminal 100, and a time code (hereinafter referred to as reference point 2) of the image signal is supplied to a terminal 129. , A telecine phase signal is supplied to a terminal 130, and a time code indicating an editing point is supplied to a terminal 131. In the type determination circuit 132,
Based on the information of the time code of the reference point 2, the telecine phase signal, and the edit point time code, the coded picture type is changed so that the B picture before the I picture becomes a repeated field with respect to the image except for the vicinity of the edit point. decide. Also,
The type determining circuit 132 converts a signal for removing a repetitive field from the input telecine image signal into an inverse telecine conversion circuit 13 based on the input time code.
Output to 3. In the inverse telecine conversion circuit 133, when receiving an instruction for repeated field removal,
A repeated field is removed from the 3: 2: 2: 3 telecine image signal and output.

According to this configuration, 3: 2:
The same processing as in the third embodiment can be realized for a 2: 3 telecine image signal.

(Sixth Embodiment) FIG. 10 is a block circuit diagram showing the configuration of a sixth embodiment of the image compression encoding apparatus according to the present invention. In FIG. 10, the same parts as those in FIG. 1 are denoted by the same reference numerals, and redundant description will be omitted.
FIG. 11 shows the inverse image used in the present embodiment.
A specific configuration of the telecine conversion circuit 134 is shown. Also,
FIG. 12 shows the processing contents of automatic detection of the telecine phase, inverse telecine processing, and determination of the coding type according to the present embodiment.

In the present embodiment, a 3: 2 telecine phase is automatically detected from an input telecine image signal, and a B signal before an I picture is detected.
The coded picture type is determined so that the picture becomes a repeated field. FIG.
, The 3: 2 telecine image signal input from the terminal 100 is supplied to the two-field delay circuits 135 and 136 and the subtractor 137. Two-field delay circuit 135
And the image signal input to the input terminal 100 are subtracted for each pixel by a subtractor 137, and are cumulatively added by a cumulative adder 138. The accumulator 138 is reset for each field, and its output is the cumulative sum of the difference between the current input image field and the image field two fields before.

Here, referring to FIG. 12, if the field images F1, F2, F3,... Are images from the same film, the difference value cumulative sum (di) of F3 and F1
f31) becomes almost zero. When the cumulative value of the difference values is equal to or smaller than a certain value, the repeated field determination circuit 139 determines that there is a repeated field. The result of the repeated field determination is supplied to the field / frame conversion circuit 140 and output to the type determination circuit 142 through the terminal 141.

The field / frame conversion circuit 140 takes in the output of the two-field delay circuit 136 and outputs an image signal converted into a write frame signal to a top-field or bottom-field memory bank. Reading is stopped for the field (F3) for which it is determined that there is a repeated field. Therefore, image signals F1, F2, F3,.
With regard to, the reading of F3 is stopped, and a frame image signal composed of F1 and F2 is output.
This readout period is equal to three field periods.

Next, since F4 and F5 and F6, F7 and F8 are signals from the same film, the cumulative sum of difference values Dif42 and Dif
53, Dif64, Dif75 are a repetition field determination circuit 139
Becomes larger than the judgment value of, and dif86 becomes smaller than the judgment value. Therefore, field F8 is determined to be a repetitive signal. In the field / frame conversion circuit 140, F4
And F5 are read and converted to a frame image and output. Next, since F6 is also read continuously, the reading of the frame signal composed of F4 and F5 takes two field periods. After F6 and F7 are read, F8 is determined to be a repeated field, so that the image composed of F6 and F7 is a signal of three field periods.

As is apparent from the above description, a pattern having a 3: 2 telecine phase (hereinafter, referred to as a telecine pattern) is automatically detected from an input image by using an accumulated value of an inter-field difference separated by two fields. be able to. The output of the repetition field determination circuit 139 is input to the type determination circuit 142 via the terminal 141. The type determination circuit 142 determines the coding type based on the repetition field determination result. For example, when the repetition field is determined as 3 → 2 → 3 → 2 → 3 → 2 as shown in this example, the coding type is set to the repetition of BIBBP as shown in FIG. As a result, the image before the second and subsequent (I8) I pictures becomes an image having a repeated field.

As described above, while automatically detecting the telecine pattern from the input image, the image before the I picture according to the present invention can be controlled to an image having a repetitive field. Although FIG. 12 describes a 3: 2 telecine image signal, the same can be applied to a 3: 2: 2: 3 telecine image signal.

(Seventh Embodiment) FIG. 13 shows, as a seventh embodiment of the image compression / encoding apparatus according to the present invention, an arrangement characterized by a method of determining an encoding type for a telecine image having an editing point. It is a block circuit diagram shown. In FIG. 13, the same parts as those in FIG. 1 are denoted by the same reference numerals, and the description thereof will not be repeated.

In FIG. 13, an inverse telecine conversion circuit 144 automatically determines whether or not a field is repeated from an input telecine image signal. The result of the field determination is supplied to a type determination circuit 145.
The type determination circuit 145 stops outputting a field removal signal instructing field removal to the inverse telecine conversion circuit 144 when the field repetition that should be detected is not detected.

FIG. 14 shows an example of a case where the presence / absence of field repetition is automatically determined. Here, it is assumed that there is an editing point between the image 9 and the image a, and the 3: 2 telecine pattern is disturbed.

First, in the part of the image a, since the field repetition which should be detected is not detected, the type determination circuit 145 sets the inverse telecine conversion circuit 14
The output of the field removal signal instructing the field removal to No. 4 is stopped. In the next image b, the field repetition is detected, but the type determination circuit 145 also keeps the output stop state of the field removal signal.
Thereafter, the output stop state of the field removal signal is maintained even if the field repetition is detected, and the output of the field removal signal is started for the first time in the image f. As a result, as shown in FIG. 14, with respect to the image j and thereafter, the image preceding the I picture can be set as a B picture having a repeated field.

In addition, according to the automatic detection method, it is possible to change the cycle of the I picture. Then, image a
Is originally a B picture, but it is changed to a P picture, and thereafter, the coding type is changed to IBBPI. As a result, with respect to the picture h and thereafter, the picture before the I picture can be set as a B picture having a repeated field.

FIG. 15 shows an example of determining the coding type when the editing point and the reference point 1 are input. Here, it is assumed that there is an editing point between the image 9 and the image a, and that the reference point 1 is located at positions 6 and f.

In this example, similarly to the above case, if the inverse telecine conversion is controlled by the type determination circuit 145 so as to stop the field removal after the edit point, the image preceding the I picture is repeated. It can be a B picture with a field. In particular, in this case, since the reference point 1 and the edit point are known before encoding, FIG.
As shown in FIG. 5, the coding type can be determined ahead of the editing point, and the cycle of the I picture can be changed by changing the images 9 and a1 to B pictures.

The method of determining the coding type for an image having an edit point for 3: 2 telecine has been described above, but the same concept can be applied to 3: 2: 2: 3 telecine.

(Eighth Embodiment) In a signal compressed by the MPEG method, a B picture and a P picture cannot be decoded only by codes for a specific coded picture. Therefore, when it is necessary to randomly access a specific image, the specific image is
It must be encoded as a picture.

FIG. 16 is a block circuit diagram showing a configuration in which a specific image is encoded as an I picture as an eighth embodiment of the image compression encoding apparatus according to the present invention. In FIG. 16, the same parts as those in FIG. 1 are denoted by the same reference numerals, and redundant description will be omitted.

In FIG. 16, a terminal 146 receives a time code synchronized with an input telecine image signal, and a terminal 147 receives a time code of a specific image to be encoded by an I picture before encoding. Decision circuit 1
48. The type determination circuit 148 detects the timing at which the synchronization time code matches the specific image time code, and sets the image at this time as an I picture. Subsequently, the output of the field removal signal to the inverse telecine conversion circuit 149 is stopped until the next I picture regardless of the content of the repeated field determination signal from the inverse telecine conversion circuit 149.

FIG. 17 shows an example of coding type determination in this embodiment. Here, an image to be coded by an I picture is referred to as an image 9. When the image 9 is set as an I picture, an instruction is issued to the inverse telecine conversion circuit 149 to stop the subsequent field removal (output stop of the field removal signal). The removal of the repeated field is started again. Then, for subsequent coded picture types,
The B picture (image g) before the I picture becomes an image having a repeated field.

As shown in FIG. 17, by not changing the field removal and setting the encoding type of the picture a and thereafter to BBI and temporarily changing the cycle at which the I picture is inserted, The B picture (image e) before the picture can be an image with a repeated field. In the present embodiment, a description has been given of the 3: 2 telecine, but the same processing can be performed on a 3: 2: 2: 3 telecine image.

In the above embodiment, a case has been described where an I picture is inserted at a specific position in an input image group for random access or the like.
It is also possible to insert an I picture in accordance with (Pictures) information. That is, when the GOP information is given, the generation interval of the I picture generated by the start of the GOP is determined, and when the generation interval does not become 6, 12 or 18 pictures, it occurs next to the I picture. By adjusting the generation interval of the I picture, the coded picture immediately before the I picture can be an image signal having the repetition field.

(Ninth Embodiment) FIG. 18 is a block circuit diagram showing a configuration of a ninth embodiment of the image compression encoding apparatus according to the present invention. 18, the same components as those in FIG. 1 are denoted by the same reference numerals, and the duplicate description will be omitted.

The image compression encoding apparatus according to the present embodiment has three components:
2 telecine image signals, 3: 2: 2: 3 telecine image signals, and non-telecine image signals can be handled, and even if different types of image signals are continuously input to the terminal 100, the input image The operation at the time of switching can be stabilized by indicating which kind of signal the signal is.

In FIG. 18, a signal indicating the telecine type is input to a terminal 150, and the inverse telecine processing circuit 151 operates according to the input signal. Each operation is as described in the above embodiment. In the case of an image signal that has not been telecineed, it is output as it is without performing field removal. This makes it possible to cope with an image signal in which the telecine type is changed.

Finally, the present invention is not limited to the above embodiment, but is applicable when the coded picture before the I picture is wider than the display interval of other coded pictures.

[0075]

As described above, according to the present invention, even when the telecine image signal is subjected to inverse telecine processing to remove the repetitive fields and then to be subjected to the encoding processing, immediately before the decoding of the I picture, , MPEG
Provided is an image encoding device capable of optimizing an amount of data remaining in an encoded data buffer memory arranged at an input stage of a decoder and obtaining a high-quality compressed encoded image of a film material. Can be.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a first embodiment of an image compression encoding apparatus according to the present invention when a telecine image signal is compression-encoded in accordance with the MPEG-2 video standard.

FIG. 2 is a diagram for explaining an operation of the inverse telecine conversion circuit according to the first embodiment;

FIG. 3 is a diagram for explaining the operation of the type determination circuit according to the first embodiment.

FIG. 4 shows that the I picture interval is 6, 1 in the first embodiment.
The figure which shows the arrangement | sequence at the time of controlling to 2 and 18.

FIG. 5 is a block circuit diagram showing a configuration in a case where a telecine phase indication signal synchronized with a telecine image signal is provided as a second embodiment of the image compression encoding apparatus according to the present invention.

FIG. 6 is a block circuit diagram showing a configuration in a case where an edit point signal indicating an edit location of a telecine image signal is provided as a third embodiment of the image compression encoding apparatus according to the present invention.

FIG. 7 is a diagram showing an example of a telecine pattern of an image signal between editing points according to the third embodiment.

FIG. 8 is a block circuit diagram showing a configuration when processing a 3: 2: 2: 3 telecine image signal as a fourth embodiment of the image compression encoding apparatus according to the present invention.

FIG. 9 is a block circuit diagram showing a configuration for processing a 3: 2: 2: 3 telecine image signal having an editing point as a fifth embodiment of the image compression encoding apparatus according to the present invention.

FIG. 10 is a block circuit diagram showing a configuration of a sixth embodiment of the image compression encoding apparatus according to the present invention.

FIG. 11 is a block circuit diagram showing a specific configuration of an inverse telecine conversion circuit used in a sixth embodiment.

FIG. 12 is a diagram showing processing contents of automatic detection of a telecine phase, inverse telecine processing, and determination of a coding type in a sixth embodiment.

FIG. 13 is a block circuit diagram showing, as a seventh embodiment of the image compression encoding apparatus according to the present invention, a configuration characterized by a method of determining an encoding type for a telecine image having an editing point.

FIG. 14 is a diagram showing an example of a case in which the presence or absence of field repetition is automatically determined in the seventh embodiment.

FIG. 15 is a diagram showing an example of determining an encoding type when an edit point and a reference point 1 are input in the seventh embodiment.

FIG. 16 is a block circuit diagram showing a configuration for encoding a specific image as an I picture as an eighth embodiment of the image compression encoding apparatus according to the present invention.

FIG. 17 is a view showing an example of coding type determination in the eighth embodiment.

FIG. 18 is a block circuit diagram showing a configuration of a ninth embodiment of an image compression encoding apparatus according to the present invention.

[Explanation of symbols]

100: Telecine image signal input terminals, 101, 120,
121, 127, 133, 144, 149, 151: inverse telecine conversion circuit, 102: motion detection circuit,
103: image rearrangement circuit, 104: subtractor, 105:
DCT (discrete cosine) transform circuit, 106 quantizer,
107: inverse quantizer, 108: inverse DCT converter, 109
... adder, 110 ... motion compensation image memory, 112, 12
2, 128, 132, 142, 145, 148: type determination circuit, 113: grammar generation circuit, 115: code amount control circuit, 116: encoded data buffer memory, 117 ...
Fixed-rate coded data output terminal, 118... I-picture interval indicating signal input terminal, 119... Telecine phase indicating signal, 123 129, 146.
124, 130 ... telecine phase signal input terminal, 125,
131: edit point signal input terminal; 126: field removal signal input terminal; 135, 136 ... two-field delay circuit; 137 ... subtractor; .., 141... Repeated field determination signal input terminal, 147... Specific image time code input terminal, 15.
0 ... Telecine type signal input terminal.

Claims (9)

[Claims] 1. An ISO / IEC 13818-2 (MPEG)
An image compression encoding apparatus for compressing and encoding a telecine image signal in which fields are repeatedly inserted according to a predetermined pattern in accordance with a standard, wherein a repeated field removal for removing a repeated field in the telecine image signal is performed. Means, for an image signal of each field of the telecine image signal from which repeated fields have been removed by this means, an I picture (Intra-coded picture) defined by the standard,
Coded picture type determining means for determining which type of coded picture, a P picture (Predictive-code picture) or a B picture (Bi-directionally predictive-coded picture), and a type determined by this means Compression encoding means for compressing and encoding the image signal of each field output from the field removing means based on the standard, and wherein the encoded picture type determining means comprises: An image compression encoding apparatus characterized in that a certain image signal is of an encoded picture type immediately before said I picture. 2. The coded picture type determining means according to claim 1, wherein said telecine image signal is a 3: 2 telecine signal in which an image signal of a repetitive field and an image signal of a non-repeated field are alternately arranged. , The P picture or I picture generation interval is set to three picture intervals, and the I picture generation intervals are set to 6, 12, 18, 2,
4, wherein the telecine image signal is an image signal group of these 12 fields in which the image signal of the repeated field is continuous twice, and the image signal without the repeated field is continuous three times. Is a 3: 2: 2: 3 telecine image signal repeatedly arranged, the generation interval of the P picture or the I picture is set to 3 picture intervals, and the generation interval of the I picture is set to 12 or 24 picture intervals. The image compression encoding apparatus according to claim 1, wherein: 3. When a repetition field indication signal indicating a position of a repetition field of the telecine image signal is given, the repetition field removal means removes an image signal of the repetition field according to the repetition field indication signal. Picture type determination means determines an encoded picture type of an image signal of each field according to the repeated field indication signal such that an image signal having the repeated field becomes an encoded picture type immediately before the I picture. 2. The method according to claim 1, wherein
The image compression encoding apparatus according to claim 1. 4. When a discontinuous point indicating signal indicating a position at which a telecine pattern of the telecine image signal is discontinuous is provided, the repetitive field removing means detects a repetitive field based on the discontinuous point indicating signal. Removing the image signal, the coded picture type determining means determines an image signal of a field to be the I picture based on the discontinuity indication signal, and the coded picture type immediately before the I picture is 2. The image compression encoding apparatus according to claim 1, wherein an encoded picture type of an image signal of each field is determined so that the image signal has the repeated field. 5. A telecine pattern discriminating means for discriminating a telecine pattern of an input image signal, wherein said repetitive field removing means converts an image signal of a repetitive field according to the pattern discriminated by said telecine pattern discriminating means. The coded picture type determining means receives telecine pattern information from the telecine pattern discriminating means, and, based on this information, sets the coded picture type immediately before an I picture to an image having the repetition field. 2. The image compression encoding apparatus according to claim 1, wherein an encoded picture type of an image signal of each field is determined so as to be a signal. 6. The coded picture type determining means includes:
When the telecine pattern discriminating means cannot discriminate the telecine pattern, the interval of the I picture is temporarily changed, and after the telecine pattern discriminating means discriminates the pattern, the I-picture is determined based on the telecine pattern information. 6. The image compression encoding apparatus according to claim 5, wherein the encoding picture type of the image signal of each field is determined so that the immediately preceding encoding picture type becomes the image signal having the repeated field. 7. The repetitive field removing means, when the telecine pattern discriminating means cannot discriminate the telecine pattern, temporarily stops removing the image signal of the repetitive field and changes the interval between I pictures. 6. The image compression encoding apparatus according to claim 5, wherein after the pattern is determined by said telecine pattern determining means, the image signal of the repeated field is removed based on the result of the determination. 8. The coded picture type determining means includes:
When the image signal at the specific position is designated as an I picture, the interval between the I pictures is temporarily changed, and the image signal at the specific position is set as an I picture, and
Determining the coded picture type of the image signal of each field so that the coded picture type immediately before the picture becomes the image signal having the repetition field, and thereafter returning the interval of the I picture to the original. The image compression encoding apparatus according to claim 1. 9. The repetition field removing means, when the image signal at a specific position is designated as an I picture, temporarily stops the removal of the repetition field image signal and changes the I picture interval. 2. The image compression encoding apparatus according to claim 1, wherein after outputting the image signal of the field at the specific position, the removal of the image signal of the repeated field is restarted.