JPH11252515A - Signal detector and its method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a device capable of convert-processing so as to rightly convert a telecine picture signal back to an original video signal based on a second field by providing a means detecting the second field based on the detecting result of the number of times of judging prescribed number of pixels to be different from each other. SOLUTION: An inputted video field signal S11 and a comparing video field signal S12 are successively compared by the unit of a block to detect the number of the times of judging corresponding blocks to be different from each other based on the comparing result. Then, based on the detected number of times, whether the field S11 is a repeated field or not is judged. Then, even in a case when noise is generated at the signals S11 and S12, whether the signal S11 is the repeated field or not is securely judged by the number of judging the corresponding blocks to be different from each other and the repeated field can be detected.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[Table of Contents] The present invention will be described in the following order.

TECHNICAL FIELD The prior art (FIGS. 16 to 20) Problems to be solved by the invention (FIGS. 21 and 22) Means for solving the problems (FIGS. 1 to 15) Embodiment (1) Configuration of Encoding Device According to Embodiment (FIGS. 1 to 13) (2) Controller 11 and Repeat Field Detection 1
(3) Various processing in the encoding device (3-1) Repeat field determination processing (FIGS. 5 and 6) (3-2) Telecine check processing (3-3) Reverse Telecine departure processing (FIGS. 7 and 8) (3-4) Fixed length processing (FIGS. 9 to 14) (4) Encoding processing procedure (FIG. 15) (5) Operation and effect of the present embodiment (FIG. (1 to 15) (6) Other Embodiments (FIGS. 1 to 15) Effects of the Invention

[0003]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a signal detection apparatus and a signal detection method, for example, converting a 24 [Hz] movie signal generated from an image recorded on a movie film into a 30 [Hz] video signal. (Hereinafter referred to as telecine image signal in particular) and 30 [Hz] of commercial video
The present invention is suitably applied to an encoding device that encodes a video signal in which a non-telecine image signal composed of a video signal is mixed.

[0004]

2. Description of the Related Art Conventionally, in a telecine image signal, as shown in FIG. 16, a four-frame movie signal ES1 has a top field tf of the second frame between the second and third frames from the beginning. Repeatedly insert the bottom field bf of the fourth frame at the end.
Is repeatedly inserted to convert the signal into a signal of 5 frames, and further, the top field tf and the bottom field bf are rearranged so as to be positioned alternately.

Accordingly, the telecine image signal TS1 is composed of a frame FR1, in which a so-called repeat field rf, which is obtained by repeating the top field tf or the bottom field bf, is located every five fields, and has two fields.
A pattern (hereinafter, referred to as a telecine pattern) representing a frame configuration peculiar to the telecine image signal TS1 in which the frame FR2 and the frame FR2, which is composed of three fields including the repeat field rf, and the frame FR4 are sequentially and alternately located.

Then, such a telecine image signal TS1
Is encoded, the telecine image signal TS1 is converted so as to return to the original movie signal ES1, and the encoding efficiency is improved by encoding after reducing the number of fields. As an encoding device that performs such encoding processing, there is an encoding device configured as shown in FIG.

[0007] That is, as shown in FIG. 17, in the encoding device 1, under control of the controller 2, the telecine image signal TS 1 for at least four fields is converted under the control of the controller 2. The data is sequentially written into the writable memory 3 in units of fields.

In the encoding apparatus 1, a telecine image signal TS1 supplied from the outside is also taken into the repeat field detecting section 4 sequentially in units of fields, and the repeat field detecting section 4 stores one field. Every time a telecine image signal (hereinafter, referred to as an input telecine field signal) TS2 is taken in, the memory 3 under the control of the controller 2 is two fields earlier than the input telecine field signal TS2 by time. A telecine image signal for one field (hereinafter, referred to as a comparison telecine field signal) TS3
And supplies the same to the repeat field detector 4.

In this case, the controller 2 receives the input telecine field signal TS2 from the repeat field detector 4.
And a comparison result with the comparison telecine field signal TS3. If a repeat field is detected once based on the comparison result, the repeat field is supplied every five fields thereafter based on the detected repeat field. The timing for reading the repeat field from the memory 3 is generated.

The controller 2 reads out the telecine image signal TS1 from the memory 3 in the order in which the telecine image signal TS1 is written in the field unit. However, when reading out the repeat field included in the telecine image signal TS1, the controller 2 does not read out the repeat field. Is written in the area where is written, a new field of the telecine image signal TS1 for one field supplied from the outside is discarded, and thus the repeat field of the telecine image signal TS1 from the memory 3 is discarded. (Hereinafter referred to as inverse telecine processing) so as to return to the original movie signal ES1 while reading.
Then, the movie signal ES1 is given to the encoder 5.

At this time, the controller 2 has a memory 3
Each time the movie signal ES1 for one frame is read out and supplied to the encoder 5, the movie signal ES1 for one frame is read out.
, A top field fast flag TF1 indicating whether the top field is a top field or a bottom field, and the movie signal ES for one frame.
1 is a repeat first field flag RF1 indicating whether or not the first field is a field that has been added to the repeat field when the telecine image signal is generated.
Are given to the encoder 5.

The encoder 5 sequentially converts the movie signal ES1 supplied from the memory 3 into MPEG2
(Moving Picture Experts Group Phase2) encoding, generating a coded signal FS1 by storing a top field fast flag TF1 and a repeat fast field flag RF1 corresponding to the header portion, and outputting this to the outside. .

When the encoded signal FS1 is decoded by a decoder (not shown), the top field fast flag TF stored in the header portion is used.
The telecine image signal TS1 can be generated from the coded signal FS1 by decoding based on 1 and the repeat fast field flag RF1.

Here, in practice, in the repeat field detector 4, as shown in FIG. 18, an input telecine field signal TS2 supplied from the outside and a corresponding comparison telecine field signal TS3 read from the memory 3 are output. The difference is given to the differentiator 7 in synchronization with each pixel, and the subtracter 7 sequentially subtracts, for example, the luminance level of the corresponding pixel between the input telecine field signal TS2 and the comparison telecine field signal TS3, and obtains the absolute value of the obtained difference. (Less than,
This is called an absolute difference value).

At this time, the controller 9 controls the vertical synchronizing signal VS1 supplied from the outside in synchronization with the telecine image signal TS1.
, And starts counting at the timing at which the vertical synchronizing signal VS1 is applied, and the adding unit 8 sequentially adds the absolute difference values at the timing of counting as described above.

After that, when the control section 9 counts the number corresponding to the number of pixels for one field (that is, when the vertical synchronizing signal VS1 is supplied), the control section 9 supplies the operation reset signal S1 to the adding section 8 to thereby calculate the number. The obtained addition result (that is, the sum of the absolute values of the differences for one field) is used as a comparison result, and a residual signal S2 representing the result is sent to the controller 2, and the calculation result in the addition section 8 is reset. , Again add a new absolute difference value for one field.

In this case, the controller 2 (FIG. 17) compares the sum of the absolute values of the differences for one field obtained based on the residual signal S2 with a preset threshold value and, based on the obtained comparison result. If the sum of the absolute differences is greater than the threshold value, the corresponding input telecine field signal TS2 is not a repeat field obtained by repeating the comparison telecine field signal TS3 when generating the telecine image signal TS1. If the sum of the absolute values of the differences is smaller than a threshold value based on the comparison result, it is determined that the input telecine field signal TS2 is a comparison telecine field signal TS3 repeat field. Thus, the repeat field is detected.

In the coding apparatus 1 (FIG. 17),
By setting the signal to be subjected to the encoding process from the telecine image signal TS1 to the non-telecine image signal HS1, the non-telecine image signal HS1 can also be encoded. In this case, The non-telecine image signal HS1 supplied from the outside is
Under the above control, the non-telecine image signal HS1 is sequentially written in the memory 3 in units of fields, and is read out from the memory 3 in accordance with the order in which the non-telecine image signals HS1 are sequentially written in units of fields.

At this time, the controller 2
Non-telecine image signal HS for one frame read from
Every time 1 is given to the encoder 5, the corresponding top field first flag TF1 and the repeat first field flag RF1 are given to this encoder 5.

Thus, the encoder 5 sequentially encodes the non-telecine image signal HS1 supplied from the memory 3 in frame units, and stores the top field fast flag TF1 and the repeat fast field flag RF1 corresponding to the header portion. To obtain the encoded signal FS2
And output this to the outside.

In this way, in the encoding device 1, as shown in FIG. 19, when the telecine image signal TS1 is supplied from the outside, the controller 2 writes the telecine image signal TS1 into the memory 3 and simultaneously By reading the telecine image signal TS1 from the memory 3 except for the repeat fields fd13 and fd18 detected via the repeat field detection unit 4, the telecine image signal TS1 is subjected to inverse telecine processing and given to the encoder 5.

At this time, the controller 2 controls each frame FR1 of the movie signal ES1 read from the memory 3.
1, if the first field of FR12, FR13 is the top field tf11, tf15, tf17, the top field fast flag TF1 set to "1" indicating this is given to the encoder 5, and the top field is also given. Is the bottom field bf12, bf14, b
In the case of f16, the top field fast flag TF1 set to "0" indicating this is given to the encoder 5.

The controller 2 also controls each frame F
If the first field of R11, FR12, and FR13 is a field based on the repeat fields fd13 and fd18, a repeat fast field flag RF1 set to "1" indicating this is sent to the encoder 5. In addition, if the first field is not the field that is the origin of the repeat fields fd13 and fd18, the repeat fast field flag RF1 set to "0" indicating this is encoded by the encoder 5.
Give to.

On the other hand, in this encoding device 1, FIG.
As shown by 0, when the non-telecine image signal HS1 is supplied from outside, the controller 2 writes the non-telecine image signal HS1 into the memory 3 once, and sequentially outputs the non-telecine image signal HS1 from the memory 3. The data is read out in the order of writing in units of fields and supplied to the encoder 5.

At this time, the controller 2 sets the first field of each frame FR21, FR22, FR23, FR24 of the non-telecine image signal HS1 read from the memory 3 to the top field fd21, fd23, fd.
In the case of 25 and fd27, the top field fast flag TF1 set to "1" indicating this is supplied to the encoder 5, and in the case where the top field is the bottom field fd22, fd24, fd26, fd28, Is supplied to the encoder 5 (in this case, each frame FR21, FR22, FR23, FR24).
Is the top field fd21, fd23, fd2
5, fd27), and the repeat fast field flag RF1 set to "0" is given to the encoder 5 because there is no repeat field.

Thus, the encoding device 1 can encode the telecine image signal TS1 or the non-telecine image signal HS1.

[0027]

By the way, in the coding apparatus 1 having such a configuration, when a repeat field is detected, even if the input telecine field signal TS1 is actually a repeat field of the corresponding comparison telecine field signal HS1, As shown in FIGS. 21A and 21B, an image based on the input telecine field signal TS2 (hereinafter, referred to as an input telecine image) T1 (FIG. 21A) and a corresponding comparison telecine field signal If noise is generated throughout the image based on TS3 (hereinafter, referred to as a comparative telecine image) H1 (FIG. 21B), the size of the entire image is increased to some extent by comparing these. Since the sum of the absolute difference values is obtained, the input telecine field signal with noise S2 it is necessary to set the threshold to avoid that determined not to be erroneous connexion repeat field.

However, as shown in FIGS. 22 (A) and (B), an input telecine image T2 in which the whole is a scene with almost no movement but only a very small bird BI moves, and a corresponding comparison When the telecine image H2 is compared with the telecine image H2, there is a case where the total value of the difference absolute values almost the same as when the input telecine field signal TS1 in which the above-described noise is generated and the corresponding comparison telecine field signal HS1 are compared is obtained. In this case, the input telecine image T
There is a problem that the input telecine field signal TS2 of No. 2 is erroneously determined to be a repeat field.

That is, in the encoding device 1, when a telecine image signal TS1 is supplied which is a repeat field but generates noise or is not a repeat field but has a slight movement, the discrimination is made. Because of this, the repeat field was erroneously detected, and it was difficult to convert the telecine image signal TS1 back to the original movie signal.

The present invention has been made in view of the above points, and provides a signal detecting apparatus and a signal detecting method capable of converting a telecine image signal into an original movie signal correctly based on a second field. It is something to propose.

[0031]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, according to the present invention, when generating a telecine image signal from a telecine image signal obtained by converting a movie signal into a video signal, a predetermined first signal of the movie signal is used. In the signal detection device for detecting the second field inserted at a position separated by a predetermined number of fields by repeating the field of the above, the predetermined field of the telecine image signal and the field at a position corresponding to the predetermined predetermined number of fields are sequentially determined. A comparing means for comparing in a unit of a predetermined number of pixels, and judging whether or not the corresponding predetermined number of pixels are the same based on a comparison result obtained from the comparing means, and comparing the corresponding predetermined number of pixels based on the judgment result. Number detecting means for detecting the number of times that is determined to be different, and a second file based on a detection result of the number of times obtained from the number detecting means. It was provided a field detection means for detecting a Rudo. A signal detection device comprising:

As a result, even when noise is generated in the fields or there is only a slight movement between the fields, it is possible to accurately detect the second field by the number of times that the predetermined number of pixels are determined to be different. it can.

Further, in the present invention, when generating the telecine image signal from the telecine image signal obtained by converting the movie signal into a video signal, a predetermined first field of the movie signal is repeated and separated by a predetermined number of fields. In a signal detection method for detecting a second field inserted at a position, a first step of sequentially comparing a predetermined field of a telecine image signal and a field at a position away from the corresponding predetermined number of fields by a predetermined number of pixels. And whether the corresponding predetermined number of pixels are the same based on the comparison result obtained in the first step, and the number of times the corresponding predetermined number of pixels are determined to be different based on the determination result. And a second field for detecting the second field based on the number of detections obtained in the second step. That it was set to the third provided a step.

As a result, even when noise is generated in the fields or there is only slight movement between the fields, it is possible to accurately detect the second field by the number of times that the predetermined number of pixels are determined to be different. it can.

[0035]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below in detail with reference to the drawings.

(1) Configuration of Encoding Device According to the Present Embodiment In FIG. 1, reference numeral 10 denotes an encoding device according to the present invention as a whole, in which telecine image signals and non-telecine image signals supplied from the outside are mixed. Under the control of the controller 11, the video signal S10 is written into the memory 12 in which the video signal S10 for at least four fields can be written.

At this time, in the encoding device 10,
The video signal S10 supplied from the outside is also taken in by the repeat field detection unit 13, and the repeat field detection unit 13 supplies one field of video signal (hereinafter, referred to as an input video field signal) S1.
1 is fetched from the memory 12 under the control of the controller 11 every time the input video field signal S11
A video signal for one field two fields past in time (hereinafter referred to as a comparison video field signal)
S12 is read out, and this is read by the repeat field detector 1
Give to 3.

In this case, the controller 11 sends the input video field signal S11 from the repeat field detection unit 13.
And a comparison result with the comparison video field signal S12 corresponding thereto are sequentially provided, a repeat field signal is detected based on the comparison result, and reading of the video signal S10 from the memory 12 is controlled based on the obtained detection result. Accordingly, the telecine image signal included in the video signal S10 is subjected to inverse telecine processing as necessary, and the non-telecine image signal is read as it is, and the video signal including the read telecine image signal, movie signal, and non-telecine image signal is read. S13 is given to the encoder 14.

At this time, the controller 11 reads the movie signal S13 for one frame from the memory 12 and supplies it to the encoder 5 every time the movie signal S13 for one frame is read.
13, a top field first flag TF2 indicating whether the top field is a top field or a bottom field, and the top field of the movie signal S13 for one frame is a source of a repeat field when a telecine image signal is generated. Repeat first field flag R indicating whether or not the field is
And F2 to the encoder 14.

Thus, the encoder 14 sequentially encodes the movie signal S13 supplied from the memory 12 in a frame unit by, for example, the MPEG2 system so as to have a fixed length.
Top field first flag TF2 and repeat first field flag RF2 corresponding to the header portion
To generate an encoded signal S14, and output this to the outside.

(2) Detailed Configuration of Controller 11 and Repeat Field Detector 13 As shown in FIG. 2, the controller 11 is provided with a state machine 20.
As shown in FIG. 3, when the entire coding apparatus 10 is reset as shown in FIG. 3, the processing shifts from the standby state to the normal processing mode. In this normal processing mode, the video signal S10 is externally supplied to the coding apparatus 10. Therefore, this video signal S1
The entire controller 11 is controlled so that 0 is sequentially supplied to the encoder 14 in frame units irrespective of the telecine image signal and the non-telecine image signal included therein.

When the state machine 20 detects a repeat field through the repeat field detector 13 by supplying the video signal S10 consisting of a telecine image signal to the encoding device 10 in the normal processing mode, the state machine 20 executes the telecine check. In the processing mode, the video signal S10 is sequentially supplied to the encoder 14 in frame units in the same manner as in the normal processing mode.
The controller 11 is controlled so as to execute a telecine check process for checking whether or not the repeat field is correctly detected from the video signal S10 supplied to the controller 11.

Further, in the state machine 20, the video signal S1 supplied by detecting a predetermined number of repeat fields in the telecine processing mode.
When it is confirmed that 0 is a telecine image signal, the mode shifts to an inverse telecine processing mode, and the entire controller 11 is controlled to execute an inverse telecine process so as to return the telecine image signal to a movie signal. Signal S1
When 0 changes from a telecine image signal to a non-telecine image signal, the inverse telecine processing is terminated, and the process returns to the normal processing mode.

When the state machine 20 erroneously detects a repeat field from the video signal S10 supplied from the outside in the telecine processing mode, the state machine 20 shifts to the normal processing mode.

In practice, in the controller 11 (FIG. 2), in the normal processing mode, the memory controller 21
By generating a recording address signal S17 and supplying it to the memory 12, the video signal S10 supplied from the outside is sequentially written to the memory 12 in units of fields, and the second field after the writing of the video signal S10 to the memory 12 is started. When the writing of the video signal S10 for one field is started, a read address signal S18 is generated and applied to the memory 12, so that the video signal S10 for one field written first from the memory 12 is read out as the comparison video field signal S12. This is given to the repeat field detection unit 13.

As described above, the memory controller 21
Thereafter, each time one field of the video signal S10 is started to be written into the memory 12 (in all modes), one field of the video signal S10 is sequentially read from the memory 12.
Are read out in the order in which they were written as the comparison video field signal S12, and this is read out by the repeat field detection unit 1.
Give to 3.

Further, the memory controller 21
When the video signal S10 for one field of the fourth field is started to be written after the video signal S10 is started to be written to the memory 2, the read address signal S18 is given to the memory 12 again, and the first written data from the memory 12 is read.
The video signal S10 for the field is read out and supplied to the encoder 14.

After this, the memory controller 21
Each time one field of video signal S10 is started to be written to memory 12 (in the normal processing mode and telecine check processing mode), one field of video signal S10 is read out from memory 12 in the order in which it was written and encoded. Each time a video signal S10 for one frame is read from the memory 12, a frame read signal S19 representing the read video signal S19 is supplied to a top field fast operation unit 22.

Top field first operation unit 22
Is provided with a vertical synchronization signal VS2 supplied in synchronization with the video signal S10 from the outside, and the video signal first written in the memory 12 based on the field identification information included in the vertical synchronization signal VS2. It is determined whether S10 is a top field or a bottom field. If the video signal S10 for one field initially written in the memory 12 is a top field based on the detection result, the top field fast flag is set. TF2 is set to "1", and when the video signal S10 for one field is a bottom field, the top field first flag TF2 is set to "0".

In the normal processing mode (and in the telecine check processing mode), the top field first calculating unit 22 sets the top field first set every time the frame read signal S19 is given from the memory controller 21. The flag TF2 is provided to the encoder 14 and the repeat fast field calculator 23.

Repeat fast field computing section 23
In the normal processing mode (and in the telecine check processing mode), the video signal S10 written in the memory 12 as described above is read out irrespective of the telecine image signal and the non-telecine image signal, and , The repeat first field flag RF
2 is set to “0”, and this is set by the top field first arithmetic unit 22 to the top field first flag T.
Each time F2 is provided, it is provided to the encoder 14.

Incidentally, the fixed-length operation unit 24 includes the encoder 1
4 should be fixed length (for example, 1
5 frames previously set as the number of frames (to be coded with the coding amount of [Mbit]) (hereinafter referred to as a fixed length unit) are used as the fixed length setting signal S20.
4 to thereby set a fixed length unit for the encoder 14.

Thus, in the normal processing mode (and in the telecine processing mode), the encoder 14 operates in the memory 12
Is sequentially encoded in frame units so as to have a preset code amount for each fixed length unit, and the top field fast flag TF2 and the repeat fast field flag RF2 corresponding to the header portion are encoded. Is stored in the encoded signal S14.
And output this.

In the repeat field detector 13, as shown in FIG. 4, an input video field signal S11 supplied from outside (in all modes).
And the corresponding comparison video field signal S12 read from the memory 12 are given to the differentiator 27 in synchronization with each other on a pixel basis, and the input video field signal S1 and the comparison video field signal S2 are , For example, are sequentially subtracted on a pixel-by-pixel basis, and the obtained absolute difference value is provided to the adding unit 28.

At this time, the control unit 29 starts counting from "0" at the timing when the vertical synchronizing signal VS2 is given from the outside, and supplies the arithmetic reset signal S23 to the adding unit 28 every time a multiple value of 50 is counted. At the same time, each time the number corresponding to the number of pixels for one field is counted, a determination result reset signal S24 is provided to the determination unit 30.

The adder 28 sequentially adds the difference absolute values sequentially given from the differentiator 27, and each time the control reset signal S23 is given from the control unit 29, determines the addition result of the difference absolute value calculated at that time. And the addition result is reset, and the difference absolute value provided from the differentiator 27 is added again. This addition process is sequentially repeated. In this way, the adder 28 gives the determination unit 30 the result of adding the absolute value of the difference sequentially obtained from the block of 50 pixels corresponding to the input video field signal S11 and the comparison video field signal S12.

The judging section 30 compares the addition result sequentially given from the adding section 28 with a preset threshold value (for example, 500).
If the addition result is equal to or greater than the threshold based on the obtained comparison result, it is determined that the corresponding blocks used to obtain the addition result are different from each other, and the addition is performed. If the result is smaller than the threshold value, it is determined that the corresponding blocks are the same.

Each time the determination result reset signal S24 is supplied from the control unit 29, the determination unit 30 outputs the input video field signal S11 and the comparison video field signal S12.
As a comparison result, the controller 11 outputs a determination result signal S25 representing the number of times corresponding blocks are determined to be different from each other (hereinafter referred to as the number of mismatches).
(FIG. 2) to the repeat field determination unit 33 (FIG. 2).

The repeat field judgment section 33 (FIG. 2)
Compares the number of mismatches obtained based on the determination result signal S25 with a preset threshold value (for example, 3). If the number of mismatches is equal to or less than the threshold value based on the obtained comparison result, It is determined that the corresponding input video field signal S11 is a repeat field, the repeat field detection signal S27 is set to "1" to indicate this, and if the number of mismatches is greater than a threshold value, It is determined that the input video field signal S11 is not a repeat field, and the repeat field detection signal S27 is set to "0" to indicate this.
Set to.

Then, the repeat field judging section 33
Is thus obtained as the repeat field detection signal S27
Is set to "1" or "0", this is supplied to the state machine 20, the telecine check counter 34, the fixed length operation unit 24 and the repeat fast field operation unit 23.

Here, the state machine 20 outputs “1” from the repeat field determination unit 33 in the normal processing mode.
When the repeat field detection signal S27 set to is provided, the mode shifts to the telecine processing mode, and the status signal S28 indicating this is sent to the telecine pattern generation unit 35.
Give to.

The telecine pattern generation unit 35 recognizes that a repeat field has been detected based on the status signal S 28 given from the state machine 20, and thereafter, externally sends an image composed of a telecine image signal to the encoding device 10. If the signal S10 continues to be supplied, the repeat field is supplied every five fields with reference to the repeat field, so that the status signal S28 and the externally applied vertical synchronization signal VS2
Then, every time the video signal S10 for one field is supplied every five fields, a repeat signal S30 representing the repeat field is sequentially transmitted to the state machine 20,
The timing at which the repeat field is supplied is generated by giving the result to the telecine check counter 34 and the repeat fast field calculation unit 23.

At this time, the fixed length calculation section 24 counts up a field counter provided therein based on a vertical synchronization signal VS2 supplied from the outside, and sets a field to be fixed length set in advance. The number (for example, 10 fields) is repeatedly counted, thereby detecting a fixed length target section (hereinafter, referred to as a fixed length section) of the video signal S10 supplied from the outside to the encoding device 10.

When the fixed-field lengthening operation section 24 receives the repeat field detection signal S27 set to "1" from the repeat field judgment section 33 when the mode is switched to the telecine processing mode, the fixed-length operation section 24 first detects the signal. The count value of the field counter when the repeat field detection signal S27 set to “1” is again supplied from the repeat field determination unit 33 is detected from the break position of the fixed length section, and the detected count value is determined. Based on the detected telecine pattern, a predetermined number of repeat field detections (e.g., 6 or 7) is selected based on the detected telecine pattern based on the detected telecine pattern. S
31 is given to the telecine check counter 34.

If the repeat field detection signal S27 set to "1" is given from the repeat field determining unit 33 when the mode is shifted to the telecine processing mode, the telecine check counter 34 sets the count value as the number of times the repeat field is detected. For example, it is set to “10” which is set in advance, and when the set number signal S31 is given from the fixed length calculation unit 24 thereafter, the set number signal S31
, The count value is reset to the same value as the number of times the repeat field is detected according to the telecine pattern.

The telecine check counter 34
The reset count value is provided to the state machine 20 and the fixed length calculation unit 24 as a repeat field detection count signal S32, and thereafter, the repeat field detection signal S27 set to "1" from the repeat field determination unit 33 is output. The count value is decremented by one each time a repeat signal S30 is supplied from the telecine pattern generation unit 35, and the count value is supplied to the state machine 20 and the fixed length calculation unit 24 as a repeat field detection number signal S32.

In this case, the state machine 20 sequentially determines whether or not the repeat field has been correctly detected until the repeat count detection signal S32 representing the count value of "0" is given from the telecine check counter 34. If the repeat signal S30 from the telecine pattern generation unit 35 and the repeat field detection signal S27 set to “1” from the repeat field determination unit 33 are given at different timings, for example, Video signal S1 consisting of an image
It is determined that 0 has been mistakenly detected as a repeat field, and the process returns to the normal processing mode.

On the other hand, when the repeat signal S30 is supplied and the repeat field detection signal S27 set to "1" is supplied, the state machine 20 determines that the repeat field is correctly detected. In this way, the telecine check counter 34 outputs "0".
Repeat field detection count signal S representing the count value of
If it is determined that the repeat fields are sequentially detected correctly until 32 is given, the operation shifts to the reverse telecine processing mode when the repeat field detection number signal S32 representing the count value of “0” is given.

In the inverse telecine processing mode, the repeat fast field computing unit 23 sets the repeat fast to "0" every time the top field fast flag TF2 is given from the top field fast computing unit 22 in the same manner as described above. Although the stfield flag RF2 is provided to the encoder 14, each time the repeat signal S30 is provided from the telecine pattern generation unit 35 and the repeat field detection signal S27 set to "1" is provided from the repeat field determination unit 33. Instead of the repeat fast field flag RF2 set to "0", the repeat fast field flag RF2 set to "1" is replaced by the encoder 14 and the memory controller 21.
Give to.

The memory controller 21 sequentially reads out the video signal S10 from the memory 12 in units of fields in the same manner as described above, and supplies it to the encoder 14. However, the repeater controller 23 sets the repeater signal set to "1". When the stop field flag RF2 is given, the video signal S10 for one field to be read from the memory 12 at this time is recognized as a repeat field, and is not read out. The repeat field is discarded by overwriting the area where the video signal S10 for one field is recorded with the video signal S10 for one field newly supplied from the outside.

As described above, the memory controller 21
Performs an inverse telecine process while reading a telecine image signal from the memory 12 so as to discard the repeat field, thereby providing a video signal S13 composed of a movie signal to the encoder 14. By the way, the memory controller 21
Each time the repeat field is discarded without being read from the memory 12, the read suspension signal S indicating this is discarded.
35 is given to the top field first calculation unit 22.

Top field first calculating section 22
If the memory controller 21 prevents the repeat field from being read from the memory 12, the video signal S1 for one frame read from the memory 12 thereafter
3 is changed to a bottom field or a top field which is opposite to the top field or the bottom field up to that point, the first field of the video signal S13 for this one frame is detected based on the read temporary stop signal S35, The setting of the top field fast flag TF2 is changed in accordance with the detection result, and the result is given to the encoder 14.

By the way, the fixed-length operation section 24 receives the repeat field detection number signal S32 representing the count value of "0" from the telecine check counter 34 (ie, when the mode is shifted to the reverse telecine processing mode). Of the fixed-length section detected immediately before the repeat field detection number signal S32 is given (that is, in the telecine mode), based on the detected count value (hereinafter, referred to as the first position). A telecine pattern from the first break position to a next break position (hereinafter, this is called a second break position) is detected.

Based on the detected telecine pattern, the fixed-length operation unit 24 determines whether or not a frame consisting of two or three fields of the telecine image signal is applied to the second break position of the fixed-length section. Changes the length, which is the number of fields of the fixed-length section, by shifting the second delimiting position so as to be located between the frame and the frame adjacent thereto, and accompanying this change. If the number of frames when actually encoding the telecine image signal in the fixed length section is different from the five frames previously set as the fixed length unit, the number of frames is used as a fixed length setting signal S20 as an encoder. 14, and the fixed length unit in the encoder 14 is reset according to the number of frames at the time of actual encoding.

In this case, the fixed-length operation unit 24 determines that the telecine image signal for 10 fields is located in the fixed-length section after the second delimiter position of the fixed-length section. Is used as the fixed length unit for this fixed length section when the inverse telecine processing is performed on the video signal, and the fixed length unit is reset when the mode is shifted to the inverse telecine processing mode. If the fixed length unit differs from the fixed length unit, the fixed length setting unit S20 is supplied to the encoder 14 again to set the fixed length unit again.

Thus, the encoder 14 encodes the video signal S13 supplied from the memory 12 by fixing the video signal S13 in a fixed-length unit reset as necessary, and the top field fast corresponding to the header portion. Flag TF2
And a repeat fast field flag RF2 to generate an encoded signal S14, which is output.

Thereafter, in the inverse telecine processing mode, the state machine 20 switches the video signal S10 supplied from the outside to the encoder 10 from a telecine image signal to a non-telecine image signal. When the repeat field detection signal S27 set to "0" is supplied from the repeat field determination unit 33 when the repeat signal S30 is supplied, the mode shifts to the normal processing mode, and the inverse telecine processing ends.

As described above, in the reverse telecine processing mode, for example, a video signal S10 in which a non-telecine image signal follows the first field of a frame consisting of three fields of the telecine image signal is supplied, and the last field of the telecine image signal is supplied. Since the repeat field for the tail field cannot be detected by the non-telecine image signal, the mode shifts to the normal processing mode.

At this time, in the controller 11, at the joint between the telecine image signal of the video signal S10 and the non-telecine image signal, the field of the telecine image signal and the non-telecine image signal is left as it is so that the inverse telecine image signal is left. 1st inverse telecine departure processing which shifts from processing to normal processing, or 1 at the end of telecine image signal
One of the second inverse telecine departure processing that shifts from the inverse telecine processing to the normal processing by generating a frame consisting of three fields using the field and discarding the first two fields of the non-telecine image signal accordingly. Execute

In practice, the repeat fast field computing unit 23 receives the first inverse telecine departure processing or the second
The reverse telecine departure mode signal S37 which designates one of the reverse telecine departure processing is provided. When the mode is shifted from the reverse telecine processing mode to the normal processing mode, the signal is set to “0” by the repeat field determination unit 33. When the repeat field detection signal S27 is given, if the reverse telecine release mode signal S37 is set to, for example, “0” that designates the first reverse telecine release process, the first reverse telecine release process is performed. Since the repeat field corresponding to the above-described comparison video field signal S12 is missing due to the non-telecine image signal,
The repeat fast field flag RF2 is set to "0", and this is set to the encoder 14 and the memory controller 2
Give to 1.

On the other hand, the repeat first field calculation unit 23 is based on the field identification information included in the vertical synchronization signal VS2 supplied from the outside in advance, and the one field first written in the memory 12 after resetting. It recognizes whether the video signal S10 is a top field or a bottom field, and stores it in an internal storage unit (not shown). The reverse telecine release mode signal S37 performs the second reverse telecine release processing. If it is set to, for example, "1", the field at the end of the telecine image signal represented by the top field fast flag TF2 given from the top field fast calculation unit 22 when the mode shifts to the normal processing mode , Is compared with the field stored in the storage unit.

When the fields match based on the comparison result, the repeat fast field computing unit 23 executes the first inverse telecine leaving process although the second inverse telecine leaving process is designated. , Sets the repeat fast field flag RF2 to “0”,
This is given to the encoder 14 and the memory controller 21. If the fields do not match based on the comparison result, a second inverse telecine departure process is executed, the repeat fast field flag RF2 is set to "1", and this is set to the encoder 14 and the memory controller 21. And a repeat field duplicate signal S38 for discarding the first two fields of the non-telecine image signal following the telecine image signal.
Give to.

In this case, when the repeat fast field computing unit 23 executes the first reverse telecine leaving process, the memory controller 21 executes the first reverse telecine leaving process in accordance with the first reverse telecine leaving process, and is set to “0”. Based on the repeat fast field flag RF2.
, The last one field of the telecine image signal and the subsequent non-telecine image signal are sequentially read out one field at a time in the order in which they were written.
Give to 4.

On the other hand, the memory controller 21
When the repeat fast field computing unit 23 executes the second inverse telecine departure process, the repeat field duplicate signal S38 is given from the repeat fast field computing unit 23, so that the repeat field duplicate signal S38 is given to the repeat field duplicate signal S38. A second reverse telecine departure process is executed based on this.

In this case, the memory controller 21 reads the last one field of the telecine image signal from the memory 12 and supplies it to the encoder 14, and then reads the first one field of the subsequent non-telecine image signal at the time when the first field should be read. Is read out again and the last one field of the telecine image signal is read out and supplied to the encoder 14, and one field supplied from the outside is supplied to the area where the first field of the non-telecine image signal is written. The video signal S10 is written, and the first field of the non-telecine image signal is discarded.

Thereafter, the memory controller 21 sets "1"
Repeat first field flag RF set to
At the time when the second field from the beginning of the non-telecine image signal is to be read from the memory 12 based on 2, the external field is supplied to the area where the second field from the beginning of the non-telecine image signal is written. The second field from the beginning of the non-telecine image signal is discarded by writing the video signal S10 for one field.

Then, from the third field from the head of the non-telecine image signal, the data is read out in the order of writing in the field unit and applied to the encoder 14, and in this way, the tail of the telecine image signal is In other words, inverse telecine processing is performed, and no inverse telecine processing is performed on non-telecine image signals.

Thus, when the encoder 14 receives the last one field of the telecine image signal from the memory 12 and the first one field of the subsequent non-telecine image signal at the time of the first inverse telecine departure processing, As one frame, and a top field fast operation unit 2 according to the last one field of the telecine image signal.
2 from the top field fast flag T
The encoded signal S14 is generated such that F2 and the repeat fast field flag RF2 set to "0" given from the repeat fast field calculation unit 23 are stored in the header.

When the non-telecine image signal for one frame subsequent to the first field of the non-telecine image signal is sequentially supplied, the encoder 14 encodes the non-telecine image signal for one frame, and then encodes the non-telecine image signal for the one frame. The coded signal S14 is generated by storing the top field first flag TF2 and the repeat first field flag RF2 corresponding to the first field of the image signal in the header.

On the other hand, when the last one field of the telecine image signal is given twice from the memory 12 at the time of the second inverse telecine departure processing, the encoder 14 codes this as one frame, and A top field fast flag TF2 representing the last one field of the telecine image signal provided from the field fast calculating section 22, and a repeat fast field calculating section 23
The coded signal S14 is generated by storing the repeat first field flag RF2 set to "1" given by the first and second in the header portion.

Then, when the non-telecine image signal for one frame is sequentially supplied from the third field from the head of the non-telecine image signal, the encoder 14 encodes the non-telecine image signal for the one frame. A top field first flag TF2 corresponding to the first field of the image signal, and a repeat first field calculation unit 23
The coded signal S14 is generated by storing the repeat fast field flag RF2 set to "0" given by the above and in the header portion.

By the way, the fixed-length operation unit 24 performs the first inverse telecine departure process and the second inverse telecine departure process from the repeat fast field operation unit 23 when shifting from the inverse telecine process mode to the normal process mode. Is given, and based on the leaving processing signal S40 and the count value of the field counter at this time, the fixed signal detected immediately before (that is, in the inverse telecine processing mode) is given. A telecine pattern from a break position of the lengthened section (hereinafter, referred to as a third break position) to a next break position (hereinafter, referred to as a fourth break position), and a non-telecine image following the telecine pattern A pattern representing the frame configuration of the signal after the first or second inverse telecine departure processing (hereinafter referred to as a non-telecine pattern) Out to.

When the frame of the non-telecine image signal is applied to the fourth delimiter position based on the detected telecine pattern and the non-telecine pattern, the fixed-length calculating section 24 determines that The length of the fixed-length section is changed by shifting the fourth delimiter position between adjacent frames, and the fixed-length unit is set to the fixed length previously set according to this change. If it is different from the conversion unit, the fixed length setting signal S20
Is given to the encoder 14, and the encoder 14 sets the fixed length unit again.

By the way, the fixed-length operation section 24 has 10 fixed-length sections in the fixed-length section after the fourth delimiter position.
Since the non-telecine image signal for the field is located, the number of frames (5 frames) when the non-telecine image signal is actually encoded is set as a fixed length unit for the fixed length section, and the fixed length unit is set. Is different from the fixed length setting unit reset when the normal processing mode is entered, the fixed length setting unit S20 is supplied to the encoder 14 again to set the fixed length unit again.

Thus, the encoder 14 outputs the video signal S13
Are fixed length-encoded in accordance with the fixed length unit set by the fixed length calculation unit 24 and are encoded.

(3) Various Processes in Encoding Device (3-1) Repeat Field Judgment Process In practice, the repeat field detection unit 13 compares the input video field signal S11 with the corresponding comparison video field signal S12. In this case, an image based on the input video field signal S11 (hereinafter referred to as an input field image) and a comparison video field signal signal S12
(Hereinafter referred to as a comparison field image) is divided into a plurality of blocks, for example, in units of 50 pixels, and the corresponding blocks of the input field image and the comparison field image are sequentially compared with each other. Get the number of miss matches.

Then, the repeat field judgment unit 33
Based on the number of mismatches obtained from the repeat field detection unit 13, it is determined whether or not the input field image matches the comparison field image, that is, whether or not the input field image is a repeat field of the comparison field image.

Accordingly, in the repeat field judgment processing for judging whether or not the input field image is a repeat field of the comparison field image, FIG.
As shown in (A) and (B), in the comparative field image HG1 (FIG. 5 (A)) and the input field image NG1 (FIG. 5 (B)) which are scenes with almost no motion, for example, a very small bird BD Since the input field image NG1 has moved, the input field image NG1 is actually a comparison field image HG1.
Even if the input field image NG1 is erroneously determined to be a repeat field because the residual is hardly obtained as a whole field only by comparison on a pixel basis as in the related art even though it is not a repeat field of Of the input field image NG1
Is not a repeat field of the comparison field image HG1.

In this repeat field determination process, as shown in FIGS. 6A and 6B, a comparative field image HG2 (FIG. 6A) composed of a scene with no motion.
And in the input field image NG2 (FIG. 6 (B)), although this input field image NG2 is actually a repeat field of the comparison field image HG2,
For example, since noise is generated, a difference of a certain level or more is obtained as a whole field only by comparison on a pixel-by-pixel basis as in the prior art, and even if the input field image NG2 is erroneously determined not to be a repeat field, The input field image NG1 is a repeat field of the comparison field image HG1 by setting the threshold value to be compared with the residual obtained by the comparison between the blocks to be determined so as not to be judged to be different from the noise level. That can be detected.

(3-2) Telecine Processing In the telecine processing for checking the telecine pattern in the telecine processing mode, a repeat field is detected from the telecine pattern generation unit 35 by the state machine 20 after a repeat field is detected in the normal processing mode. Is provided, the repeat field determination unit 33 detects a plurality of times whether or not the repeat field detection signal S27 set to “1” is provided.

When the repeat signal S30 and the repeat field detection signal S27 are sequentially given at substantially the same timing, it is determined that the video signal S10 written to the memory 12 at this time is a telecine image signal.

On the other hand, when the state machine 20 receives the repeat field detection signal S27 set to “1” at a timing different from the timing at which the repeat signal S30 is supplied, the state machine 20 becomes a still image in the memory 12, for example. It is determined that the repeat field has been erroneously detected by, for example, sequentially writing the video signal S10.

Thus, in this telecine checking process,
In the inverse telecine processing mode, it is possible to prevent erroneous execution of the inverse telecine processing on a non-telecine image signal.

(3-3) Reverse Telecine Departure Processing When shifting from the reverse telecine processing mode to the normal processing mode, either the first reverse telecine departure processing or the second reverse telecine departure processing is performed as the reverse telecine departure processing as described above. Do one or the other.

In this case, in the first reverse telecine departure process,
As shown in FIG. 7A, after the video signal S10 has, for example, a bottom field fd51 used for obtaining a repeat field of the telecine image signal TS10, a plurality of frames FR51 having a top field tf52 at the top,
FR52, FR53, FR54, FR55, FRF56
As shown in FIG. 7 (B), the non-telecine image signal HS10 composed of the following non-telecine image signal HS10 is connected to the last bottom field fd51 of the telecine image signal TS10 and the head of the non-telecine image signal HS10. A frame FR511 is constituted by the top field fd52, and the frames FR521, FR531, FR541, FR5 are arranged so that the bottom field fd53 is successively the first in this non-telecine image signal HS10.
51, FR561.

In the first inverse telecine departure process, the top field f of the non-telecine image signal HS10 is obtained.
The frame configuration in which d52 is the top is the bottom field f
Since the frame structure is changed to the frame structure in which d53 is the leading frame, these newly formed frames FR511, FR521,
By providing a top field fast flag TF2 set to "0" in accordance with FR531, FR541, FR551, and FR561 to the encoder 14, the non-telecine image signal HS starts from the end of the telecine image signal TS10.
The encoding can be sequentially performed in units of newly constructed frames up to 10.

On the other hand, in the second inverse telecine departure process, FIG. 8 having the same configuration as the video signal S10 shown in FIG.
In the video signal S10 shown in (A), as shown in FIG. 8B, the first two fields fd52 and fd53 of the non-telecine image signal HS10 are discarded, and the last bottom field fd51 of the telecine image signal TS11 is correspondingly discarded.
From the position of the top field fd52 at the head of the non-telecine image signal HS11 and the following bottom field fd
53, the bottom field f at the end of the telecine image signal TS11.
One frame F consisting of three fields using d51
R512 and the non-telecine image signal HS1
1 is the second and subsequent frames FR52, FR52
53, FR54, FR55, and FRF56.

In the second inverse telecine departure process, the last bottom field fd51 of the telecine image signal TS11 is supplied to the encoder 14 twice.
Top field first flag T set to "0"
F2 and a repeat fast field flag RF2 set to "1", and then a new non-telecine image signal HS11 and a top field fast flag TF2 set to "1" provide this telecine. From the end of the image signal to the non-telecine image signal, it can be sequentially encoded in frame units.

(3-4) Fixed Length Processing As shown in FIG. 9 (A), fixed length processing is carried out by the fixed length sections KT1 and KT2 set to 10 fields, and the length of the telecine image signal is reduced. Frame F made of fields
In a case where R61 and FR62 are divided into two fields on the front side and one field on the rear side, as shown in FIG. 9B, the length of the front fixed length section KT1 is extended by, for example, one field. The fixed length section KT3 is changed to the fixed length section KT3 of 11 fields, and the fixed length section KT2 on the rear side is changed accordingly.
Extended by one field, for example, and shifted its position
The number of fields of the telecine image signal (or non-telecine image signal) to be fixed in length is managed by changing to a fixed length section KT4 of 10 fields.

In the fixed length processing, however, the break position of the fixed length section can be changed in a range of two fields before and after a position M (hereinafter referred to as a reference position) M before the change. As shown in FIG. 9C, the fixed length sections KT1 and KT2 shown in FIG.
The length of the front fixed length section KT1 is shortened by two fields to be changed to the eight field fixed length section KT5, and the rear fixed length section KT2 is correspondingly moved two fields forward. Can be changed to the fixed length section KT6.

Therefore, in this fixed length processing, the amount of change in the length of the fixed length section is changed to the reference position M of the shifted partition position.
SUM (reference position is set to “0” and “+”
If the fixed length processing is performed a plurality of times in the coding process of the video signal S10, the value is set to "0" when the coding device 10 is reset. The length of the fixed length section is changed so that the variable SUM becomes closer to “0” each time the fixed length process is repeated.

In practice, in this fixed length processing, when the length of the fixed length section is changed when the mode shifts to the inverse telecine processing mode, the telecine pattern is detected by the fixed length calculation unit 24 as described above, The variable SUM is set based on this detection result.

In this case, the telecine patterns detected by the fixed-length operation unit 24 include the ones shown in FIGS.
As shown in (B), when the top field is a top field in the fixed length section KT8 between the first break position K1 and the second break position K2 of the fixed length section, this fixed length is used. There are five types of patterns depending on the position of the repeat field in the segmentation section KT8 (FIG. 10).
(A) to FIG. 11 (B)). Also, when the top field is a bottom field in the fixed length section KT8, there are five types of patterns depending on the position of the repeat field in the fixed length section KT8 (FIGS. 12A to 13B). )).

By the way, the fixed length calculation unit 24 can detect the repeat field the same number of times when the mode is shifted to the telecine processing mode until the mode is shifted to the inverse telecine processing mode regardless of the telecine pattern. ,
A first telecine pattern, a third telecine pattern, a fifth telecine pattern, a sixth telecine pattern, an eighth
When the telecine pattern and the tenth telecine pattern are detected, the number of times of detection of the repeat field in the telecine check processing mode is set to, for example, seven times, and the second telecine pattern, the fourth telecine pattern, and the seventh telecine pattern are set. If the telecine pattern and the ninth telecine pattern are detected, the number of times the repeat field is detected in the telecine processing mode is set to, for example, six.

That is, the telecine image signals composed of the first to tenth telecine patterns are shown in FIGS.
In the telecine processing mode, up to the frame indicated by the dotted line of the telecine image signal according to the number of times of detection of the repeat field set as shown in FIG.
Encoding is performed for each field, and from the frame indicated by the solid line of the telecine image signal, the mode is shifted to the inverse telecine processing mode, and inverse telecine processing is performed.

The fixed-length operation unit 24 first determines that the top field is a top field in the fixed-length section KT8, and that the first field is a repeat field in the first telecine pattern (FIG. 10A). Since the second delimiter position K2 is located between the front two fields and the rear one field of the frame FR71 consisting of three fields, the variable SUM is set to "-2" and the fixed length section KT8 is set. Can be reduced by two fields, or the variable SUM can be set to "+1" to extend the length of the fixed length section KT8 by one field.

In the second telecine pattern (FIG. 10B) in which the repeat field is located at the second field from the first break position K1 of the fixed length section KT8, the second break position K2 is three. The variable SUM is set to “−1” to reduce the length of the fixed-length section KT8 by one field because it is located between one field on the front side and two fields on the rear side of the frame FR72 composed of fields. Alternatively, the variable SUM can be set to “+2” to change the length of the fixed length section KT8 so as to be extended by two fields.

Further, in the third telecine pattern (FIG. 10C) in which the repeat field is located in the third field from the first break position K1 of the fixed length section KT8, the second break position K2 is 2 Since it is located between the frame FR73 consisting of fields and the frame (not shown) consisting of three fields, the length of the fixed length section KT8 is not changed. However, before the fixed length processing by the third telecine pattern is executed, the fixed length processing is already executed, and at this time, the variable SUM is set to “+1”.
Alternatively, when the length of the fixed length section is changed by setting to “+2”, since the variable SUM is managed to approach “0” as described above, the variable SUM is set to “−2”.
To reduce the length of the fixed length section by two fields.

Further, in the fourth telecine pattern (FIG. 11A) in which the repeat field is located at the fourth field from the first break position K1 of the fixed length section KT8, the second break position K2 is 2 The variable SUM is set to "-1" to reduce the length of the fixed-length section KT8 by one field, or the variable SUM is set to "+1" because it is located in the middle of the frame FR74 composed of fields. Thus, the length of the fixed length section KT8 can be changed so as to be extended by one field.

Further, in the fifth telecine pattern (FIG. 11B) in which the repeat field is located at the fifth field from the first break position K1 of the fixed length section KT8, the second break position K2 is 3 The length of the fixed length section KT8 is not changed because it is located between the frame FR75 consisting of fields and the frame (not shown) consisting of two fields.

On the other hand, if the first field is a bottom field in the fixed length section KT8, the sixth to tenth telecine patterns determined by the position of the repeat field (FIGS. 12A to 13B) ), Only the top field and the bottom field of the above-described first to fifth telecine patterns are reversed, so that the position of the repeat field from the first break position K1 is the same. The fixed length processing can be performed as in the case of the first to fifth telecine patterns.

Incidentally, the fixed length processing is also executed when the mode is shifted from the inverse telecine processing mode to the normal processing mode as described above. However, in the fixed length processing at this time,
When setting the variable SUM, in addition to the telecine pattern, the reverse telecine departure processing is performed by the first reverse telecine departure processing and the second telecine departure processing.
The information on which of the reverse telecine departure processing is executed is also used.

That is, in the fixed length processing, when the length of the fixed length section is changed in accordance with, for example, the first to fifth telecine patterns detected when the mode is shifted to the inverse telecine processing mode, the fixed length section is changed. Depending on whether the length is changed by extending or shortening the length, the fixed length sections KT10, KT1 having, for example, 10 fields corresponding to the telecine pattern divided by the changed fixed length sections.
1, a telecine pattern delimited by KT12 and KT13 follows. Incidentally, when the third and fifth telecine patterns are detected, the length of the fixed length section is not changed.

In this fixed-length processing, the telecine image signal in the video signal S10 as shown in FIG. 7A (similarly, FIG.
10, the fixed length section K
Since the last field of the telecine image signal is located in the third field from the break position K5 on the rear side of T10, the fixed length section KT15 of 10 fields following this fixed length section KT10 is separated from the first inverse teletenance. In the process, the process is not changed, and in the second reverse telecine departure process,
The length is changed to the fixed length unit KT16 extended by one field or shortened by one field.

When the telecine image signal has a fixed length section KT
11, the last field of the telecine image signal is located in the sixth field from the break position K6 on the rear side of the fixed length section KT11. In the first inverse telecine departure process, the fixed length extension section KT18 of the subsequent 10 fields is changed to a fixed length extension unit KT17 extended by one field or shortened by one field, and in the second inverse telecine departure process, Don't change it.

Further, if the telecine image signal has a fixed length section K
If the telecine image signal is divided sequentially by T12, the last field of the telecine image signal is located in the eighth field from the break position K7 on the rear side of the fixed length section KT12. The subsequent 10-field fixed length section KT20 is changed to the fixed length unit KT19 extended by 1 field or shortened by 1 field in the first inverse telecine release processing, and in the second reverse telecine release processing, Don't change it.

Further, if the telecine image signal has a fixed length section K
If the telecine image signal is divided sequentially by T13, the last field of the telecine image signal is located in the first field from the break position K8 on the rear side of the fixed length section KT13. The fixed length section KT21 of the following 10 fields is not changed in the first inverse teletenet departure process, and is extended by one field or shortened by one field in the second inverse telecine departure process. Change to KT22.

In this fixed length processing, even when the sixth to tenth telecine patterns are detected when the mode is shifted from the inverse telecine processing mode to the normal processing mode, the video signal S10 (FIG. 7A, Since only the top field and the bottom field in FIG. 8A are reversed, the fixed length processing can be performed in substantially the same manner as in the first to fifth telecine patterns.

(4) Encoding process procedure Here, in practice, as shown in FIG. 15, in this encoding device 10, when the encoding device 10 is reset,
The state machine 20 shifts from the standby state to the normal processing mode. In this state, for example, a predetermined section at the beginning is a telecine image signal from outside, and thereafter a non-telecine image signal and a telecine image signal are mixed alternately and sequentially. A signal S10 is provided.

In this case, the memory controller 21 sequentially writes the video signal S10 supplied from the outside into the memory 12 in units of fields, and starts to write the video signal S10 for one field of the second field to the memory 12; 12, the video signal S10 is read out in the order in which the video signal S10 is sequentially written in units of fields, and the readout signal is started to be supplied to the repeat field detection unit 13 as a comparison video field signal S12. , The video signal S10 is read out from the memory 12 in the order in which the video signal S10 is sequentially written in units of fields, and is given to the encoder 14.

At this time, every time the memory controller 21 reads the video signal S10 for one frame from the memory 12 by the memory controller 21, the topfield first arithmetic unit 22 sets the top field of the video signal S10 supplied from the outside to the topfield. Therefore, the top field fast flag TF2 set to "1" is given to the encoder 14, and the repeat fast field calculation unit 23
The repeat fast field flag RF2 set to “0” is given to the encoder 14.

Thus, the encoder 14 sequentially encodes the video signal S10 for each fixed-length unit consisting of 10 fields provided from the memory 12 in a frame unit so as to have a set code amount, and also has a header. A coded signal S14 is generated by storing the top field first flag TF2 and the repeat first field flag RF2 in the portion, and this is output.

Incidentally, the memory controller 21 outputs the video signal S10 of the third field supplied from the outside to the memory 1
2, the determination result signal S25 obtained by sequentially comparing the input video field signal S11 and the comparison video field signal S12 from the repeat field detection unit 13 in block units is used as the repeat field determination unit 3.
3, the repeat field judgment unit 33
When the number of mismatches obtained based on the determination result signal S25 is equal to or smaller than the threshold value, a repeat field detection signal S27 set to "1" is given to the state machine 20.

When the repeat field detection signal S27 set to "1" is given, the state machine 20 shifts to the telecine processing mode, and gives the status signal S28 to the telecine pattern generation unit 35, thereby the telecine pattern generation unit 35. 35 supplies the repeat signal S30 to the state machine 20 and the telecine counter 34 each time the video signal S10 for one field is supplied every five fields.

At this time, the fixed-length operation section 24 gives the repeat field detection signal S27 set to "1" to the telecine pattern in the fixed-length section KT30 which is first detected after the shift to the telecine processing mode. When the detected telecine pattern is, for example, the fourth telecine pattern, the count value of the telecine check counter 34 is set to 6 in accordance with the detected telecine pattern.

Thus, the telecine check counter 34
Calculates the count value as the repeat signal S30 from the telecine pattern generation unit 35 and the repeat field determination unit 3
Every time the repeat field detection signal S27 set to “1” from 3 is applied at substantially the same timing, the countdown is sequentially performed, and the repeat field detection number signal S32 is applied to the state machine 20 and the fixed length calculation unit 24.

The state machine 20 determines whether or not the repeat signal S30 and the repeat field detection signal S27 set to “1” are sequentially applied until the telecine check counter 34 counts “0”. It is determined whether or not the externally supplied video signal S10 is a telecine image signal. In a state where the externally supplied video signal S10 is determined to be a telecine image signal, a repeat representing a count value of "0" is performed. When the field detection number signal S32 is given, the mode shifts to the inverse telecine processing mode, and the inverse telecine processing is executed.

The fixed length calculation unit 24 determines the fixed length section K starting from the immediately preceding break position K31 based on the count value of the field counter detected at this time.
The telecine pattern in T31 is detected. In this case, the fixed length calculation unit 24 determines that the last two fields of the detected telecine pattern are located at the rear break position K32 of the fixed length section KT31. Change the length.

By the way, the fixed length calculation unit 24 resets the variable SUM to "0" at the time of reset, and first changes the length of the fixed length section KT31 after this reset, for example, by changing the length of the fixed length section KT31. The length is set to be extended and changed. Accordingly, the variable SUM is set to "+1", whereby the length of the fixed length section KT31 is changed to be extended by one field.

At this time, the fixed length calculation unit 24 changes the number of fields of the fixed length section KT31 from 10 fields at this time.
This fixed length section KT is changed to 11 fields
The first four fields of the telecine image signal in 31 are read out from the memory 12 as a two-frame video signal S13 and supplied to the encoder 14, and the seven fields following the four fields are converted into three frames by inverse telecine processing. Encoder 14 as video signal S13 composed of a movie signal
, The fixed length unit in the encoder 14 is not changed.

In the fixed length sections KT32 and KT33 consisting of 10 fields following the changed fixed length section KT31, the fixed length calculation section 24 converts the telecine image signal into a four-frame movie by inverse telecine processing. Since the video signal S13 is supplied to the encoder 14 as a video signal, the fixed length unit in the encoder 14 is changed from 5 frames to 4 frames prior to the encoding processing of the telecine image signal in the fixed length section KT32. Reset to.

By the way, when the repeat fast field operation section 23 shifts to the inverse telecine processing mode, the repeat fast field flag RF2 set to "0" is given to the encoder 14 as described above, but the telecine pattern generation section 23 generates the telecine pattern. When the repeat signal S30 is provided from the section 35 and the repeat field detection signal S27 set to "1" is provided from the repeat field determination section 33, the repeat fast field flag RF2 set to "1" is encoded. 14 and the memory controller 21.

The memory controller 21 sequentially reads the video signal S10 from the memory 12 in units of fields in the same manner as described above and supplies the read video signal S10 to the encoder 14, but the repeat fast field flag RF2 set to "1".
, The repeat field written in the memory 12 is discarded, the video signal S10 written in the memory 12 is read out while being converted back to the original movie signal, and is supplied to the encoder 14. .

Further, when the repeat field is discarded by the memory controller 21, the top field first calculation unit 22 determines that the first field of the video signal S 13 for one frame read from the memory 12 thereafter is different from the previous field. Since the fields are different, a top field fast flag TF2 corresponding to this is provided to the encoder 14.

As a result, when the encoder 14 shifts to the inverse telecine processing mode, first, the video signal S1 composed of the telecine image signal and the movie signal supplied from the memory 12 is output.
3 in 5-frame fixed length units, and the top field fast flag TF corresponding to the header portion thereof.
2 and a repeat first field flag RF2 are stored to generate an encoded signal S14, and then a video signal S13 composed of a movie signal supplied from the memory 12.
Is encoded in fixed-length units of four frames, and the top field fast flag TF2 corresponding to the header portion is encoded.
Further, the encoded signal S14 is generated so as to store the repeat fast field flag RF2.

Thereafter, when the video signal S10 is switched from a telecine image signal to a non-telecine image signal, for example, at the third field from the beginning of the fixed length section KT34, the repeat field detecting section 13 sets the three fields of the fixed length section KT34. Since the eye cannot be detected as a repeat field, the state machine 20 shifts to the normal processing mode based on the repeat field detection signal S27 set to “0” given from the repeat field determination unit 33 at this time, and ends the inverse telecine processing. Let it.

At this time, if the second reverse telecine departure processing is specified, the repeat first field calculation section 23 performs the field (top field) of the video signal S10 for one field initially written in the memory 12.
Is compared with the last field (bottom field) of the telecine image signal indicated by the top field fast flag TF2 provided from the top field fast calculation unit 22. In this case, since the fields do not match, "1" Is supplied to the encoder 14 and the memory controller 21 and the repeat field duplicate signal S38 is supplied to the memory controller 21.

Thus, based on the repeat field duplicate signal S38, the memory controller 21 reads out twice to repeat the last one field of the telecine image signal and supplies the same to the encoder 14.
Two fields from the beginning of the non-telecine image signal are discarded. Thereafter, the memory controller 21 sequentially reads out the second and subsequent fields from the head of the non-telecine image signal in units of fields, and supplies this to the encoder 14.

Incidentally, the fixed-length operation unit 24 includes a count value of the field counter at the time of shifting to the normal processing mode, and a separation processing signal S given from the fixed-length operation unit 24.
49, a telecine pattern in the fixed length section KT34 and a pattern of a non-telecine image signal following the telecine pattern are detected. Then, the fixed length calculation unit 24 changes the length of the fixed length section KT34 because the frame of the pattern of the detected non-telecine image signal is applied to the rear break position K33 of the fixed length section KT34. .

In this case, the fixed-length operation unit 24 sets the variable SUM to “+1” when the mode is shifted to the inverse telecine processing mode, and in this case, sets the variable SUM to “0” so as to reduce the variable SUM. As a result, the length of the fixed length section KT34 is changed so as to be shortened by one field.

However, at this time, the fixed-length operation section 24 changes the number of fields of the fixed-length section KT34 from 10 fields to 9 fields.
The first three fields in 34 are subjected to inverse telecine processing to be supplied to the encoder 14 as a video signal S13 consisting of a one-frame movie signal, and a non-telecine image signal of six fields following the three fields is converted to three frames. The fixed length unit in the encoder 14 is not changed since the video signal S13 is supplied to the encoder 14 as the video signal S13.

In the fixed length section KT35 consisting of 10 fields following the fixed length section KT34, the fixed length calculation section 24 converts the non-telecine image signal in the fixed length section KT35 into a 5-frame video signal. Since this is given to the encoder 14 as S13, this fixed length section KT
Prior to the encoding processing of the non-telecine image signal in 35, the fixed length unit in the encoder 14 is reset from 4 frames to 5 frames.

Thus, when the encoder 14 shifts to the normal processing mode, first, one frame obtained by repeating the last one field of the telecine image signal supplied from the memory 12 twice, and the non-telecine image signal 2 from the beginning of
Video signal S1 consisting of three frames after the field
3 is encoded in 4-frame fixed length units, and the corresponding top field first flag TF2 and repeat first field flag RF2 are stored in the header to generate an encoded signal S14. The video signal S13, which is a non-telecine image signal, is encoded in fixed-length units of 5 frames, and the corresponding top-field first flag TF2 and repeat-first field flag RF2 are stored in the header, thereby generating the encoded signal S14. Generate

Thereafter, the state machine 20 switches the video signal S10 from the non-telecine image signal to the telecine image signal in the fixed length section KT36, and sets the repeat field detection signal S27 set to "1" by the repeat field detection unit 13. Is given again, the mode shifts again to the telecine processing mode.

When the mode shifts to the telecine processing mode again, the fixed-length operation unit 24, based on the count value of the field counter detected at this time, determines the fixed length section detected first after shifting to the telecine processing mode. A telecine pattern in the KT 37 is detected. If the detected telecine pattern is, for example, the first telecine pattern, the fixed-length calculation unit 24 sets the count value of the telecine check counter 34 to 7 in accordance with this.

The telecine check counter 34 counts down the count value in the same manner as described above, and while detecting the repeat field seven times, the video signal S10 supplied from the outside to the state machine 20 is a telecine image signal. Let them determine if there is.

When the state machine 20 determines that the video signal S10 supplied from the outside is a telecine image signal, the state machine 20 provides a repeat field detection count signal S32 representing a count value of "0" from the telecine check counter 34. Then, the mode is shifted to the reverse telecine processing mode, and the reverse telecine processing is executed.

The fixed length calculation unit 24 determines the fixed length section K starting from the immediately preceding delimiter position K34 based on the count value of the field counter detected at this time.
The telecine pattern in T38 is detected. Then, the fixed length calculation unit 24 determines that the frame consisting of the last three fields of the detected telecine pattern is located at the rear break position K35 of the fixed length section KT38. Change the length.

In this case, since the variable SUM was set to "0" at the time of the previous inverse telecine departure processing, the fixed length calculation unit 24 sets the variable SUM to extend and change the length of the fixed length section KT38. SUM is set to “+1”, thereby changing the length of the fixed length section KT38 to be extended by one field.

At this time, the fixed length calculation unit 24 changes the number of fields of the fixed length section KT38 from 10 fields at this time.
This fixed length section KT is changed to 11 fields
In the telecine image signal 38, the first six fields are read out from the memory 12 as a three-frame video signal S13 and supplied to the encoder 14, and the five fields following the six fields are converted into two frames by inverse telecine processing. Encoder 14 as video signal S13 composed of a movie signal
, The fixed length unit in the encoder 14 is not changed.

The fixed length calculation unit 24 determines that the fixed length sections KT39, KT40, and KT41 consisting of 10 fields subsequent to the changed fixed length section KT36 are the same as those of the fixed length sections KT32 and KT33 described above. Similarly, since the telecine image signal is subjected to inverse telecine processing and given to the encoder 14 as a 4-frame video signal S13,
Prior to the encoding process of the four-frame video signal S13, the fixed length unit in the encoder 14 is reset from five frames to four frames.

Thus, when the encoder 14 shifts to the inverse telecine processing mode, the encoder 14 first receives the video signal S1 composed of the telecine image signal and the movie signal supplied from the memory 12.
3 in 5-frame fixed length units, and the top field fast flag TF corresponding to the header portion thereof.
2 and a repeat first field flag RF2 are stored to generate an encoded signal S14, and then a video signal S13 composed of a movie signal supplied from the memory 12.
Is encoded in fixed-length units of four frames, and the top field fast flag TF2 corresponding to the header portion is encoded.
Further, the encoded signal S14 is generated so as to store the repeat fast field flag RF2.

Thereafter, when the video signal S10 is switched from a telecine image signal to a non-telecine image signal at, for example, the third field from the beginning of the fixed length section KT42, the repeat field detection section 13 causes the five fields of the fixed length section KT34 to change. Since the eye cannot be detected as a repeat field, the state machine 20 returns to the normal processing mode again based on the repeat field detection signal S27 set to “0” given from the repeat field determination unit 33 at this time, and performs the inverse telecine processing. finish.

At this time, if the first inverse telecine departure process is designated, the repeat fast field calculation unit 23 sets the repeat fast field flag RF2 to “0”, and sets this to the encoder 14 and the memory. To the controller 21 and the memory controller 21
Reads the non-telecine image signal from the memory 12 in units of fields based on the repeat fast field flag RF2 set to "0", supplies the read signal to the encoder 14, and furthermore, the top field fast operation unit 22 is a top field fast flag TF set for the last frame of the telecine image signal.
2 is supplied to the encoder 14 as it is.

The fixed-length operation unit 24 counts the count value of the field counter at the time of transition to the normal processing mode, and the separation processing signal S given from the fixed-length operation unit 24.
49, a telecine pattern in the fixed length section KT42 and a pattern of a non-telecine image signal following the telecine pattern are detected. In this case, the fixed-length operation unit 24 determines that the detected telecine pattern and the non-telecine image signal have a fixed-length section KT without multiplying a frame at the separation position.
42, the length of the fixed length section KT42 is not changed.

At this time, the fixed-length operation unit 24 performs the inverse telecine processing on the first two fields in the fixed-length section KT42, although the number of fields in the fixed-length section KT42 does not change to 10 fields. The encoder 1 is used as a video signal S13 composed of a one-frame movie signal.
4 and a non-telecine image signal of 8 fields following the 2 fields is a video signal S of 4 frames.
Since 13 is provided to the encoder 14, the fixed length unit in the encoder 14 is reset from 4 frames to 5 frames. Then, since a 10-field non-telecine image signal is located in the fixed length section KT43, the fixed length unit in the encoder 14 is fixed at 5 frames.

Thus, when the encoder 14 shifts to the normal processing mode, the encoder 14 first receives one frame of the movie signal supplied from the memory 12 and 4 frames from the head of the non-telecine image signal.
The video signal S13 composed of frames is encoded in fixed length units of 5 frames, and the corresponding top field fast flag TF2 and repeat fast field flag RF2 are stored in the header to generate an encoded signal S14. Then, the video signal S13, which is a non-telecine image signal provided from the memory 12, is encoded in fixed-length units of 5 frames, and the corresponding top field fast flag TF2 and repeat fast field flag RF2 are stored in the header. As a result, an encoded signal S14 is generated.

In this way, in the encoding device 10, the video signal S10 supplied from the outside is switched from the non-telecine image signal to the telecine image signal again, and the fixed length unit and the top are maintained until the operation is shifted to the inverse telecine processing mode. The setting of the field first flag TF2 and the setting of the repeat first field flag RF2 are maintained, and in this state, the video signal S10 is encoded.

(5) Operation and effect of the present embodiment In the above configuration, the encoding device 10 corresponds to the input video field signal S11 via the differentiator 27 and the adder 28 during the repeat field determination processing. The comparison image field signal S12 is sequentially compared with each other on a block basis, and the difference absolute value obtained by the comparison between the blocks is compared with a predetermined threshold value by the determination unit 30, so that the blocks are different from each other. The number of mismatches determined as "?" Is detected.

By giving the number of miss matches to the repeat field judging unit 33, the repeat field judging unit 33 compares the number of miss matches with a predetermined threshold value, based on the comparison result obtained. To detect.

In this case, the repeat field detection unit 13 applies the above-mentioned FIG. 6A to the input video field signal S11 and the corresponding comparison video field signal S12.
Even if noise occurs as a whole as shown in (B), the threshold value set in the determination unit 30 for comparison of the absolute value of the difference is set to the difference of the relatively large value obtained from the entire field as in the related art. Rather than a value set for the absolute value, the difference between the corresponding blocks differs according to the absolute value of the difference between the relatively small values obtained by comparing the blocks markedly smaller than the whole field. May be selected so as to prevent erroneous determination that

By setting the threshold value of the determination unit 30 in this manner, for example, as shown in FIGS. 5A and 5B, the input video field signal S11 and the corresponding comparison video field signal S12 are very different. Small bird BD
Even if there is little movement in other parts only when the movement of the bird BD is detected, a comparison result between the block in which the bird BD is located and the corresponding block can detect a difference absolute value larger than the difference absolute value due to noise. Therefore, it is possible to determine that the corresponding blocks are different.

Therefore, in this repeat field determination processing, even if noise occurs or the input video field signal S11 and the corresponding comparison video field signal S12 have only a slight movement, the mismatch obtained from the determination unit 30 can be obtained. It is possible to accurately determine whether or not the input video field signal S11 is a repeat field by distinguishing them according to the number of times.

Since the telecine image signal is subjected to the inverse telecine processing based on the repeat field detected in this manner, the telecine image signal and the non-telecine image signal are prevented from being erroneously inversely telecine processed. It is possible to prevent the image based on the image signal S10 from being disturbed.

According to the above arrangement, the input video field signal S11 and the comparison video field signal S12 are sequentially compared in units of blocks, and the number of times that the corresponding blocks are determined to be different based on the obtained comparison result is determined. Detected, by determining whether or not the input video field signal S11 is a repeat field based on the number of times detected, noise has occurred in the input video field signal S11 and the comparison video field signal S12, or Even if there is only a slight movement, it is possible to accurately determine whether or not the input video field signal S11 is a repeat field based on the number of times that the corresponding blocks are determined to be different from each other, thereby detecting a repeat field. Telecine image based on field No. signal detection apparatus and a signal detection method capable conversion is processed to return correctly to the original movie signal can be realized.

(6) Other Embodiments In the above-described embodiment, a case has been described in which the present invention is applied to the encoding device 10. However, the present invention is not limited to this, and the present invention is not limited to this. Such as an encoding device to which only the telecine image signal or the video signal S10 is supplied on a recording medium, a transmission device for transmitting the telecine image signal or the video signal S10, and the like. The present invention can be applied to various other devices as long as the device performs a conversion process to return to a signal.

In the above-described embodiment, the case has been described where the input video field signal S11 and the corresponding comparison video field signal S12 are sequentially compared in the block unit by the repeat field detection unit 13. The present invention is not limited to this, and the input video field signal S11 and the corresponding comparison video field signal S12 may be sequentially compared on a line basis with a predetermined number of pixels.

Further, in the above-described embodiment, a differentiator 27 is used as comparing means for sequentially comparing a predetermined field of a telecine image signal and a field at a position separated by a corresponding predetermined number of fields in units of a predetermined number of pixels. Part 2
8. The case where the control unit 29 is applied has been described. However, the present invention is not limited to this, and a predetermined field of a telecine image signal and a field at a position corresponding to a predetermined predetermined number of fields are sequentially converted to a predetermined number of pixels. As long as the comparison can be performed in units, a comparing means having various other configurations may be applied.

Further, in the above-described embodiment, it is determined whether or not the corresponding predetermined number of pixels are the same based on the comparison result obtained from the comparing means, and based on the determination result, the corresponding predetermined number of pixels is determined. Has been described in which the control unit 29 and the determination unit 30 are applied as the number-of-times detecting means for detecting the number of times that it is determined that they are different. However, the present invention is not limited to this, and based on the comparison result obtained from the comparing means. It is determined whether or not the corresponding predetermined number of pixels are the same, and if it is possible to detect the number of times that the corresponding predetermined number of pixels is determined to be different based on the determination result,
In addition, a frequency detecting means having various configurations may be applied.

Further, in the above-described embodiment, the repeat field determination unit 33 is applied as a field detection unit for detecting a second field (ie, a repeat field) based on the detection result of the number of times obtained from the number of times detection unit. I mentioned the case where I tried to do
The present invention is not limited to this. If the second field can be detected based on the detection result of the number of times obtained from the number of times detecting means, a field detecting means having various other configurations may be applied. good.

[0181]

As described above, according to the present invention, a comparison means for sequentially comparing a predetermined field of a telecine image signal and a field at a position separated by a corresponding predetermined number of fields in units of a predetermined number of pixels, and this comparing means A number detection means for determining whether the corresponding predetermined number of pixels are the same based on the comparison result obtained from the means, and detecting the number of times the corresponding predetermined number of pixels are determined to be different based on the determination result; By providing the field detection means for detecting the second field based on the detection result of the number of times obtained from the number detection means, noise is generated in the field, or slight movement between the fields. Even when there are only two pixels, the second field can be accurately detected based on the number of times that the predetermined number of pixels is determined to be different, and thus the second field can be detected. Field can be achieved signal detection apparatus capable of conversion to return correctly to the original movie signal telecine image signal based on.

A first step of sequentially comparing a predetermined field of the telecine image signal and a field at a position separated by a predetermined number of corresponding fields in units of a predetermined number of pixels;
Based on the comparison result obtained in the first step, it is determined whether or not the corresponding predetermined number of pixels are the same, and based on the determination result, the number of times that the corresponding predetermined number of pixels is determined to be different is detected. By providing a second step for performing the operation and a third step for detecting the second field based on the number of detections obtained in the second step, noise may be generated in the field. , Or even when there is only slight movement between the fields, the second field can be accurately detected by the number of times that the predetermined number of pixels are determined to be different. Thus, the telecine image signal can be detected based on the second field. Can be realized so that the signal can be converted so as to correctly restore the original movie signal.

[Brief description of the drawings]

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

FIG. 2 is a block diagram showing a configuration of a controller.

FIG. 4 is a conceptual diagram for explaining a mode transition in a state machine.

FIG. 4 is a block diagram showing a configuration of a repeat field detector.

FIG. 5 is a schematic conceptual diagram illustrating a repeat field determination process.

FIG. 6 is a schematic conceptual diagram illustrating a repeat field determination process.

FIG. 7 is a schematic conceptual diagram illustrating a first reverse telecine departure process;

FIG. 8 is a schematic conceptual diagram illustrating a second reverse telecine departure process;

FIG. 9 is a schematic conceptual diagram illustrating fixed lengthening processing.

FIG. 10 is a schematic conceptual diagram for explaining a telecine pattern when the top is a top field;

FIG. 11 is a schematic conceptual diagram for explaining a telecine pattern when the top is a top field;

FIG. 12 is a schematic conceptual diagram for explaining a telecine pattern when a head is a bottom field.

FIG. 13 is a schematic conceptual diagram for explaining a telecine pattern when a head is a bottom field.

FIG. 14 is a schematic conceptual diagram illustrating fixed length processing at the time of inverse telecine processing according to various telecine patterns.

FIG. 15 is a schematic conceptual diagram showing an encoding processing procedure.

FIG. 16 is a schematic conceptual diagram for explaining a telecine image signal.

FIG. 17 is a block diagram showing a configuration of a conventional encoding device.

FIG. 18 is a block diagram showing a configuration of a conventional repeat field detection unit.

FIG. 19 is a schematic conceptual diagram used for describing a telecine image signal encoding process in the encoding device.

FIG. 20 is a schematic conceptual diagram used to describe an encoding process of a non-telecine image signal in the encoding device.

FIG. 21 is a schematic conceptual diagram illustrating a conventional repeat field determination process.

FIG. 22 is a schematic conceptual diagram illustrating a conventional repeat field determination process.

[Explanation of symbols]

10 ... Encoding device, 11 ... Controller, 12 ...
Memory, 13 ... Repeat field detector, 14 ...
Encoder 27 Differentiator 28 Adder 29 29
Control unit 30, determination unit 33, repeat field determination unit, S10, S13 video signal, S11 input video field signal, S12 comparison video field signal, S14 encoded signal, S25 … Judgment result signal,
S27: Repeat field detection signal.

[Procedure amendment]

[Submission date] April 17, 1998

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Brief description of drawings

[Correction method] Change

[Correction contents]

[Brief description of the drawings]

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

FIG. 2 is a block diagram showing a configuration of a controller.

FIG. 3 is a conceptual diagram for explaining a mode transition in a state machine.

FIG. 4 is a block diagram showing a configuration of a repeat field detector.

FIG. 5 is a schematic conceptual diagram illustrating a repeat field determination process.

FIG. 6 is a schematic conceptual diagram illustrating a repeat field determination process.

FIG. 7 is a schematic conceptual diagram illustrating a first reverse telecine departure process;

FIG. 8 is a schematic conceptual diagram illustrating a second reverse telecine departure process;

FIG. 9 is a schematic conceptual diagram illustrating fixed lengthening processing.

FIG. 10 is a schematic conceptual diagram for explaining a telecine pattern when the top is a top field;

FIG. 11 is a schematic conceptual diagram for explaining a telecine pattern when the top is a top field;

FIG. 12 is a schematic conceptual diagram for explaining a telecine pattern when a head is a bottom field.

FIG. 13 is a schematic conceptual diagram for explaining a telecine pattern when a head is a bottom field.

FIG. 14 is a schematic conceptual diagram illustrating fixed length processing at the time of inverse telecine processing according to various telecine patterns.

FIG. 15 is a schematic conceptual diagram showing an encoding processing procedure.

FIG. 16 is a schematic conceptual diagram for explaining a telecine image signal.

FIG. 17 is a block diagram showing a configuration of a conventional encoding device.

FIG. 18 is a block diagram showing a configuration of a conventional repeat field detection unit.

FIG. 19 is a schematic conceptual diagram used for describing a telecine image signal encoding process in the encoding device.

FIG. 20 is a schematic conceptual diagram used to describe an encoding process of a non-telecine image signal in the encoding device.

FIG. 21 is a schematic conceptual diagram illustrating a conventional repeat field determination process.

FIG. 22 is a schematic conceptual diagram illustrating a conventional repeat field determination process.

[Description of Codes] 10 ... Encoding device, 11 ... Controller, 12 ...
Memory, 13 ... Repeat field detector, 14 ...
Encoder 27 Differentiator 28 Adder 29 29
Control unit 30, determination unit 33, repeat field determination unit, S10, S13 video signal, S11 input video field signal, S12 comparison video field signal, S14 encoded signal, S25 … Judgment result signal,
S27: Repeat field detection signal.

Claims (6)

[Claims] When generating a telecine image signal from a telecine image signal obtained by converting a movie signal into a video signal, a predetermined first field of the movie signal is repeated and inserted at a position separated by a predetermined number of fields. A signal detecting device for detecting the obtained second field; a comparing means for sequentially comparing a predetermined field of the telecine image signal and a field at a position away from the corresponding predetermined field number by a predetermined pixel number unit; Number of times of judging whether the corresponding predetermined number of pixels are the same based on the comparison result obtained from the comparing means and detecting the number of times that the corresponding predetermined number of pixels are determined to be different based on the judgment result Detecting means; and a field for detecting the second field based on a detection result of the number of times obtained from the number of times detecting means. Signal detecting apparatus characterized by comprising left and means. 2. The apparatus according to claim 1, wherein the comparing means compares the field of the telecine image signal with the field which is separated by a predetermined number of fields from the telecine image signal in blocks of a predetermined number of pixels. 3. The signal detection device according to claim 1. 3. The apparatus according to claim 1, wherein said comparing means sequentially compares the field of the telecine image signal and the field separated by a predetermined number of fields in a line unit having a predetermined number of pixels. 3. The signal detection device according to claim 1. 4. When a telecine image signal is generated from a telecine image signal obtained by converting a movie signal into a video signal, a predetermined first field of the movie signal is repeated and inserted at a position separated by a predetermined number of fields. A signal detecting method for detecting the obtained second field, comprising: a first step of sequentially comparing a predetermined field of the telecine image signal and a field at a position away from the corresponding predetermined field by a predetermined number of pixels. It is determined whether or not the corresponding predetermined number of pixels are the same based on the comparison result obtained in the first step, and it is determined that the corresponding predetermined number of pixels is different based on the determination result. A second step of detecting the number of times, and the second field based on the detection result of the number of times obtained in the second step. Third signal detecting method characterized by comprising the step of detecting a. 5. In the first step, the field of the telecine image signal and the field at a position separated by a predetermined number of fields corresponding to the telecine image signal are sequentially compared in units of blocks each including a predetermined number of pixels. The signal detection method according to claim 4. 6. The apparatus according to claim 4, wherein said comparing means compares the field of the telecine image signal with the field which is away from the corresponding predetermined number of fields by a predetermined number of pixels. 3. The signal detection method according to 1.