JPH08214315A - Encoded image cut detector - Google Patents

Abstract

PURPOSE: To provide an encoded image cut detector which detects the cut of an encoded image with a small scale, a low cost and at high speed. CONSTITUTION: This detector is equipped with a control part 3 which sets an encoded image data group comprised of encoded image data encoded by variable length encoding as an input image data group, and generates a control signal for the extraction of data quantity of each image from the input image data group, a counter 2 which sets the input image data group and the control signal from the control part 3 as input, and counts the data quantity of a part or all the input image data corresponding to the control signal from the control part 3, a storage part 4 which stores at least one of values of the counter 2, and a comparison arithmetic part 5 which decides the presence/absence of the cut between the image data in the input image data group by comparing the value stored in the storage part 4 with a threshold value set in advance, or comparing plural values stored in the storage part 4 with a value obtained by comparison operation and the threshold value set in advance.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a coded image cut detecting device for detecting a cut in a coded image data group.

[0002]

2. Description of the Related Art As a highly efficient coding method for image data,
There is a variable length coding system. There is a demand for a coded image cut detection device for supporting content search and processing / editing of image data coded by this variable length coding method. An example of a conventional coded image cut detection device will be described below with reference to the drawings.

FIG. 14 shows the configuration of a conventional coded image cut detection device. In FIG. 14, 1 is a recording medium for recording image data already encoded by variable-length encoding, 8 is a decoding unit for decoding encoded image data, 4 is a memory unit, and 4C and 4D are already used. An image data memory for storing the decoded image data, and 5 is a comparison calculation unit for comparing and calculating each image data output from the image data memory 4C and the image data memory 4D to determine the presence or absence of a cut. Is.

The operation of the coded image cut detection device configured as described above will be described below with reference to FIGS. 14 and 15. FIG. 15 shows the structure of encoded image data. First, FIG. 15 will be described. The intra-picture coded data I obtained by coding the intra-picture information by the variable length coding is indicated by I () in FIG. 15, and the difference between the reference picture and the target picture is defined by using the picture located in front in the reproduction time as the reference picture. The forward predictive coded data P in which the value is encoded by the variable length coding is indicated by P () in FIG. 15, and the forward image or the backward image or the interpolated images of both the forward and backward images at the reproduction time are referred to. As the image, bidirectional predictive encoded data B obtained by encoding the difference value between the reference image and the image of interest by variable length encoding is denoted by B () in FIG.
Indicated by A coded image data group composed of only the intra-image coded data I is shown in the structure A in FIG.
An encoded image data group composed of intra-image encoded data I and forward predictive encoded data P is shown in configuration B in FIG. In addition, a coded image data group composed of intra-coded coded data I, forward predictive coded data P, and bidirectional predictive coded data B is shown in configuration C in FIG.
In the following description, I (), P (), and B () are referred to as image unit I (), image unit P (), and image unit B (), respectively.
The numbers in parentheses () in each image in FIG. 15 indicate the order in which the images are reproduced. In the configuration C, for example, the image unit B
To decode (13) and the image unit B (14),
Since the image unit I (12) and the image unit P (15) must be decoded first, the image positioned in the rear in the reproduction order than the image unit B (13) in consideration of the decoding. The unit P (15) is rearranged so as to come before the image unit B (13). The image units in the other parts are rearranged for the same reason. Further, what is obtained by arbitrarily dividing each image is a primary divided area, and what is obtained by arbitrarily dividing all the primary divided areas is a secondary divided area, and the forward predictive encoded data and the bidirectional predictive encoded data are It is encoded by taking the difference from the reference image in units of each secondary divided area.

The unit of the image coded in FIG. 15 may be a frame or a field.

In the following description of the operation, it is assumed that the coded image data having the structure A shown in FIG. 15 is recorded on the recording medium 1 shown in FIG. 14 as an example.

In FIG. 14, first, the image unit I (10) of the configuration A of FIG. 15 is read from the recording medium 1.
The read image unit I (10) is input to the decoding unit 8 and is decoded here. The SW1 is initially connected to the a side, and the image unit I (1
The 0) decoded image data is stored in the image data memory 4C. Image unit I (1
When all the decoded image data of 0) is stored in the image data memory 4C, SW1 is switched to the b side.

Next, from the recording medium 1, the image unit I (1
1) is read. Read image unit I (11)
Is decoded by the decoding unit 8. Since SW1 is connected to b, the decoded image data of the image unit I (11) is stored in the image data memory 4D. When all the decoded image data of the image unit I (11) output from the decoding unit 8 is stored in the image data memory 4D, SW1 is switched to the side a. Next, the decoded image data is read from the image data memory 4C and the image data memory 4D,
It is input to the comparison calculation unit 5. The comparison calculation unit 5 compares the values input from the image data memory 4C and the image data memory 4D, and between the decoded image data of the image unit I (10) and the decoded image data of the image unit I (11). Determine whether there is a cut.

Thereafter, each time the storage of either the image data memory 4C or the image data memory 4D is updated by the same procedure as described above, the comparison operation unit 5 inputs from the image data memory 4C and the image data memory 4D. The determined values are compared to determine whether or not there is a cut between the decoded image data stored in the image data memory 4C and the decoded image data stored in the image data memory 4D.

[0010]

However, in the above-mentioned configuration, the encoded image data read from the recording medium 1 in FIG. 14 must be decoded.
Since the comparison calculation unit 5 cannot perform the comparison calculation for detecting the cut and the process of determining the presence or absence of the cut,
Decoding section 8 was required.

If the decoding unit 8 decodes moving image data for 2 hours in real time, that is, for 2 hours, it requires at least 2 hours or more to detect all cuts. And lacks high speed. In addition, when the time required for the decoding in the decoding unit 8 is to be reduced to, for example, one half, the decoding unit 8 must be doubled in scale, and the decoding unit 8 is large and high in size. It becomes a cost thing.

The memory unit 4 stores a large amount of decoded image data.
Therefore, a huge amount of image data memory is required in the memory unit 4, which also causes a large scale and high cost.

Further, since the comparison calculation unit 5 performs comparison calculation and cut presence / absence determination processing using a large amount of decoded image data, these processings require a great deal of time.

To summarize the above, the conventional coded image cut detecting device as shown in FIG. 14 has the problems that it lacks in high speed operation and is large in scale and high in cost.

An object of the present invention is to solve the above problems and to provide a coded image cut detection device which detects a cut between coded image data at a small scale and at low cost and at high speed.

[0016]

[Means for Solving the Problems]

(1) In order to solve the above problems, the coded image cut detection apparatus of the present invention uses a coded image data group composed of coded image data coded by variable length coding as an input image data group. A control unit for generating a control signal for extracting the data amount of each image from the input image data group, and a control signal from the input image data group and the control unit as input, A counter that counts a part or all of the data amount of the input image data group according to a control signal, a storage unit that stores at least one value of the counter, and a value that is stored in the storage unit in advance. Image data in the input image data group by comparing a value obtained by comparing a plurality of values stored in the storage unit with a preset threshold value. It is obtained by a configuration having a comparison operation unit for determining the presence or absence of cut between data.

(2) In order to solve the above problems, the coded image cut detection apparatus of the present invention uses a coded image data group composed of coded image data coded by quantization as an input image data group. And a control unit for generating a control signal for extracting a quantization parameter of each image from the input image data group, and a control signal from the input image data group and the control unit, A quantization parameter detection unit for detecting a part or all of the quantization parameters of the input image data group according to the control signal from
A storage unit that stores at least one value of the quantization parameter, compares the value stored in the storage unit with a preset threshold value, or compares and operates a plurality of values stored in the storage unit. It is configured to have a comparison calculation unit that determines the presence or absence of a cut between the image data in the input image data group by comparing the obtained value with a preset threshold value.

(3) In order to solve the above problems, the coded image cut detection apparatus of the present invention is a coded image coded by variable length coding, quantization, or variable length coding and quantization. A control unit for generating a control signal for extracting a data amount of each image from the input image data group or detecting a quantization parameter, using a coded image data group composed of data as an input image data group. A counter that receives the input image data group and a control signal from the control unit as an input and counts a part or all of the data amount of the input image data group according to the control signal from the control unit; A quantization parameter detection unit that detects a quantization parameter of part or all of the data group, and a storage unit that stores at least one value obtained by multiplying the value of the counter by the value of the quantization parameter. By comparing a value stored in the storage unit with a preset threshold value, or by comparing a value obtained by performing a comparison operation on a plurality of values stored in the storage unit with a preset threshold value, It is configured to have a comparison operation unit that determines whether or not there is a cut between image data in the input image data group.

[0019]

[Action]

(1) With the above-described configuration, the present invention utilizes the characteristic that the amount of coded image data is significantly different between the image unit before cutting and the image unit after cutting in the input image data group, and control is performed. The control unit generates a control signal for extracting the data amount of each image unit, the counter counts the data amount of each image unit, stores the value in the storage unit, and the comparison calculation unit stores it in the storage unit. The image in the input image data group by comparing the preset value with a preset threshold value or a value obtained by comparing and computing a plurality of values stored in the storage unit. Determine whether there is a cut between data.

As a result, the work of decoding the coded image data read from the recording medium 1 which is necessary in the conventional example shown in FIG. 14 is unnecessary, and instead, the data amount for extracting the data amount of each image unit is extracted. Since the counting work is performed, it becomes possible to detect the cut in a shorter time.
Further, unlike the conventional example, a large-scale image data memory for storing the decoded image data is not required, and only the value of the data amount need be stored instead. And the cost can be reduced. Furthermore, the comparison calculation unit 5
In the conventional example, the comparison calculation processing for detecting a cut is performed using a huge amount of image data, whereas it is only necessary to compare the values of a small amount of data. The time required for can be greatly reduced.

In summary, it is possible to speed up the process, reduce the size of the device, and reduce the cost of the entire device, as compared with the conventional example. Further, the cut detection is effective for encoded data whose data amount changes depending on the cut.

(2) According to the present invention, the quantization parameter of the coded image data is significantly different between the image unit before the cut and the image unit after the cut in the input image data group due to the above configuration. Using the characteristics, the control unit generates a control signal for detecting the quantization parameter of each image unit, the quantization parameter detection unit detects the quantization parameter of each image unit, and stores the value in the storage unit. A value obtained by comparing and storing a value stored in the storage unit and a preset threshold value by a comparison calculation unit or a plurality of values stored in the storage unit and a preset threshold value. The presence or absence of a cut between the image data in the input image data group is determined by comparing.

As a result, the work of decoding the coded image data read from the recording medium 1 which is required in the conventional example shown in FIG. 14 becomes unnecessary, and instead the work of detecting the quantization parameter for each image is performed. , It becomes possible to detect the cut in a shorter time. Further, unlike the conventional example, a large-scale image data memory for storing the image data after decoding is not required, and only the value of the quantization parameter is stored instead. And cost reduction are possible. Further, in the comparison calculation unit 5, in the conventional example, a comparison calculation process for detecting a cut is performed using a large amount of image data, whereas it is only necessary to compare the values of a small number of quantization parameters. The time required for the processing in the comparison calculation unit 5 can be greatly reduced.

In summary, it is possible to speed up the process, reduce the size of the device, and reduce the cost of the entire device, as compared with the conventional example. Further, even if the data amount does not change due to the cut, the quantization parameter is effective for the coded data that changes due to the cut.

(3) According to the present invention, the data amount of the encoded image data or the quantization parameter is different between the image unit before cutting and the image unit after cutting in the input image data group. Utilizing the feature of being significantly different, the control unit generates a control signal for extracting the data amount of each image unit and detecting the quantization parameter, counts the data amount with the counter, and uses the quantization parameter detection unit for each. The quantization parameter for each image is detected, the count value is multiplied by the value of the quantization parameter, the value is stored in the storage unit, and the value stored in the storage unit is preset by the comparison operation unit. In the input image data group by comparing a threshold value set in advance with a value obtained by performing a comparison operation on a plurality of values stored in the storage unit. Determining the presence or absence of a cut between the image data.

As a result, the work of decoding the coded image data read from the recording medium 1 which is necessary in the conventional example shown in FIG. 14 becomes unnecessary, and instead the extraction of the data amount of each image unit and the quantization parameter setting are performed. Since the detection is performed and the data amount and the quantization parameter are calculated, the cut can be detected in a shorter time. or,
A large-scale image data memory for storing the image data after decoding as in the conventional example is not required, and instead only the data amount and the calculated value of the quantization parameter are stored. It enables downsizing and cost reduction.
Further, in the comparison calculation unit 5, in the conventional example, a comparison calculation process for detecting a cut is performed using a large amount of image data, whereas only a small amount of data and a calculation value of a quantization parameter are compared. Therefore, the time required for the processing in the comparison calculation unit 5 can be significantly reduced. Summary,
Compared with the conventional example, the entire device can achieve faster processing, smaller device size, and lower cost. Further, the present invention can be applied to both cases where the input data is characterized by a change in the data amount and cases where it is characterized by a change in the quantization parameter, and it is not necessary to understand the characteristics of the input data in advance.

[0027]

【Example】

(Embodiment 1) FIG. 1 shows the configuration of a coded image cut detection apparatus according to the first embodiment of the present invention.

In FIG. 1, 1 is a recording medium for recording image data already encoded by variable-length encoding, 2 is a counter for counting the amount of encoded image data read from the recording medium 1, and 3 is Control part,
An internal unit 3A of the control unit 3 is an image unit detection unit for detecting a start position and an end position of an image unit for one frame or one field in the encoded image data read from the recording medium 1, and 4 is a memory unit. 4A and 4B are memories that store the value indicated by the counter 2, that is, the amount of data, and 5 is a comparison calculation of the values of the respective amounts of data output from the memories 4A and 4B, and the presence / absence of a cut is compared. It is a calculation unit.

The operation of the coded image cut detecting device configured as described above will be described below with reference to FIGS. 1 and 9. Note that FIG. 9 is the same as FIG. 15 used in the description of the conventional coded image cut detection device.

In the present embodiment, the presence / absence of a cut is determined between image data that has been intra-coded as in the configuration A in FIG. From the recording medium 1 of FIG. 1 to FIG.
Of the encoded image data shown in FIG.
Data is sequentially read from the beginning of (10) and input to the image unit detection unit 3A and the counter 2 in FIG. The image unit detection unit 3A first detects that the data is the head data of an image unit, and transmits a zero clear signal and a count start signal to the counter 2 through the line f. When the counter 2 receives the zero clear signal and the count start signal, the counter 2 clears the count value to zero and starts counting the data amount of the image unit I (10) input through the line e. The image unit I (10) is read from the recording medium 1 one after another, the image unit detection unit 3A detects the end of the image unit I (10), and a count stop signal is sent to the counter 2 through the line f in FIG. Then, the counting by the counter 2 is stopped. And since SW1 is initially connected to the a side,
The count value of the counter 2 is stored in the memory 4A.
Then, SW1 is switched to the b side.

Next, of the encoded image data shown in FIG. 9 from the recording medium 1, the data is sequentially read from the beginning of the image unit I (11) of the configuration A, and the image unit detection unit 3A and the counter of FIG. Entered in 2. Image unit detection unit 3A
Then, first, it is detected that it is the head data of each image unit, and the zero clear signal and the count start signal are transmitted to the counter 2 through the line f. When the counter 2 receives the zero clear signal and the count start signal, the counter 2 clears the count value to zero and starts counting the amount of coded image data input through the line e. Image units I (11) are read from the recording medium 1 one after another, and eventually the image unit detection unit 3A
Detects the end of the image unit I (11), a count stop signal is sent to the counter 2 through the line f, and the counter 2
Counting is stopped. At this time, since SW1 is connected to the b side, the count value of the counter 2 is stored in the memory 4B. Then, SW1 is switched to the a side.

Next, a comparison operation of the values stored in the memory 4A and the memory 4B in the comparison operation unit 5 is performed, and the reproduced image of the image unit I (10) and the image unit I (1) of the configuration A of FIG.
Whether or not there is a cut between the reproduced images in 1) is determined. The process inside the comparison operation unit 5 will be described below. First, the image unit I (10) of the configuration A of FIG. 9 stored in the memories 4A and 4B of FIG. 1 respectively.
Then, the absolute value α of the difference between the data amount of 1 and the data amount of the image unit I (11) is obtained. If the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11) are exactly the same, the image unit I (10) and the image unit I (11) have the same value, and the data The amount is exactly the same. Therefore, at this time, the value of α becomes 0. Even if the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11) are not exactly the same, the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11) are If there is no cut, the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11) are generally similar to each other, and even though they are encoded by the variable length encoding method, The data amount of the image unit I (10) and the data amount of the image unit I (11) are close to each other, and α is a value close to 0.

On the other hand, when there is a cut between the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11), there is no correlation between the reproduced images and the variable length coding method is used. Since the data is encoded, the data amount of the image unit I (10) and the data amount of the image unit I (11) are generally quite different values. In that case, the value of α is much larger than 0. Become. Therefore, the comparison calculation unit 5 sets a certain threshold value β, and when the value of α is larger than the threshold value β, there is a cut between the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11). The cut detection signal is turned on and output, and if the value of α is less than or equal to the threshold value β, the image unit I
It is determined that there is no cut between (10) and the image unit I (11), and the cut detection signal is turned off and output.

Image unit I (10) in the comparison operation unit 5
When the determination of the presence / absence of the cut between the reproduced image of No. 1 and the reproduced image of the image unit I (11) is completed, the next encoded image data is read from the recording medium 1 thereafter, and the counter 2 is read in the same procedure as above. The amount of data is counted in the memory 4A
The new value of the counter 2 is stored and updated in the one of the memory 4B and the memory 4B that stores the data amount of the old encoded image data in the order of the reproduced image. A process is performed to determine the presence / absence of a cut between reproduced images.

As an example, FIG. 11 shows an intra-image coded data from a series of coded image data groups having 6 cuts in a reproduced image and coded by the same structure as the structure C of FIG. 7 is a graph showing only the absolute value of the difference in the data amount between the intra-image coded data after extracting only the data. In FIG. 11, X
The axis shows the number of frames of the reproduced image, and the Y axis shows the absolute value of the difference in the data amount between the intra-image encoded data. The six cuts are indicated by □ in FIG. As is clear from FIG. 11, a large value appears on the Y-axis where there is a cut. Therefore, by setting a certain threshold β and comparing the Y-axis value with the threshold β, the presence or absence of the cut can be determined.

Note that the comparison calculation function of the values input from the memories 4A and 4B in the comparison calculation unit 5 is not limited to the above-described one, and the ratio of the input values is used for the cut operation. The presence / absence may be determined. In addition to the memories 4A and 4B, a plurality of memories such as a memory 4C and a memory 4D are provided in the memory unit 4, and each memory stores a different data amount of each image unit, and the comparison operation unit 5
In the above, the time change of the data amount of each image unit may be detected from the output of each memory, and the presence or absence of the cut may be determined from the change rate of the data amount.

In the above, an example is shown in which the presence / absence of a cut is determined between adjacent image units, such as between the image unit I (10) and the image unit (11) in the configuration A of FIG. However, for example, image unit I (10) and image unit I (1
The data amount of 3) is counted by the counter 2, the results are stored in the memory A or the memory B, respectively, and the stored values are compared by the comparison calculation unit 5, so that the images not adjacent to each other are compared. The presence / absence of the cut may be determined.

Further, in this embodiment, as shown in FIG. 1, the memory 4A and the memory 4B are arranged in parallel, and the switch 4 is used while being switched. However, the memory 4A and the memory 4B are arranged in series. The new value may be stored in the memory 4A and the old value may be stored in the memory 4B. It should be noted that the control unit 3 is provided with a coding unit detection unit that detects whether the coded image data is coded in a frame or field image unit, and the comparison operation unit 5 is provided according to the detection result.
Value of the threshold value in the image is changed, or the data amount of two pieces of encoded image data obtained by encoding a field as an image unit is counted by the counter 2, and the count value and the frame are encoded as an image unit. The data amount of the image data is compared by the comparison operation unit 5, or two data amounts of the encoded image data obtained by encoding the field as an image unit are stored in the memory unit 4, and the comparison operation unit 5 causes the memory unit 4 to perform the operation. The values for the two data amounts of the encoded image data obtained by encoding the fields stored in
It may be configured to compare the total value and the data amount of encoded image data obtained by encoding the frame as an image unit,
In that case, the two encoded image data for comparing the data amounts have different encoded image units such that one is a frame encoded and the other is a field encoded. But you don't have to erroneously judge whether a cut is present or not.

As described above, according to the present embodiment, the coded image data group formed from the coded intra-picture data obtained by coding the intra-picture information by the variable length coding is used as the input picture data group, and the input picture data A control unit that generates a control signal for extracting the data amount of each image from the data group, and a control signal from the input image data group and the control unit, that is, a count start signal, a count stop signal, and a zero clear signal. A counter for inputting and counting a part or all of the data amount of the input image data group according to a control signal from the control unit, a storage unit for storing at least one value of the counter, and the storage unit Comparing a stored value with a preset threshold value, or comparing a value obtained by comparing and calculating a plurality of values stored in the storage unit with a preset threshold value By providing a comparison operation unit for determining whether or not there is a cut between the intra-image encoded data in the input image data group, the encoded image read from the recording medium 1 which was necessary in the conventional example shown in FIG. 14 is provided. Since the work of decoding the data becomes unnecessary, it becomes possible to detect the cut in a shorter time.

Further, since a large-scale image data memory for storing the decoded image data as in the conventional example is not required, the apparatus can be downsized and the cost can be reduced. Further, in the comparison calculation unit 5, the comparison calculation process for detecting a cut is performed using a huge amount of image data in the conventional example, whereas the comparison of a small number of values stored in the memory unit 4 in FIG. Since it suffices to perform the above, the time required for the processing in the comparison calculation unit 5 can be greatly reduced. In summary, it is possible to speed up the process, reduce the size of the device, and reduce the cost of the entire device, as compared with the conventional example.

(Embodiment 2) Embodiment 2 is almost the same as Embodiment 1, except that input image data, that is, data recorded on a recording medium, is encoded by using a quantization method. It is the coded image data (except when the quantization parameter is fixed) and the point where the quantization parameter detection unit 9 is provided instead of the counter.

FIG. 16 shows the structure of the coded image cut detecting device in the second embodiment. The operation of the coded image cut detection device configured as described above will be described below with reference to FIGS. 16, 9 and 10.

FIG. 9 is the same as FIG. 15 used in the description of the conventional coded image cut detecting device. In the present embodiment, the presence / absence of a cut is determined between image data that has been intra-image coded as in configuration A in FIG. Of the encoded image data shown in FIG. 9, the data is read in order from the beginning of the image unit I (10) of the configuration A from the recording medium 1 of FIG. 16, and the image unit detection unit 3A of FIG. It is input to the parameter detection unit 9.

First, the image unit detecting section 3A detects the data indicating the start position of the encoded image data, and sends the start signal to the quantization parameter detecting section 9 through the line f.
Upon receiving the start signal, the quantization parameter detection unit 9 starts detecting the quantization parameter of the image unit I (10) input through the line e. Then, the value detected by the quantization parameter detection unit 9 is stored in the memory 4A through the SW1 connected to the s side. Then, the image unit detection unit 3A detects the data indicating the end of the image unit I (10), a stop signal is sent to the quantization parameter detection unit 9 through the line f, and the detection by the quantization parameter detection unit 9 is stopped. To be done. In addition, a stop signal is sent to SW1 and S
W1 is switched to the t side.

Next, of the encoded image data shown in FIG. 9 from the recording medium 1, the data is sequentially read from the beginning of the image unit I (11) of the configuration A, and the image unit detection unit 3A of FIG. It is input to the conversion parameter detection unit 9. The image unit detection unit 3A first detects the data indicating the start position of the encoded image data, and transmits the start signal to the quantization parameter detection unit 9 through the line f. Upon receiving the start signal, the quantization parameter detection unit 9 starts detecting the quantization parameter of the image unit I (11) input through the line e. Then, the value detected by the quantization parameter detecting unit 9 is stored in the memory 4B through the SW1 connected to the t side. Then, the image unit detection unit 3A detects the data indicating the end of the image unit I (11), and the line f
A stop signal is sent to the quantization parameter detection unit 9 through, and the detection in the quantization parameter detection unit 9 is stopped.
A stop signal is also sent to SW1, and SW1 is switched to the s side.

Next, by the comparison operation of the values stored in the memories 4A and 4B in the comparison operation unit 5, whether or not there is a cut between the image unit I (10) and the image unit I (11) of the configuration A in FIG. Is determined. The process inside the comparison operation unit 5 will be described below. First, the difference between the value of the quantization parameter of the image unit I (10) of the configuration A of FIG. 9 and the value of the quantization parameter of the image unit I (11) stored in the memory 4A and the memory 4B of FIG. The absolute value α of is obtained. If there is no cut between the image unit I (10) and the image unit I (11), the image unit I (10)
(10) and the image unit I (11) are generally similar, and the value of the quantization parameter of the image unit I (10) and the value of the quantization parameter of the image unit I (11) are close values. , Α becomes 0 or a value close to 0. On the other hand, when there is a cut between the image unit I (10) and the image unit I (11), there is no correlation between the images and the image unit I (1
The value of the quantization parameter of 0) and the value of the quantization parameter of the image unit I (11) are generally quite different values, in which case the value of α is considerably larger than 0.

Therefore, the comparison calculation unit 5 sets a certain threshold β, and when the value of α is larger than the threshold β, the image unit I (1
0) and the image unit I (11), it is determined that there is a cut, and the cut detection signal is turned on and output. If the value of α is less than or equal to the threshold value β, the image unit I (10) and the image unit I (11) It is determined that there is no cut between the two) and the cut detection signal is turned off and output. When the comparison operation unit 5 completes the determination of the presence / absence of a cut between the image unit I (10) and the image unit I (11), the next encoded image data is read from the recording medium 1 thereafter, and the same as above. The quantization parameter is detected by the quantization parameter detection unit 9 according to the procedure, and the one of the memories 4A and 4B that stores the value of the quantization parameter of the encoded image data in the oldest reproduction image order is quantized. The new value of the conversion parameter is stored and updated, and each time the comparison calculation unit 5 performs the same process as described above, the presence or absence of a cut between images is determined.

The comparison calculation function of the values input from the memories 4A and 4B in the comparison calculation unit 5 is not limited to the above-described one, and the ratio of the input values is used for the cut operation. The presence / absence may be determined. Further, in addition to the memories 4A and 4B, a plurality of memories such as a memory 4C and a memory 4D are provided in the memory unit 4, and a quantization parameter for each image is stored in each memory. In 5, the time change of each image unit may be detected from the output of each memory, and the presence or absence of the cut may be determined from the change rate of the quantization parameter.

Further, in the above, an example is shown in which the presence / absence of a cut is determined between adjacent image units such as between the image unit I (10) and the image unit (11) in the configuration A of FIG. However, for example, image unit I (10) and image unit I (1
3) The quantization parameter is detected, the results are stored in the memory 4A or the memory 4B, respectively, and the stored values are compared by the comparison operation unit 5, thereby cutting between non-adjacent images. It may be possible to determine the presence or absence of. In addition, in the above, as shown in FIG.
The memory 4B and the memory 4B are arranged in parallel and used while being switched by the SW1, but the memory 4A and the memory 4B are arranged in series, the value of the memory 4A is stored in the memory 4B, and the memory 4A newly detects the value. The value of the quantization parameter may be stored.

Further, the control unit 3 is provided with a coding unit detecting unit for detecting whether the coded image data is coded in the image unit of the frame or the field, and the comparison operation unit according to the detection result. Change the threshold value within 5, or
Depending on the detection result, the comparison unit 5 controls whether the comparison unit is a field unit comparison or a frame unit comparison.

As described above, according to this embodiment, a group of coded image data composed of data obtained by coding an image by a quantization method is used as an input image data group, and each image is selected from the group of input image data. Control unit for generating a control signal for the detection of the quantization parameter, the input image data group and a control signal from the control unit, that is, a detection start signal, a detection stop signal, etc. are input, A quantization parameter detection unit that detects a quantization parameter of a part or all of the input image data group according to a control signal, a storage unit that stores at least one value of the quantization parameter, and a storage unit that stores the storage unit. Comparing the calculated value with a preset threshold value, or comparing a value obtained by comparing and calculating a plurality of values stored in the storage unit with a preset threshold value. Therefore, by providing a comparison operation unit for determining whether or not there is a cut between the intra-image encoded data in the input image data group, the encoded image data read from the recording medium 1 which was necessary in the conventional example shown in FIG. Since the decoding work is not necessary, it becomes possible to detect the cut in a shorter time.

Further, unlike the conventional example, a large-scale image data memory for storing the decoded image data is not required, so that the apparatus can be downsized and the cost can be reduced. Further, in the comparison calculation unit 5, the comparison calculation process for detecting a cut is performed using a huge amount of image data in the conventional example, whereas the comparison of a small number of values stored in the memory unit 4 in FIG. Since it suffices to perform the above, the time required for the processing in the comparison calculation unit 5 can be greatly reduced.

In summary, it is possible to speed up the process, reduce the size of the device, and reduce the cost of the device as a whole, as compared with the conventional example.

(Third Embodiment) The third embodiment is substantially the same as the second embodiment, except that the input image data, that is, the data recorded on the recording medium is either the variable length coding method or the quantization method. (However, except when the quantization parameter is fixed), or the point that the image data is encoded using the variable length coding method and the quantization method, and the counter used in the first embodiment. 2 is the point.

That is, the method is a combination of the first and second embodiments. FIG. 17 shows the configuration of the coded image cut detection device according to the third embodiment. The operation of the coded image cut detection device configured as described above will be described below with reference to FIGS. 17, 9 and 10.
Note that FIG. 9 is the same as FIG. 15 used in the description of the conventional coded image cut detection device. In the present embodiment, the presence / absence of a cut is determined between image data that has been intra-image coded as in configuration A in FIG. Of the encoded image data shown in FIG. 9, the data is read in order from the beginning of the image unit I (10) of the configuration A from the recording medium 1 of FIG. 17, and the image unit detection unit 3A and the counter 2 of FIG. Is input to the quantization parameter detection unit 9.

First, the image unit detecting section 3A detects the data indicating the start position of the encoded image data, and outputs the start signal through the line f to the counter 2 and the quantization parameter detecting section 9
Send to. The counter 2 performs the same operation as that of the first embodiment, the quantization parameter detection unit 9 also performs the same operation as that of the second embodiment, and the data amount counted by the counter 2 and the quantization parameter detection unit 9 are Multiply the detected quantization parameter. Thereafter, the switching method of SW1, the storage method in the memory unit 4, and the comparison method in the comparison operation unit 5 are the same as those in the first or second embodiment. Thereafter, the same operation as described above is repeated until the input data is completed, and the cut judgment is performed.

The comparison calculation function of the values input from the memories 4A and 4B in the comparison calculation unit 5 is not limited to the above-described one, and the presence or absence of the cut is determined by using the ratio of the input values. May be determined. Further, in addition to the memories 4A and 4B, a plurality of memories such as a memory 4C and a memory 4D are provided in the memory unit 4, and a quantization parameter for each image is stored in each memory. In 5, the time change of each image unit may be detected from the output of each memory, and the presence or absence of the cut may be determined from the change rate of the multiplication value of the data amount and the quantization parameter.

Further, in the above, an example in which the presence / absence of a cut is determined between adjacent image units, such as between the image unit I (10) and the image unit I (11) in the configuration A of FIG. As shown, for example, the image unit I (10) and the image unit I (1
Even if the result of 3) is stored in the memory A or the memory B and the stored values are compared by the comparison operation unit 5, whether or not there is a cut between non-adjacent images is determined. good. Further, in the above, the memory 4A and the memory 4B are arranged in parallel as shown in FIG.
Memory 4A was used while switching while using W1
And the memory 4B may be arranged in series, the value of the memory 4A may be stored in the memory 4B, and new data may be stored in the memory 4A.

Further, the control unit 3 is provided with an encoding unit detecting unit for detecting whether the encoded image data is encoded in the image unit of the frame or the field, and the comparison operation unit is provided according to the detection result. Change the threshold value within 5, or
Depending on the detection result, the comparison unit 5 controls whether the comparison unit is a field unit comparison or a frame unit comparison. In addition, when a cut is detected between forward predictive coded images, the multiplication value of the data amount of the image immediately after the cut and the quantization parameter becomes a large value. Therefore, one of the memories 4A and 4B is used. May be used to determine the presence or absence of a cut. Further, in the above, an example in which the data amount is multiplied by each value of the quantization parameter has been shown, but the calculation method is not limited to multiplication, and for example, exponentiation may be used.

As described above, according to the present embodiment, a group of coded image data composed of image data coded by variable length coding or quantization, or variable length coding and quantization is used as an input image. As a data group, a control unit that generates a control signal for counting data of each image and extracting a quantization parameter from the input image data group, a control signal from the input image data group and the control unit, That is, a detection start signal and a detection stop signal are input, and a counter for counting the data amount of a part or all of the input image data group and a quantization parameter detection for detecting the quantization parameter according to the control signal from the control unit. Section, a storage section that stores at least one value obtained by calculating a data amount and a quantization parameter, and a value stored in the storage section and a preset threshold value are compared. Alternatively, the presence or absence of a cut between intra-coded data in the input image data group is determined by comparing a value obtained by comparing and calculating a plurality of values stored in the storage unit with a preset threshold value. By providing a comparison operation unit that enables the input data of either encoded data in which the amount of data encoded using the variable-length encoding method fluctuates or encoded data encoded using the quantization method, Cut detection is possible.

Further, in the case of coded data using quantization and variable length coding like MPEG, in the case of coded data in which the quantization parameter is constant or almost constant and the data size changes greatly by cutting, Even if the quantization parameter changes greatly due to the cut and the encoded data is almost constant encoded data, the cut can be detected without depending on the content of such data. Further, in the case of coded data in which the change in the data amount due to the cut of the input data is small and the change in the quantization parameter is small, the method of determining the cut by the change of the data amount can be the method of determining the cut by the change of the quantization parameter. However, it is difficult to detect the cut, but by calculating the data amount and the quantization parameter, the change in the data due to the cut can be greatly expressed, and the cut is judged by the change in the data amount only and the change in the quantization parameter only. More accurate cut detection than the method is possible. Further, since the work of decoding the coded image data read from the recording medium 1 which is required in the conventional example shown in FIG. 14 is unnecessary, it is possible to detect the cut in a shorter time. Further, unlike the conventional example, a large-scale image data memory for storing the decoded image data is unnecessary, so that the apparatus can be downsized and the cost can be reduced.

Further, in the comparison operation unit 5 in the conventional example, a large amount of image data is used to perform the comparison operation process for detecting a cut, whereas the comparison operation unit 5 stores a small number of data stored in the memory unit 4 in FIG. Since it suffices to compare the values, the time required for the processing in the comparison calculation unit 5 can be greatly reduced.

In summary, the processing speed of the entire apparatus can be increased, and the apparatus can be downsized and the cost can be reduced as compared with the conventional example.

(Embodiment 4) A fourth embodiment of the present invention will be described below with reference to the drawings.

FIG. 2 shows the configuration of a coded image cut detection device according to the fourth embodiment of the present invention. In FIG. 2, 1 is a recording medium, 2 is a counter, 3 is a control unit, 3A is an image unit detection unit, 4 is a memory unit, and 4A and 4
B is a memory and 5 is a comparison operation unit. The above is the same as the configuration of the first embodiment of the present invention shown in FIG.

The difference from FIG. 1 is that the control unit 3 is provided with a region detection unit 3B for detecting each primary divided region in each image, and the inside of the memories 4A and 4B. Respectively (a1, a2, ... an) and (b1, b
2, ... bn), SW2, SW3, S
The point is that the input and output can be switched using W4 and SW5. The operation of the coded image cut detection device configured as described above will be described below with reference to FIGS. 2, 9, and 10.

FIG. 10 shows the structure when an image unit is divided into each primary divided area. In this embodiment, the presence / absence of a cut is determined between the encoded image data having the configuration A in FIG. In the encoded image data shown in FIG. 9, the data is read in order from the beginning of the image unit I (10) of the configuration A in FIG. 2 from the recording medium 1.
It is input to the image unit detection unit 3A, the area detection unit 3B, and the counter 2 of FIG. The image unit detection unit 3A first detects that it is the head data in image units. When the leading data of each image is detected, the image detection unit 3A sends an area detection start signal to the area detection unit 3B. Area detector 3B
Detects the leading data of the coded image data in the area 1 shown in FIG. 10, and transmits the zero clear signal and the count start signal to the counter 2 through the line f in FIG.

When the counter 2 receives the zero clear signal and the count start signal, the counter 2 clears the count value to zero and starts counting the data amount of the coded image data in the area 1 of FIG. 10 input through the line e. The coded image data of the area 1 in FIG. 10 is read from the recording medium 1 one after another, and the area detection unit 3B eventually detects the end of the coded image data of the area 1 in FIG. A count stop signal is sent to 2, and the counter 2 stops counting. And at first SW1 is a and SW
Since 2 is connected to a1, the count value of the counter 2 is stored in a1 of the memory 4A. Then, SW2 is switched to a2.

Next, the head data of the coded image data of the area 2 shown in FIG. 10 is read from the recording medium 1, and the head of the coded image data of the area 2 shown in FIG. 10 is read by the area detector 3B of FIG. Data is detected. Then, the zero clear signal and the count start signal are transmitted to the counter 2 through the line f in FIG.

When the counter 2 receives the zero clear signal and the count start signal, the counter 2 clears the count value to zero and starts counting the data amount of the coded image data in the area 2 of FIG. 10 input through the line e. The coded image data of the area 2 in FIG. 10 is read from the recording medium 1 one after another, and the area detection unit 3B eventually detects the end of the coded image data in the area 2 of FIG. A count stop signal is sent to 2, and the counter 2 stops counting. And SW1 is a and SW2 is a
Since it is connected to 2, the count value of the counter 2 is stored in a2 of the memory 4A. Then, SW2 is switched to a3.

Thereafter, in the same procedure, the data amount values of the coded image data corresponding to the areas 3 to n in FIG. 10 are stored in a3 to an in the memory 4A in FIG.
W2 is switched to a1.

The image unit detecting section 3A has the structure A shown in FIG.
The end of reading of the image unit I (10) is detected, and an area detection stop signal is transmitted to the area detection unit 3B. Then SW
1 is switched to the b side.

Next, of the encoded image data shown in FIG. 9 from the recording medium 1, the data of the image unit I (11) of the configuration A is read out, and the SW3 is switched in the same manner as the above SW2. , FIG. 1 in image unit I (11)
The values of the data amounts of the coded image data in the areas 1 to n shown in 0 are stored in b1 to bn in the memory 4B of FIG. 2, respectively.

Next, by the comparison operation of the values stored in the memories 4A and 4B in the comparison operation unit 5, whether or not there is a cut between the image unit I (10) and the image unit I (11) of the configuration A of FIG. Is determined. The process inside the comparison operation unit 5 will be described below.

First, SW4 and SW5 are the memory 4 respectively.
Connected to a1 of A and b1 of the memory 4B and stored in a1 via line c and line d, respectively, and b1
The value stored in is input to the comparison calculation unit 5. a1
The value stored in the image unit I (1
0) is the data amount of the coded image data corresponding to the area 1 in FIG. 10 and is referred to as data C. The value stored in b1 is the data amount of the coded image data corresponding to the area 1 in FIG. 10 in the image unit I (11) of the configuration A in FIG. 9, which is called data D. That is, in order to compare the data C and the data D, the comparison operation unit 5 first obtains the absolute value α1 of the difference between the value of the data C and the data D.

Next, SW4 and SW5 are respectively connected to the memory 4A.
A2 and b2 of the memory 4B, and inputs them to the comparison operation unit 5. In the comparison operation unit 5, the value read from a2 is set as new data C, the value read from b2 is set as new data D, and the values of data C and data D are compared in order to compare data C and data D. Absolute value of value difference α2
Ask for.

After that, SW4 and SW5 follow the same procedure as above.
While switching, the values from α3 to αn are obtained.

If the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11) are exactly the same,
The image unit I (10) and the image unit I (11) have the same value, and naturally the data amount is exactly the same. That is, data C
And the data D have the same value. Therefore, at this time, the values of α1 to αn are all 0. Even if the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11) are not exactly the same, the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11) are If there is no cut in,
Generally, the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11) are similar to each other, and even though they are encoded by the variable length encoding method, the image unit I (10) And the data amount of the image unit I (11) are close to each other, that is, the data C and the data D are close to each other. Therefore, most of the values of α1 to αn are close to zero.

On the other hand, when there is a cut between the reproduced image of the image unit I (10) and the reproduced image of the image unit I (11), there is no correlation between the reproduced images, and the variable length coding method is used. Since the data is encoded, the data amount of the image unit I (10) and the data amount of the image unit I (11) are generally quite different values, that is, the values of the data C and the data D are very different values. become. Therefore, in this case, among the values of α1 to αn, those having values considerably larger than 0 appear.

Therefore, the comparison operation unit 5 sets a certain threshold value β, and when some of the values of α1 to αn are larger than the threshold value β, the image unit I (10) and the image unit I (1
It is determined that there is a cut during 1), the cut detection signal is turned on and output, and if some of the values of α1 to αn are less than or equal to the threshold β, the image unit I (10) and the image unit I (1
It is determined that there is no cut during 1) and the cut detection signal is turned off and output.

Image unit I (10) in the comparison operation unit 5
When the determination of the presence / absence of the cut between the image unit I (11) and the image unit I (11) is completed, the data of the next image unit is read from the recording medium 1 and the data amount is counted by the counter 2 in the same procedure as above. , Of the memories 4A and 4B,
The value of the new counter 2 is stored and updated in the memory that stores the data amount of the old data in the order of the reproduced images, and the processing similar to the above is performed in the comparison calculation unit 5 each time, and the cut between the reproduced images is performed. The presence or absence of is determined.

The function of comparing the values input from the memory 4A and the memory 4B in the comparison operation unit 5 is not limited to the above, and the presence / absence of a cut is determined by using the ratio of the values input. May be determined. Further, the presence or absence of the cut may be determined by comparing a part or all of the values α1 to αn or the average value thereof with a certain threshold value. Further, although the structure in which the image unit is divided into each primary divided area is shown in FIG. 10, the dividing method is not limited to this, and it may be divided in the horizontal direction or in a matrix. In addition to the memories 4A and 4B, a plurality of memories such as a memory 4C and a memory 4D are provided in the memory unit 4,
The amount of data for each image unit is stored in each memory,
The comparison operation unit 5 may detect the time change of the data amount of each image unit from the output of each memory and determine the presence or absence of the cut from the rate of change of the data amount.

Further, in this embodiment, as shown in FIG. 2, the memory 4A and the memory 4B are arranged in parallel, and the switch 4 is used while being switched. However, the memory 4A and the memory 4B are arranged in series. The new value may be stored in the memory 4A and the old value may be stored in the memory 4B. Note that, in the above, an example in which the presence / absence of a cut is determined between adjacent images such as between the image unit I (10) and the image unit (11) in the configuration A of FIG. 9 has been described. Unit I
The data amount of (10) and the image unit I (13) is counted by the counter 2
And the results are stored in the memory 4A or the memory 4B, respectively, and the stored values are compared by the comparison operation unit 5 to determine the presence / absence of a cut between non-adjacent images. May be.

Further, in the above, in the comparison operation unit 5 of FIG. 2, the minimum unit data amount of each area such as a1 and b1 in the memory unit 4 is compared, but the comparison unit is as described above. There is no need to limit this. For example, the sum of the values of a1 and a2 and the sum of the values of b1 and b2 may be compared, and comparison may be performed with the data amount corresponding to a plurality of areas.

The control unit 3 is provided with a coding unit detection unit for detecting whether the coded image data is coded in the image unit of the frame or the field, and the comparison operation unit according to the detection result. 5, the value of the threshold value is changed, or the data amount of two pieces of encoded image data obtained by encoding a field as an image unit is counted by the counter 2, and the count value and the frame are encoded as an image unit. The data amount of the encoded image data is compared by the comparison operation unit 5, or two data amounts of the encoded image data obtained by encoding the field as an image unit are stored in the memory unit 4, and the comparison operation unit 5 stores the memory unit. Values of two data amounts of encoded image data obtained by encoding the field stored in 4 as an image unit are summed, and the sum and the frame are encoded as an image unit. May be configured to compare the data amount of Goka image data, in which case, 2 for comparing the data amount
One coded image data is one coded frame, the other is field coded. Complete.

Further, even when the data amounts of the images as a whole between the images before and after a certain cut are accidentally close to each other, in the present embodiment, the data amounts of the corresponding regions in each image unit are different from each other. Since each image is compared, it is possible to grasp the difference in the amount of data between regions due to changes in the contours and the existing positions of objects that exist in the reproduced image, and thus the amount of data in the entire image. Compared with the first embodiment in which the cut detection is performed, the cut detection can be performed with higher accuracy.

In the above description, the control unit 3 shown in FIG. 2 is provided with both the image unit detecting unit 3A and the region detecting unit 3B, but if the number n of regions in each image is known in advance, The image unit detection unit 3A may not be provided,
In that case, a function of managing the number of the appearing areas may be provided in the area detecting unit 3B instead, and the end of the image unit may be determined from the number of the appearing areas, or n of the memory 4A or the memory 4B may be determined. The end of the image unit may be determined by updating the value of the th memory, that is, an or bn.

In the above description, the coded data coded by the variable length coding is used as the input image data group, the counter 2 counts the data amount, and the value is stored in the memory 4. The encoded data encoded by the above may be used as an input data group, and the counter 2 may be replaced with the quantization parameter detection unit 9 to detect the quantization parameter and store it in the memory 4. In addition, variable length coding, quantization, or coded data coded by variable length coding and quantization is used as input image data, and both the counter 2 and the quantization parameter detection unit 9 are provided. A value obtained by multiplying the value and the value of the quantization parameter 9 by multiplication or exponentiation may be stored in the memory 4.

As described above, according to this embodiment, the intra-picture information is encoded by variable-length coding or quantization, or the intra-picture coded data that is coded by variable-length coding and quantization. An image data group is used as an input image data group, a control unit that extracts a data amount of each image from the input image data group or generates a control signal for detecting a quantization parameter, and the input image data group. A counter that receives a control signal from the control unit and counts a part or all of the data amount of the input image data group according to the control signal from the control unit, or a quantization parameter detection that detects a quantization parameter. Section, a storage section that stores at least one value of the counter or a quantization parameter, and a value stored in the storage section and a preset threshold value are compared, or Comparing operation for determining the presence / absence of a cut between intra-coded data in the input image data group by comparing a value obtained by performing a comparison operation of a plurality of values stored in the unit with a preset threshold value The recording medium, which is necessary in the conventional example shown in FIG. 14, is configured by including a section and the control section including an area detection section that detects each divided area in each image unit of the input image data group. Since the work of decoding the coded image data read from No. 1 is unnecessary, it is possible to detect the cut in a shorter time.

Further, since a large-scale image data memory for storing the decoded image data as in the conventional example is not required, the device can be downsized and the cost can be reduced. Further, in the comparison calculation unit 5, in the conventional example, a comparison calculation process for detecting a cut is performed using a large amount of image data, whereas a comparison of a small number of values stored in the memory unit 4 in FIG. Since it suffices to perform the above, the time required for the processing in the comparison calculation unit 5 can be greatly reduced. In summary, it is possible to speed up the process, reduce the size of the device, and reduce the cost of the entire device, as compared with the conventional example.

(Embodiment 5) A fifth embodiment of the present invention will be described below with reference to the drawings.

FIG. 3 shows the configuration of a coded image cut detection device according to the fifth embodiment of the present invention. In FIG. 3, 1 is a recording medium, 2 is a counter, 3 is a control unit, 3A is an image unit detection unit, 3B is an area detection unit, 4 is a memory unit, 4A and 4B are memories, and 5 is a comparison operation unit. The above is the same as the configuration of the fourth embodiment of the present invention shown in FIG. The difference from FIG. 2 is that an image coding type detection unit 3C that detects the coding type of each image is provided in the control unit 3.

The operation of the coded image cut detecting device configured as described above will be described below with reference to FIGS. 3, 9 and 10.

In this embodiment, the configuration A, the configuration B, and
Determining the presence / absence of a cut between intra-coded image data of configuration C, between forward predictive-coded image data and its reference image data, and between forward predictive-coded image data However, the operation will be described for each method by taking the case of using the data of the configuration B as an example.

The operation for determining the presence / absence of a cut between the intra-image coded image data of the configuration B of FIG. 9 by the method 1 will be described below. In the coded image data shown in FIG. 9, the data is sequentially read from the beginning of the image unit I (10) of the configuration B in FIG. 3 from the recording medium 1, and the image unit detection unit 3A and the area detection in FIG. It is input to the unit 3B, the image coding type detection unit 3C, and the counter 2. The image unit detection unit 3A first detects that it is the head data in image units. When the leading data of the image unit is detected, the image coding type detection start signal is transmitted from the image unit detection unit 3A to the image coding type detection unit 3C.

Then, the image coding type detection unit 3C determines which coding type the image unit I (10) is, of intra-picture coding, forward predictive coding and bidirectional predictive coded image, It can be seen that the image unit I (10) is due to intra-picture coding. Then, the image coding type detection unit 3C
The area detection start signal is transmitted from the area detection unit 3B.
Then, the area detector 3B detects the leading data of the encoded image data of the area 1 shown in FIG. When the leading data of the coded image data in the area 1 is detected, the zero clear signal and the count start signal are transmitted to the counter 2 through the line f in FIG. When the counter 2 receives the zero clear signal and the count start signal, the counter 2 clears the count value to zero and starts counting the data amount of the coded image data input through the line e. Recording media one after another
The encoded image data of the area 1 in FIG. 10 is read from
Eventually, the area detection unit 3B detects the end of the coded image data of the area 1 in FIG. 10, a count stop signal is sent to the counter 2 through the line f in FIG. Since SW1 is initially connected to a and SW2 is connected to a1, the count value of the counter 2 is stored in a1 of the memory 4A. And SW2 is a2
Is switched to.

Next, the head data of the coded image data of the area 2 shown in FIG. 10 is read from the recording medium 1, and the head of the coded image data of the area 2 shown in FIG. 10 is read by the area detector 3B of FIG. Data is detected. Then, the zero clear signal and the count start signal are transmitted to the counter 2 through the line f in FIG. When the counter 2 receives the zero clear signal and the count start signal, the counter 2 clears the count value to zero and starts counting the data amount of the coded image data input through the line e. The encoded image data in the area 2 in FIG. 10 is read from the recording medium 1 one after another, and the area detection unit 3B eventually detects the end of the encoded data in the area 2 in FIG. 10, and the counter 2 is detected through the line f in FIG. A count stop signal is sent to, and the count in the counter 2 is stopped.

Since SW1 is connected to a and SW2 is connected to a2, the count value of the counter 2 is stored in the memory 4
It is stored in a2 of A. Then, SW2 is switched to a3.

By the same procedure as above, the data amount values of the coded image data corresponding to the areas 3 to n in FIG. 10 are stored in a3 to an in the memory 4A in FIG.
2 is switched to a1. Then, the image unit detection unit 3A detects the end of reading of the image unit I (10) of the configuration B of FIG. 9, and SW1 is switched to the b side. Next, of the encoded image data shown in FIG. 9 from the recording medium 1, the data of the image unit of the configuration B that has already been read is read. Then, the image coding type detection unit 3A skips the data until the next intra-coded data, that is, the image unit I (15) of the configuration B of FIG. 9 is detected.

When the image unit I (15) is detected, SW
When 3 is the image unit I (10) described above, the same switching as SW2 is performed, so that the image unit I (1
5), the value of the data amount of the encoded image data in the areas 1 to n shown in FIG. 10 is b in the memory 4B in FIG.
Stored from 1 to bn. Next, by the comparison calculation of the values stored in the memories 4A and 4B in the comparison calculation unit 5, the image unit I (10) and the image unit I of the configuration B in FIG.
The presence / absence of a cut is determined during (15).

The process inside the comparison operation unit 5 will be described below. First, SW4 and SW5 are the memory 4 respectively.
Connected to a1 of A and b1 of the memory 4B and stored in a1 via line c and line d, respectively, and b1
The value stored in is input to the comparison calculation unit 5. a1
The value stored in the image unit I (1
0) is the data amount of the coded image data corresponding to the area 1 in FIG. 10 and is referred to as data C. The value stored in b1 is the data amount of the encoded image data corresponding to the area 1 in FIG. 10 in the image unit I (15) of the configuration B in FIG. 9 and is called data D. That is, in order to compare the data C and the data D, the comparison operation unit 5 first obtains the absolute value α1 of the difference between the value of the data C and the data D.

Next, SW4 and SW5 are respectively set in the memory 4A.
A2 and b2 of the memory 4B, and inputs them to the comparison operation unit 5. In the comparison operation unit 5, the value read from a2 is the new data C, the value read from b2 is the new data D, and the difference between the value of the data C and the data D is set in order to compare the data C with the data D. The absolute value α2 of is calculated.

After that, in the same procedure as above, SW4 and SW5
While switching, the values from α3 to αn are obtained.

If the reproduced image of the image unit I (10) and the reproduced image of the image unit I (15) are exactly the same,
The image unit I (10) and the image unit I (15) have the same value, and naturally the data amount is exactly the same. That is, data C
And the data D have the same value. Therefore, at this time, the values of α1 to αn are all 0. Even if the reproduced image of the image unit I (10) and the reproduced image of the image unit I (15) are not exactly the same, between the reproduced image of the image unit I (10) and the reproduced image of the image unit I (15). If there is no cut in,
Generally, the reproduced image of the image unit I (10) and the reproduced image of the image unit I (15) are similar to each other, and even though they are encoded by the variable length encoding method, the image unit I (10) And the data amount of the image unit I (15) are close to each other, that is, the data C and the data D are close to each other. Therefore, most of the values of α1 to αn are close to zero.

On the other hand, when there is a cut between the reproduced image of the image unit I (10) and the reproduced image of the image unit I (15), there is no correlation between the reproduced images and the variable length coding method is used. Since it is encoded, the data amount of the image unit I (10) and the data amount of the image unit I (15) are generally quite different values, that is, the values of the data C and the data D are very different values. become. Therefore, among the values of α1 to αn, those having values considerably larger than 0 appear.

Therefore, the comparison operation unit 5 sets a certain threshold value β and compares the values α1 to αn themselves with the threshold value β.

If some of the values α1 to αn are larger than the threshold value β, it is determined that there is a cut between the image unit I (10) and the image unit I (15), and the cut detection signal is turned on. If some of the values of α1 to αn are less than or equal to the threshold value β, the image unit I (10) and the image unit I
It is determined that there is no cut during (15) and the cut detection signal is turned off and output.

After that, the next data is read from the recording medium 1, the data amount is counted by the counter 2 in the same procedure as described above, and the data amount of the old data in the order of reproduced images in the memory 4A and the memory 4B is determined. The new value of the counter 2 is stored and updated in the stored memory, and each time the comparison calculation unit 5 performs the same process as described above, the presence or absence of a cut between reproduced images is determined.

The operation of determining the presence / absence of a cut between the forward predictive-coded image data of the configuration B of FIG. 9 and the reference image thereof by the method 2 will be described below.

In the encoded image data shown in FIG. 9 from the recording medium 1 in FIG. 3, the image unit I (1
Data is sequentially read from the beginning of (0) and is input to the image unit detection unit 3A, the area detection unit 3B, the image coding type detection unit 3C, and the counter 2 of FIG. Image unit detection unit 3
In A, it is detected that the image unit is input, and then the image coding type detection unit 3C detects that the image unit I (10) is intra-image coded. Next, the image unit P (11) is read. In the image unit detection unit 3A,
It is detected that it is the head data of an image unit, and an image coding type detection start signal is transmitted from the image unit detecting unit 3A to the image coding type detecting unit 3C. Then, the image coding type detection unit 3C detects that the image unit P (11) is forward predictively coded. As a result, a region detection start signal is transmitted from the image coding type detection unit 3C to the region detection unit 3B. Then, in the area detection unit 3B,
The leading data of the coded image data of the area 1 indicated by 0 is detected.

When the leading data of the coded image data in the area 1 is detected, the zero clear signal and the count start signal are transmitted to the counter 2 through the line f in FIG.

Upon receiving the zero clear signal and the count start signal, the counter 2 clears the count value to zero and starts counting the amount of coded image data input through the line e. The coded image data is read from the recording medium 1 one after another, and the area detection unit 3B eventually detects the end of the coded image data of the area 1 in FIG. 10, and the line f in FIG.
A count stop signal is sent to the counter 2 through the, and the count in the counter 2 is stopped. And first SW1
Is connected to a and SW2 is connected to a1, so the count value of the counter 2 is stored in a1 of the memory 4A. Then, SW2 is switched to a2.

Next, the head data of the coded image data of the area 2 shown in FIG. 10 is read from the recording medium 1, and the head of the coded image data of the area 2 shown in FIG. 10 is read by the area detector 3B of FIG. Data is detected. Then, the zero clear signal and the count start signal are transmitted to the counter 2 through the line f in FIG.

Upon receiving the zero clear signal and the count start signal, the counter 2 clears the count value to zero and starts counting the amount of coded image data input through the line e. The coded image data is read from the recording medium 1 one after another, and the area detection unit 3B eventually detects the end of the coded image data of the area 2 in FIG. 10, and the line f in FIG.
A count stop signal is sent to the counter 2 through the, and the count in the counter 2 is stopped. And SW1 is a
Since SW2 is connected to a2, the count value of the counter 2 is stored in a2 of the memory 4A. Then, SW2 is switched to a3.

Thereafter, in the same procedure, the data amount values of the encoded image data corresponding to the areas 3 to n in FIG. 10 are stored in a3 to an in the memory 4A of FIG. 3, respectively, and S
W2 is switched to a1.

The image unit detecting section 3A has a structure B shown in FIG.
End of reading of the image unit P (11) of
Is switched to the b side.

Next, a comparison operation of the memory 4A and a certain threshold value βp in the comparison operation section 5 causes a reproduced image of the image unit P (11) of the configuration B of FIG. 9 and an image which is a reference image of the image unit P (11). Whether or not there is a cut between the reproduced images of the unit I (10) is determined. The process inside the comparison operation unit 5 will be described below.

First, SW4 is connected to a1 of the memory 4A, and the value stored in a1 is input to the comparison operation section 5 through the line c. The value stored in a1 is the area 1 in FIG. 11 in the image unit P (11) of the configuration B in FIG.
Is the amount of encoded image data corresponding to. The comparison operation unit 5 compares the value stored in a1 with the threshold value βp.

The image unit P (11) is the image unit I (10).
Difference between the reproduced image of P and the reproduced image of the image unit P (11)
If the image unit I (10) is
If the reproduced image of P (11) is exactly the same as the reproduced image of P (11), the value of the difference Dp is 0, and the image unit P (1
1) is the coded 0. Therefore, the data amount of the image unit P (11) becomes a certain value C0 when 0 is encoded.

Even if the reproduced image of the image unit I (10) and the reproduced image of the image unit P (11) are not exactly the same, the reproduced image of the image unit I (10) and the image unit P (1
If there is no cut between the reproduced images in 1), the image unit I
Generally, the reproduced image of (10) and the reproduced image of the image unit P (11) are similar, and the value of the difference Dp is close to zero.
Therefore, the data amount of the image unit P (11) becomes a value close to a certain constant value C0 when 0 is encoded.

On the other hand, when there is a cut between the reproduced image of the image unit I (10) and the reproduced image of the image unit P (11), there is no correlation between the reproduced images and the value of the difference Dp is 0. It is generally much higher. Therefore, the data amount of the image unit P (11) becomes a value larger than a certain value C0 when 0 is encoded.

Therefore, the comparison operation unit 5 of FIG.
A1 of A and the threshold value βp are compared. Similarly, the values from a2 to an in the memory 4A are also compared with the threshold value βp. If some of the values a1 to an in the memory 4A are larger than the threshold value βp, it is determined that there is a cut between the reproduced image of the image unit I (10) and the image unit P (11), and the cut detection signal is detected. Is turned on and output, and a in memory 4A
If some of the values 1 to an are not more than the threshold value βp, it is determined that there is no cut between the image unit I (10) and the image unit P (11), and the cut detection signal is turned off and output.

After that, if the next image unit is read from the recording medium 1 and it is forward predictive coded,
The data amount is counted by the counter 2 in the same procedure as described above, the value of the counter 2 is stored in the memory 4B this time, and the comparison operation unit 5 stores the values of b1 to bn of the memory 4B and the threshold value β.
By comparing with p, it is determined whether or not there is a cut between the reproduced image of the forward prediction coded image in which the data amount is stored in the memory 4B and the reference image.

As an example, in FIG. 12, the forward predictive coded data is selected from a series of coded image data groups having 6 cuts in the reproduced image and coded in the same configuration as the configuration C of FIG. The graph which extracted only and displayed the value of the data amount of forward prediction encoding data is shown. In FIG. 12, the X-axis shows the number of frames of the reproduced image, and the Y-axis shows the value of the amount of forward predictive encoded data. The six cuts are indicated by □ in FIG. As is clear from FIG. 12, a large value appears on the Y-axis where there is a cut, so a certain threshold βp is set, and the presence or absence of most of the cut is determined by comparing the Y-axis value with the threshold βp. it can.

In the example of FIG. 12, the X axis is 300.
Since a large value appears on the Y-axis in the data near 450 to 450, it may be erroneously detected that there is a cut. Hold the value and refer to the value to set the threshold β
The value of p may be changed, which makes it possible to prevent erroneous detection.

The operation in the case of judging the presence / absence of a cut between the forward predictive coded image data of the configuration B of FIG. 9 by the method 3 will be described below.

In the encoded image data shown in FIG. 9 from the recording medium 1 in FIG. 3, the image unit I (1
Data is sequentially read from the beginning of (0) and is input to the image unit detection unit 3A, the area detection unit 3B, the image coding type detection unit 3C, and the counter 2 of FIG. Image unit detection unit 3
In A, it is detected that the image unit is input, and then the image coding type detection unit 3C detects that the image unit I (10) is intra-image coded. Next, the image unit P (11) is read. In the image unit detection unit 3A,
It is detected that it is the first data of the image unit P (11), and the image unit detection unit 3A to the image coding type detection unit 3
An image coding type detection start signal is transmitted to C. Then, the image coding type detection unit 3C causes the image unit P (1
It is detected that 1) is forward predictive coded.
As a result, a region detection start signal is transmitted from the image coding type detection unit 3C to the region detection unit 3B. Then, the area detector 3B detects the leading data of the encoded image data of the area 1 shown in FIG.

When the leading data of the coded image data in the area 1 is detected, the zero clear signal and the count start signal are transmitted to the counter 2 through the line f in FIG.

When the counter 2 receives the zero clear signal and the count start signal, it clears the count value to zero and starts counting the data amount of the coded image data input through the line e. The coded image data of the area 1 of FIG. 10 is read from the recording medium 1 one after another, and the area detection unit 3B eventually detects the end of the coded image data of the area 1 of FIG. A count stop signal is sent to 2, and the counter 2 stops counting. Since SW1 is initially connected to a and SW2 is connected to a1, the count value of the counter 2 is stored in the memory 4A.
Stored in a1. Then, SW2 is switched to a2.

Next, the head data of the coded image data of the area 2 shown in FIG. 10 is read from the recording medium 1, and the head of the coded image data of the area 2 shown in FIG. 10 is read by the area detection unit 3B of FIG. Data is detected. Then, the zero clear signal and the count start signal are transmitted to the counter 2 through the line f in FIG.

Upon receiving the zero clear signal and the count start signal, the counter 2 clears the count value to zero and starts counting the data amount of the coded image data input through the line e. The encoded image data in the area 2 in FIG. 10 is read from the recording medium 1 one after another, and the area detection unit 3B eventually detects the end of the encoded image data in the area 2 in FIG. A count stop signal is sent to 2, and the counter 2 stops counting. Since SW1 is connected to a and SW2 is connected to a2, the count value of the counter 2 is a2 of the memory 4A.
Is stored. Then, SW2 is switched to a3.

Thereafter, in the same procedure, the data amount values of the coded image data corresponding to the areas 3 to n in FIG. 10 are stored in a3 to an in the memory 4A in FIG.
W2 is switched to a1.

The image unit detecting section 3A has a structure B shown in FIG.
End of reading of the image unit P (11) of
Is switched to the b side.

Next, of the encoded image data shown in FIG. 9 from the recording medium 1, the already-encoded encoded image data of configuration B is read out. Then, the image coding type detection unit 3A detects the data that has been forward predictively coded next, that is, the image unit P (12) of the configuration B of FIG. When the image unit P (12) is detected, SW3 is switched in the same manner as SW2 as in the image unit P (11) described above, so that the image unit P (12) shown in FIG. The data amount values of the encoded image data in the areas 1 to n are stored in b1 to bn in the memory 4B of FIG. 3, respectively.

Next, by the comparison operation of the values stored in the memories 4A and 4B in the comparison operation section 5, whether or not there is a cut between the image unit P (11) and the image unit P (12) of the configuration B of FIG. Is determined. The process inside the comparison operation unit 5 will be described below.

First, SW4 and SW5 are the memory 4 respectively.
Connected to a1 of A and b1 of the memory 4B and stored in a1 via line c and line d, respectively, and b1
The value stored in is input to the comparison calculation unit 5. a1
The value stored in the image unit P (1
The data amount of encoded image data corresponding to the area 1 in FIG. 10 in 1) is referred to as data C. The value stored in b1 is the data amount of the coded image data corresponding to the area 1 in FIG. 10 in the image unit P (12) of the configuration B in FIG. 9, which is called data D. That is, in order to compare the data C and the data D, the comparison operation unit 5 first obtains the absolute value α1 of the difference between the value of the data C and the data D.

Next, SW4 and SW5 are respectively connected to the memory 4A.
A2 and b2 of the memory 4B, and inputs them to the comparison operation unit 5. In the comparison operation unit 5, the value read from a2 is the new data C, the value read from b2 is the new data D, and the difference between the value of the data C and the data D is set in order to compare the data C with the data D. The absolute value α2 of is calculated.

Thereafter, the same procedure as above is used for SW4 and SW5.
While switching, the values from α3 to αn are obtained.

If the reproduced image of the image unit I (10), the reproduced image of the image unit P (11), and the reproduced image of the image unit P (12) are exactly the same, the reproduction of the image unit I (10) is performed. Image unit P (1
1) and the image unit P (12) coded by referring to the reproduced image of the image unit P (11) have the same value, and naturally the data amount is also exactly the same. That is, the data C and the data D have the same value. Therefore, at this time, the values of α1 to αn are all 0. If the reproduced image of the image unit I (10), the reproduced image of the image unit P (11), and the image unit P (12)
Even if the reproduced images are not the same, if there is no cut between the reproduced images, the image unit I (10)
In general, the reproduced image of P (11) is similar to the reproduced image of P (11) and the reproduced image of P (12) is similar to each other. Image unit P (11) encoded by referring to the reproduced image of I (10) and data of image unit P (12) encoded by referring to the reproduced image of image unit P (11) The amounts are close to each other, that is, the data C and the data D are close to each other. Therefore, most of the values of α1 to αn are
It is close to 0.

On the other hand, there is no cut between the reproduced image of the image unit I (10) and the reproduced image of the image unit P (11), and the reproduced image of the image unit P (11) and the image unit P (1
When there is a cut between the reproduced images in 2), the image unit P
There is no correlation between the reproduced image of (11) and the reproduced image of the image unit P (12), and the data amount of the image unit P (11) and the image unit P ( In general, the data amount of 12) is a value that is far apart from each other, that is, the values of the data C and the data D are far away from each other. Therefore, among the values of α1 to αn, those having values considerably larger than 0 appear.

Therefore, in the comparison calculation unit 5, a certain threshold value βpp
Is provided, and the reproduced image of the image unit I (10) and the image unit P
When there is no cut between the reproduced images of (11), the values themselves of α1 to αn are compared with the threshold βpp.
When some of the values of α1 to αn are larger than the threshold value βpp, the image unit P (11) and the image unit P (1
It is determined that there is a cut between 2), the cut detection signal is turned on and output, and if some of the values of α1 to αn are less than or equal to the threshold βpp, the image unit P (11) and the image unit P
It is determined that there is no cut during (12), and the cut detection signal is turned off and output.

After that, the next image unit is read from the recording medium 1, the data amount is counted by the counter 2 in the same procedure as described above, and the data amount of the old data in the reproduction image order is stored in the memory 4A and the memory 4B. The value of the new counter 2 is stored and updated in the memory that stores
A process similar to the above is performed in the comparison operation unit 5,
The presence / absence of a cut between reproduced images is determined. As an example, in FIG. 13, only forward predictive coded data is extracted from a series of coded image data groups that have six cuts in a reproduced image and are coded with the same configuration as the configuration C of FIG. Then
The graph which displayed the absolute value of the difference of the data amount between forward prediction encoding data is shown. In FIG. 13, the X axis represents the number of frames of the reproduced image, and the Y axis is the absolute value of the difference in the data amount between the forward prediction coded data. The six cuts are indicated by □ marks in FIG. As is clear from FIG. 13, a large value appears on the Y-axis where there is a cut. Therefore, by setting a certain threshold value βpp and comparing the value on the Y-axis with the threshold value βpp, the presence or absence of the cut can be determined.

The operation for determining the presence / absence of a cut between the forward-prediction-encoded image data and the reference image thereof and the forward-prediction-encoded image data of the configuration B of FIG. 9 by the method 4 will be described below. explain. The method 4 is a combination of the method 2 and the method 3.

First, encoded image data is read from the recording medium 1 of FIG. 3 from the image unit I (10) of the configuration B of FIG. Next, the image unit P (11) is read out, and the presence or absence of a cut between the image unit I (10) and the image unit P (11) is determined by the same processing as the method 2. Next, the image unit P (12) is read from the recording medium 1, and similarly, the data amount of each area is stored in the memory 4B. Then, as in the method 3, the memory 4A and the memory 4
The values stored in B are compared to determine whether or not there is a cut between the reproduced images of the image unit P (11) and the image unit P (12). After that, the presence or absence of the cut is determined by the same method as the method 3. However, when the forward predictive-coded image data in which the intra-image coded image is used as the reference image, such as the image unit P (11), is read next, the image unit I (10)
Similarly to the case where the presence / absence of the cut is determined between the image unit P (11) and the image unit P (11), the presence / absence of the cut is determined using Method 2.

The comparison function of the values input from the memories 4A and 4B in the comparison operation unit 5 is not limited to the above-described one, and the presence or absence of the cut is determined by using the ratio of the input values. May be determined. Further, the presence or absence of the cut may be determined by comparing a part or all of the values α1 to αn or the average value thereof with a threshold value.

FIG. 10 shows the structure in which the image unit is divided into each primary divided area. However, the dividing method is not limited to this, and may be divided in the horizontal direction or in a matrix form. good.

Further, in this embodiment, as shown in FIG. 3, the memory 4A and the memory 4B are arranged in parallel and the switch 4 is used by switching, but the memory 4A and the memory 4B are arranged in series. The new value may be stored in the memory 4A and the old value may be stored in the memory 4B.

Further, in the memory unit 4, the memory 4A and the memory 4 are
In addition to B, a plurality of memories such as a memory 4C and a memory 4D are provided, and the data amount of each image unit is stored in each memory. The presence or absence of a cut may be determined from the change rate of the data amount by detecting the time change of the data amount.

In the method 2 described above, the memory A
Although the example in which the memory B and the memory B are alternately used has been described, it may be configured to have only one of the memory A and the memory B, and to determine the presence / absence of the cut by using that.

It is not necessary that the operation of the method 2 and the operation of the method 3 in the method 4 are properly limited to the above.
After the image unit P (11) of the configuration B of FIG.
(14) The method 2 may be used for the determination of the presence or absence of the previous cut, or the method 3 may be used in combination. It should be noted that the control unit 3 is provided with a coding unit detection unit that detects whether the coded image data is coded in which image unit, frame or field, and according to the detection result, the comparison operation unit 5 Change the threshold value of, or the amount of data of two pieces of encoded image data obtained by encoding the field as an image unit is counted by the counter 2, and the count value and the encoded image data obtained by encoding the frame as an image unit Of the encoded image data obtained by comparing the data amount of each of the two in the comparison operation unit 5 or in the memory unit 4 and stored in the memory unit 4 in the comparison operation unit 5. The values for the two data amounts of the coded image data obtained by coding the specified fields in image units are summed, and the total value and the coded image data in which the frame is coded in image units are added. It may be configured to compare the data amount of data, in which case the two encoded image data to compare data amount,
One is a coded frame, and the other is a field coded. Therefore, even if the coded image units are different from each other, it is not necessary to erroneously determine the presence or absence of a cut.

In the above, the coded data coded by the variable length coding is used as the input data group, and the counter 2
In the configuration in which the data amount is counted and the value is stored in the memory 4, the encoded data encoded by the quantization is used as the input image data group, the counter 2 is replaced with the quantization parameter detection unit 9, A configuration may be employed in which the quantization parameter is detected and stored in the memory 4.

Further, the coded data coded by the variable length coding, the quantization, or the variable length coding and the quantization is used as the input image data, and both the counter 2 and the quantization parameter detecting section 9 are provided. A value obtained by multiplying or multiplying the value of the counter 2 and the value of the quantization parameter 9 may be stored in the memory 4. Further, in the first, second, third, and fourth embodiments, the presence / absence of a cut can be determined only in the encoded data having the configuration A in FIG. 9, but in the present embodiment, the fourth embodiment is used. In addition to having a function equivalent to that of FIG. 9, in any of the configurations A, B, and C of FIG. Between its reference images,
Also, it is possible to determine the presence / absence of a cut between image data that is forward-prediction coded by method 3. Further, in the case of forward predictive coding, since it is encoded by referring to the intra-picture coded data coded earlier than that or the forward predictive coded data, if there is a cut, the data amount or Since the value of the quantization parameter is much larger than that in the case where there is no cut, it is possible to determine whether or not there is a cut between intra-picture coded data and forward predictive coded data or between forward predictive coded data. . Also, when performing cut detection between intra-coded data and forward predictive coded data, or between forward predictive coded data, since the data amount of the image immediately after the cut or the quantization parameter becomes a large value, One of the memories 4A and 4B may be used to determine the presence / absence of a cut.

As described above, according to this embodiment, the intra-image information is encoded by variable-length encoding, or by quantization, or by the intra-image encoded data that is encoded by variable-length encoding and quantization. Using the image data group I or the intra-image coded data and the image located ahead in the reproduction time as a reference image, the difference value for each secondary divided region between the reference image and the image of interest is subjected to variable length coding or quantization. Alternatively, a coded image data group IP composed of forward predictive coded data coded by variable-length coding and quantization, or intra-picture coded data and forward predictive coded data, and a forward image at the reproduction time. Alternatively, the reference image and the target image can be used as a reference image using a rear image or an interpolated image of both the front and rear images.
Coded image data group I composed of bidirectional predictive coded data obtained by variable-length coding or quantizing the difference value for each sub-division area, or variable-length coding and quantization.
A control unit for generating any one of the PBs as an input image data group and generating a control signal for extracting the data amount of each image from the input image data group, and the input image data group and the control from the control unit. A counter that receives a signal as an input and counts a data amount of a part or all of the input image data group according to a control signal from the control unit, a storage unit that stores at least one value of the counter, and a control unit However, each image unit of the input image data group is provided with an image coding type detection unit that detects which method of intra-picture coding, forward predictive coding, or bidirectional predictive coding is coded, By comparing a value stored in the storage unit with a preset threshold value, or by comparing a value obtained by comparing and calculating a plurality of values stored in the storage unit with a preset threshold value. A comparison operation unit for determining whether there is a cut between intra-coded data in the input image data group, or between forward predictive coded data and its reference image, or between forward predictive coded data. As a result, the work of decoding the encoded image data read from the recording medium 1 which is required in the conventional example shown in FIG. 14 is unnecessary, and thus the cut can be detected in a shorter time. or,
Since a large-scale image data memory for storing the decoded image data as in the conventional example is not required, the device can be downsized and the cost can be reduced.

Further, in the comparison calculation unit 5 in the conventional example, the comparison calculation process for cut detection is performed using a large amount of image data, whereas a small number stored in the memory unit 4 in FIG. Since it suffices to compare the values, the time required for the processing in the comparison calculation unit 5 can be greatly reduced.
In summary, it is possible to speed up the process, reduce the size of the device, and reduce the cost of the entire device, as compared with the conventional example.

(Embodiment 6) A sixth embodiment of the present invention will be described below with reference to the drawings.

FIG. 4 shows the configuration of a coded image cut detection apparatus according to the sixth embodiment of the present invention. In FIG. 4, 1 is a recording medium, 2 is a counter, 3 is a control unit, 3A is an image unit detection unit, 3B is a region detection unit, 3C is an image coding type detection unit, 4 is a memory unit, and 4A and 4B.
Is a memory for storing the amount of data, 5 is a comparison operation part, and the above is the same as the configuration of the fifth embodiment of the present invention shown in FIG. What is different from FIG. 3 is that the control unit 3 reads from the image address memory 3D that stores the start address of the image unit, the read position control unit 3E that controls the read start position of the recording medium 1, and the recording medium 1. The point is that an address counter 3F for counting the value of the address of the encoded image data is provided.

The operation of the coded image cut detecting device configured as described above will be described below with reference to FIGS. 4, 9 and 10.

In this embodiment, the configuration A, the configuration B, and
The presence / absence of a cut is determined between the intra-image coded image data of configuration C, between the forward predictive-coded image data and its reference image, and between the forward predictive-coded image data. However, the operation will be described for each method by taking the case of using the data of the configuration B as an example. A combination of method 1 and method 2 of the third embodiment is referred to as method 5,
A combination of the methods 1 and 3 of the third embodiment is called a method 6, and a combination of the methods 1 and 4 of the third embodiment is called a method 7.

In the following, in the case of judging the presence / absence of a cut between the intra-coded image data of the configuration B of FIG. 9 and between the forward predictive-coded image data and its reference image by the method 5 The operation will be described.

In the encoded image data shown in FIG. 9 from the recording medium 1 in FIG. 4, the image unit I (1
Data is read in order from the beginning of 0). When the image-unit detecting unit 3A in FIG. 4 detects the leading data of the image unit, the image-unit detecting unit 3A transmits a count start signal to the address counter 3F. The address counter 3F is
Start counting the amount of data. After that, the data amount is stored in the memory 4A in the same procedure as the method 1 of the third embodiment. Next, when the image unit detection unit 3A detects the beginning of the image unit P (11) of the configuration B of FIG. 9, the address counter 3F
The count value at that time is transmitted and stored in the image address memory as the address value of the image unit P (11). After that, the presence or absence of the cut between the image unit I (10) and the image unit I (15) of the configuration B of FIG. 9 is determined by the same procedure as the method 1 of the third embodiment.

Then, the address value of the next image unit of the image unit I (15) is stored in the image address memory 3D. If it is determined that there is no cut, the presence / absence of a cut between the image unit I (15) and the next intra-coded data is determined by the same procedure as in the method 1 of the third embodiment. When it is determined that there is a cut between the image unit I (10) and the image unit I (15), the count of the address counter 3F is stopped. Then, with reference to the address value of the image unit P (11) stored in the image address memory 3D, the read position control unit 3E moves the read position of the recording medium 1 to the head position of the image unit P (11). After that, the image unit P (1
Whether there is a cut between 1) and its reference image is determined. Image unit P (12), image unit P (13), image unit P (1
The same applies to 4).

The operation for determining the presence / absence of a cut between the intra-image coded image data of the configuration B of FIG. 9 and the forward predictive-coded image data by the method 6 will be described below. Method 6 is the third method in Method 5.
This is the same as the system 3 of the third embodiment, except that the portion performing the same operation as the system 2 of the third embodiment is replaced.

In the following, according to the method 7, between the intra-image coded image data of the configuration B of FIG. 9, between the forward predictive coded image data and its reference image, and between the forward predictive coded image data. The operation for determining the presence / absence of a cut will be described. The method 7 is the same as the method 5 in which the same operation as the method 2 of the third embodiment is replaced with the method 4 of the third embodiment.

In this embodiment, the address counter 3F and the image address memory 3D are provided in order to calculate the read position from the recording medium 1 in FIG. 4, but instead of using these, each image is previously stored. Information on which address on the recording medium 1 the unit is recorded is calculated and recorded in a table,
With reference to this table, the read position control unit 3E
Then, the read position may be moved.

Further, the switching of each system may be performed by the comparison operation section 5 with reference to the output of the image coding type detection section 3C, or the control section 3 or the comparison operation section 5 or the control section 3 from the outside. Alternatively, the method switching signal may be input to both of the comparison calculation unit 5 and the comparison calculation unit 5. It should be noted that the control unit 3 is provided with a coding unit detection unit that detects whether the coded image data is coded in a frame or field image unit.
Depending on the detection result, the threshold value in the comparison operation unit 5 is changed, or the counter 2 counts the data amount of two pieces of encoded image data obtained by encoding the field as an image unit and the count value. And the data amount of the encoded image data encoded by the frame as the image unit are compared by the comparison operation unit 5, or two data amounts of the encoded image data by which the field is encoded as the image unit are stored in the memory unit 4. Then, the comparison operation unit 5 sums the values of two data amounts of the encoded image data obtained by encoding the fields stored in the memory unit 4 as an image unit, and the sum and the frame are encoded as an image unit. It may be configured to compare the data amounts of the encoded image data that have been encoded. In that case, two encoded image data items that compare the data amounts may have one encoded frame. In, so that the other is obtained by encoding the field, it is not necessary to determine erroneously whether the cut even if the image unit obtained by encoding different.

In the above description, the coded data coded by the variable length coding is used as the input image data group, the counter 2 counts the data amount, and the value is stored in the memory 4. The encoded data encoded by the above may be used as the input image data group, and the counter 2 may be replaced with the quantization parameter detection unit 9 to detect the quantization parameter and store it in the memory 4. The input data is coded data that has been coded by variable length coding, quantization, or variable length coding and quantization, and is provided with both the counter 2 and the quantization parameter detection unit 9 A value obtained by multiplying the value and the value of the quantization parameter 9 by multiplication or exponentiation may be stored in the memory 4.

As described above, according to this embodiment, a group of coded image data composed of coded image data coded by variable-length coding is used as an input image data group, and A control unit that generates a control signal for extracting the data amount of each image, the input image data group and the control signal from the control unit as inputs, and the input image data according to the control signal from the control unit. A counter that counts a part or all of the data amount of the group, a storage unit that stores at least one value of the counter, and a value stored in the storage unit and a preset threshold value are compared,
Alternatively, the presence or absence of a cut between encoded image data in the input image data group is determined by comparing a value obtained by performing a comparison operation on a plurality of values stored in the storage unit with a preset threshold value. A read / write position control section for controlling a read position of the recording medium, and a storage section for storing the read position of the recording medium; Alternatively, by adopting a configuration including both the storage unit and the read position calculation unit of the recording medium, the configuration shown in FIG.
In the conventional example shown in FIG. 4, the decoding work of the coded image data read from the recording medium 1 which is required, is unnecessary, so that the cut can be detected in a shorter time.

Further, unlike the conventional example, a large-scale image data memory for storing the decoded image data is unnecessary, so that the apparatus can be downsized and the cost can be reduced. Further, in the comparison calculation unit 5, in the conventional example, a comparison calculation process for detecting a cut is performed using a large amount of image data, whereas a comparison of a small number of values stored in the memory unit 4 in FIG. Since it suffices to perform the above, the time required for the processing in the comparison calculation unit 5 can be greatly reduced. In summary, it is possible to speed up the process, reduce the size of the device, and reduce the cost of the entire device, as compared with the conventional example.

Further, in the fifth embodiment, the range in which the presence / absence of the cut can be judged is limited by each method.
In this embodiment, which has a configuration in which the methods of the above embodiments are combined, the cut determination can be performed more finely. In addition, in any of the methods of the present embodiment, the method 1 of the fifth embodiment is used to first determine the presence / absence of a cut between intra-image coded data, and then determine between the intra-image coded images with cuts. Since each method is used to finely determine the presence or absence of a cut, if it is determined that there is no cut between intra-coded images, the presence or absence of a cut between the next intra-coded images is determined. Since it is possible to shift to the step, the number of times of processing for comparison calculation can be reduced, so that the determination of the presence or absence of the cut of the entire image data can be completed in a short time.

(Embodiment 7) A seventh embodiment of the present invention will be described below.

FIG. 5 shows the configuration of a coded image cut detection apparatus according to the seventh embodiment of the present invention. In FIG. 5, 1 is a recording medium, 2 is a counter, 3 is a control unit, 3A is an image unit detection unit, 3B is an area detection unit, 3C is an image coding type detection unit, 3D is an image address memory,
3E is a read position controller, 3F is an address counter,
Reference numeral 4 is a memory unit, 4A and 4B are memories, 5 is a comparison operation unit, and the above is the same as the configuration of the sixth embodiment of the present invention shown in FIG. What is different from FIG. 4 is that which of the front image, the rear image, and the interpolated images of the front image and the rear image, each secondary divided region in the bidirectional predictive encoded image is referred to and encoded. Reference image determination unit 6 for determination
In addition, the cut determination unit 7 that determines the position of the cut based on the determination result of the reference image determination unit 6 is provided. The operation of the coded image cut detection device configured as described above will be described below with reference to FIGS. 5 and 9. In the present embodiment, the presence or absence of a cut between each image is determined in the encoded image data of any of the configurations A, B and C of FIG. The operation will be described by taking the case of using the data of the configuration C as an example. From the recording medium 1 of FIG. 5 to the image unit I (12) to the image unit P (15) of the encoded image data of the configuration C of FIG.
In the same manner as in the embodiment of FIG.
It is assumed that there is a cut between the reproduced image of (12) and the reproduced image of the image unit P (15), and the cut detection signal is turned on and output. The details of the operation up to this point are the same as in the sixth embodiment, and will not be repeated.

Next, the image unit B (13) in FIG. 9 is read out, and it is determined that the image unit B (13) is an image unit bidirectionally predictively coded by the image unit detection unit 3A and the image coding type detection unit 3C in FIG. To be detected. To the reference image determination unit 6, the encoded image data read from the recording medium 1, the detection result from the image encoding type detection unit 3C, and the cut detection signal from the comparison calculation unit 5 are input. The reference image determination unit 6
When the output of the image coding type detection unit 3C is a signal indicating bidirectional predictive coding and the cut detection signal of the comparison calculation unit 5 is ON, the bidirectional predictive coded image currently being read It is possible to detect whether the data, that is, each secondary divided area of the image unit B (13) of the configuration B of FIG. 9 is coded by referring to which image of the front image, the rear image, and the interpolated image of the front image and the rear image. Then, the number J of secondary divided areas coded with reference to the front image and the number K of secondary divided areas coded with reference to the rear image are transmitted to the cut determination unit 7.

In the cut determination section 7, the value of J is a threshold value γ
If it is J or more, it is considered that there is no cut between the reproduced image of the image unit I (12) and the reproduced image of the image unit B (13). If the value of J is less than a certain threshold value γJ or if the value of K is more than a certain threshold value γK, it is considered that there is a cut between the reproduced image of the image unit I (12) and the reproduced image of the image unit B (13), Turn on and output the final cut detection signal.
Reproduced image of image unit I (12) and image unit B (13)
If there is no cut between the playback images of,
For the image unit B (14), the same processing as that for the image unit B (13) is performed. However, image unit B (14)
When the value of K is less than the threshold value γK, the reproduced image of the image unit B (14) and the reproduced image of the image unit P (15) are regarded as having no correlation, and the reproduced image of the image unit (14) is considered to have no correlation. Then, it is determined that there is a cut between the reproduced images of the image unit P (15).

Although it has been described above that the signals transmitted from the reference image determination unit 6 to the cut determination unit 7 are both J and K, either one of J and K is transmitted and transmitted. J
Alternatively, the presence or absence of a cut may be determined by comparing the value of K with each threshold value. In addition, the encoded image data input to the image encoding type detection unit 3C is bidirectionally predictively encoded so that the reference image determination unit 6 or the cut determination unit 7 is not provided with the output of the comparison operation unit 5. The reference image determination unit 6 may be operated when it is detected that there is. Further, in the above, the reference image determination unit 6 operates when the image encoding type detection unit 3C detects that the input encoded image data is bidirectional predictive encoded. The reference image determination unit 6 may operate when the encoding type detection unit 3C detects that the input encoded image data is forward predictive encoded.

The control unit 3 is provided with a coding unit detection unit for detecting whether the coded image data is coded in the image unit of the frame or the field, and the comparison operation unit according to the detection result. 5, the threshold value is changed, or the data amount of two pieces of encoded image data in which a field is encoded in image units is counted by the counter 2, and the count value and the frame are encoded in image units. The comparison operation unit 5 compares the data amounts of the encoded image data, or the memory unit 4 stores two data amounts of the encoded image data obtained by encoding the field by image exhaustion and compares them. The arithmetic unit 5 sums the values of two data amounts of encoded image data in which the fields stored in the memory unit 4 are encoded in image units, and the total value and the frame are encoded in image units. The coded image data may be compared in data amount, and in this case, two coded image data items in which the data amounts are compared are one in which a frame is encoded and the other in which a field is encoded. Even if the encoded image units are different from each other, it is not necessary to erroneously determine the presence or absence of a cut.

As described above, according to this embodiment, the intra-picture coded data obtained by coding the intra-picture information by the variable length coding and the picture located ahead in the reproduction time as the reference picture are considered as the reference picture. Refer to the forward predictive coded data obtained by coding the difference value for each secondary divided area from the image by variable length coding, and the forward image or the backward image or the interpolated image of the forward and backward images at the reproduction time. An input image data group is an encoded image data group IPB including bidirectional predictive encoded data obtained by encoding the difference value of each secondary divided area between the reference image and the image of interest as an image by variable length encoding. A control unit for generating a control signal for extracting a data amount of each image from the input image data group; A control signal from the input image data group and the control unit, that is, a count start signal, provided with an image coding type detection unit that detects whether the coding is performed by measurement coding or bidirectional predictive coding. A counter for counting a part or all of the data amount of the input image data group in response to a control signal from the control unit and a count stop signal or a zero clear signal, and at least one value of the counter is stored. The storage unit compares a value stored in the storage unit with a preset threshold value, or compares a value obtained by comparing and calculating a plurality of values stored in the storage unit with a preset threshold value. The input image data group and the image code have a comparison calculation unit that determines whether or not there is a cut between the encoded image data in the input image data group. The output of the type detection unit or the input image data group and the output of the image coding type detection unit and the output of the comparison operation unit are input, and the coded image data in the input image data group is the bidirectional predictive coded data. In this case, it is detected whether each secondary divided area is coded by referring to a front image in the reproduction time or a rear image in the reproduction time, and the coded image data in the input image data group is forwarded. In the case of predictive-coded data, it is detected whether each secondary divided area is encoded by referring to the preceding image at the reproduction time, and the secondary divided area of the encoded by referring to the preceding image at the reproduction time is detected. Number J, or 2 encoded with reference to a backward image in playback time
The reference image determination unit that outputs the number K of the next divided areas, or both the J and the K, and the output of the reference image determination unit, or both the output of the reference image determination unit and the output of the comparison performance unit. The conventional configuration shown in FIG. 14 has a configuration including a cut determination unit that determines the presence or absence of a cut before and after bidirectional predictive encoded data as an input, or the presence or absence of a cut between forward predictive encoded data and its reference image. In the example, the decoding work of the encoded image data read from the recording medium 1 which is necessary in the example is not necessary, so that the cut can be detected in a shorter time.

Further, unlike the conventional example, a large-scale image data memory for storing the decoded image data is not required, so that the apparatus can be downsized and the cost can be reduced. Further, in the comparison calculation unit 5, in the conventional example, a comparison calculation process for detecting a cut is performed using a large amount of image data, whereas a comparison of a small number of values stored in the memory unit 4 in FIG. And whether each of the secondary divided areas of the bidirectional prediction coded image refers to the preceding or following image, or whether each of the secondary divided areas of the forward predictive coded image refers to the forward image. Since it is only necessary to detect, the time required for the processing in the comparison calculation unit 5 can be significantly shortened, and the processing in the cut determination unit 7 can be completed in a short time. In summary, it is possible to speed up the process, reduce the size of the device, and reduce the cost of the entire device, as compared with the conventional example. Further, in the sixth embodiment, it was not possible to judge the presence / absence of a cut before and after the bidirectionally predicted image in the encoded image data of the configuration C in FIG. 9, but this embodiment is equivalent to the sixth embodiment. Besides having functions,
Even in the case of data having a configuration including a bidirectional predictive-coded image as in the configuration C of FIG. 9, it is possible to determine the presence / absence of a cut between all images.

(Embodiment 8) An eighth embodiment of the present invention will be described below with reference to the drawings. FIG. 6 shows the configuration of the coded image cut detection apparatus according to the eighth embodiment of the present invention.

In FIG. 6, 1 is a recording medium, 3 is a control unit, 3A is an image unit detection unit, 3C is an image coding type detection unit, 6 is a reference image determination unit, and 7 is a cut determination unit, which are shown in FIG. It is similar to the seventh embodiment. The operation of the coded image cut detection device configured as described above will be described below with reference to FIGS. 6 and 9. In the present embodiment, the presence / absence of a cut is determined between the forward predictive-coded image data and the reference images of the configurations B and C in FIG. 9 and before and after the bidirectional predictive-coded image data. . The operation will be described by taking the case of using the data of configuration C as an example. First, the recording medium 1 of FIG.
From the coded image data of the configuration C of FIG. 9, the coded image data of the image unit I (12) is read out, and the image unit detection unit 3A, the image coding type detection unit 3C, and the reference image determination of FIG. Input to the part 6. Image unit detection unit 3A in FIG.
The image coding type detection unit 3C detects that it is an image-coded image unit, and skips reading.

Next, the image unit B (10) of FIG. 9 is read from the recording medium 1, and the image is bidirectionally predictively coded by the image unit detection unit 3A and the image coding type detection unit 3C of FIG. The unit is detected. The reference image determination unit 6 receives the encoded image data read from the recording medium 1 and the detection result from the image encoding type detection unit 3C, and the output of the image encoding type detection unit 3C is bidirectional predictive encoding. Since it is a signal indicating that
Of each of the secondary divided areas of the image unit B (10) of the configuration B of No. 3B, the front image, the rear image, and the interpolated image of the front image and the rear image are referred to be detected to detect the front image. The number J of secondary divided areas coded by reference,
The number K of secondary divided areas coded with reference to the rear image is transmitted to the cut determination unit 7. If the value of J is equal to or greater than a certain threshold value γJ, the cut determination unit 7 determines that there is no cut between the reproduced image of the image unit B (10) and the preceding reference image in the reproduction time of the image unit B (10), If the value of J is less than the threshold value γJ, it is considered that there is a cut between the reproduced image of the image unit B (10) and the preceding reference image in the reproduction time of the image unit B (10), and the final cut detection signal is turned on. Output.

If the value of K is greater than or equal to a certain threshold value γK, it is considered that there is no cut between the reproduced image of the image unit B (10) and the reference image at the rear during the reproduction time of the image unit B (10), and the value of K is Image unit B (10) if less than the threshold γK
It is considered that there is a cut between the reproduced image of B and the reference image in the rear in the reproduction time of the image unit B (10), and the final cut detection signal is turned on and output. With respect to subsequent bidirectional predictive coded data, the presence / absence of a cut is similarly determined.

Although it has been described above that the signals transmitted from the reference image determination unit 6 to the cut determination unit 7 are both J and K, either one of J and K is transmitted and transmitted. J
Alternatively, the presence or absence of a cut may be determined by comparing the value of K with each threshold value.

Further, in the above, the image coding type detection unit 3
When it is detected that the encoded image data to which C is input is bidirectionally predictively encoded, the reference image determination unit 6 operates. However, the image encoding type detection unit 3C inputs the code. The reference image determination unit 6 may be operated when it is detected that the encoded image data is forward predictively encoded. In that case, in the forward predictive encoded data, the number J of secondary divided areas encoded by referring to the forward reference image in the reproduction time is detected, and the cut determination unit 7 sets the value of J and the threshold γJ. By comparing, it is determined whether or not there is a cut between the reproduced image of the forward predictive encoded data and the forward reference image at the reproduction time.

As described above, according to this embodiment, the intra-picture coded data in which the intra-picture information is coded by the variable length coding and the reference picture with the picture located at the front in the reproduction time as the reference picture are used. A coded image data group IP composed of forward predictive coded data obtained by coding a difference value for each secondary divided area from the image of interest by variable length coding, or intra-coded data and forward predictive coded data And the difference value for each secondary divided area between the reference image and the target image is used as a reference image using the front image or the rear image or the interpolated images of both the front and rear images in the reproduction time by variable length encoding. The encoded image data group IPB composed of the bidirectional predictive encoded data is used as an input image data group, and a control unit is provided, and the control unit detects each image unit of the input image data group. An image unit type detection unit and an image coding type detection unit that detects whether each image unit of the input image data group is coded by intra-picture coding, forward prediction coding, or bidirectional prediction coding Each of the input image data groups, and when the encoded image data in the input image data group is bidirectional predictive encoded data using the output of the input image group and the image encoding type detection unit as input. Two
It is detected whether the next divided area is coded by referring to the image ahead in the reproduction time or the image posterior in the reproduction time, and the encoded image data in the input image data group is forward predictive encoded data. In this case, it is detected whether or not each secondary divided area is encoded by referring to the preceding image in the reproduction time, and the number J of secondary divided areas encoded by referring to the preceding image in the reproduction time, or The number K of secondary divided areas coded with reference to a rear image in the reproduction time,
Alternatively, the bidirectional predictive coding with the reference image determination unit that outputs both the J and the K and the output of the reference image determination unit, or both the output of the reference image determination unit and the output of the comparison operation unit as inputs. A recording medium required in the conventional example shown in FIG. 14 is provided by providing a cut determination unit that determines the presence or absence of a cut before and after data, or the presence or absence of a cut between forward predictive encoded data and its reference image. Since the work of decoding the coded image data read from No. 1 is unnecessary, it is possible to detect the cut in a shorter time.

Further, since a large-scale image data memory for storing the decoded image data as in the conventional example is not required, the apparatus can be downsized and the cost can be reduced. Further, in the conventional example, the comparison calculation unit 5 in FIG. 14 uses a large amount of image data to perform the comparison calculation process for cut detection, whereas in the present embodiment, each of the two bidirectional predictive encoded images is processed. Since it is only necessary to detect which of the preceding and following images the next divided region refers to, or whether each secondary divided region of the forward prediction coded image refers to the preceding image, the cut determination unit 7 Processing can be done in a short time.

In summary, it is possible to increase the processing speed, reduce the size of the device, and reduce the cost of the device as a whole, as compared with the conventional example. Further, according to this embodiment, the device can be made smaller than that of the seventh embodiment.

(Embodiment 9) A ninth embodiment of the present invention will be described below with reference to the drawings. In this embodiment, the same function as that of the first embodiment of the present invention is realized by software on a computer. FIG. 7 shows a flowchart of the operation of this embodiment, and the operation will be described with reference to FIGS. 7 and 9.

In the present embodiment, as in the first embodiment of the present invention, the presence / absence of a cut is determined between the image data that has been intra-coded as in the configuration A in FIG. In FIG. 7, first, in step 1, variables C and D are initialized to 0. Both the variable C and the variable D hold the value of the data amount of each image.
Next, in step 2, image-by-image detection is performed. As an example, the first image unit is the image unit I (10) of the configuration A in FIG. Next, in step 3 of FIG. 7, the value of the variable C is substituted for the variable D. At this time, 0 is assigned to the variables C and D. Next, in step 4, the image unit I of the configuration A of FIG.
The data amount of (10) is counted, and the value of the data amount is substituted into the variable C.

Next, in step 5 of FIG. 7, the absolute value of the difference between the variable C and the variable D is calculated.
Then, it is determined that there is a cut, and if it is less than the threshold value β, it is determined in step 6 that there is no cut. The value of the data amount of the currently read image unit is assigned to the variable C, and the value of the data amount of the previous image unit in the playback time is assigned to the variable D. Then, if the preceding image unit is exactly the same in the reproduction time, the absolute value of the difference between the value of the variable C and the value of the variable D becomes zero. Even if the image unit currently being read is not exactly the same as the preceding image unit in the reproduction time, if there is no cut between them, the difference between the respective data amounts is generally small. The absolute value of the difference between the values of variable D is 0
It becomes a value close to. However, if there is a cut between the image unit that is currently being read and the image unit that is ahead in the playback time, the difference in the respective data amounts is generally large, so the difference between the value of variable C and the value of variable D is large. The absolute value of becomes large. Therefore, by setting a certain threshold value β and comparing the absolute value of the difference between the value of the variable C and the value of the variable D with the threshold value β, the presence or absence of the cut can be determined. As an example, FIG.
In the reproduced image, only the intra-image coded data is extracted from the series of coded image data groups that have 6 cuts and are encoded by the same configuration as the configuration C of FIG. The graph which displayed the absolute value of the difference of the data amount between coded data is shown.

In FIG. 11, the X-axis shows the number of frames of the reproduced image, and the Y-axis shows the absolute value of the difference in the data amount between the intra-picture coded data. The six cuts are indicated by □ in FIG. As is clear from FIG. 11, a large value appears on the Y-axis where there is a cut. Therefore, by setting a certain threshold β and comparing the Y-axis value with the threshold β, the presence or absence of the cut can be determined. Next, in step 7 of FIG. 7, it is determined whether the data has ended. If the data is finished, all processes are terminated.

If the data has not ended, step 2
Then, the process similar to the above is repeated thereafter. The above is the description of the operation of the present embodiment.

The operation for comparing the variable C and the variable D in step 5 in FIG. 7 is not limited to the above, and the presence / absence of the cut is determined using the ratio of the variable C and the variable D. You may. Further, in addition to the variables C and D, a plurality of variables are provided, each variable stores a different data amount of each image unit, the time change of the data amount of each image unit is detected, and the change of the data amount is detected. The presence or absence of the cut may be determined from the rate.

The presence / absence of a cut may be determined not only between continuous image units as described above but also between non-adjacent images.

A means for detecting whether the encoded image data is encoded in the image unit of the frame or the field is provided, and the threshold value in step 5 of FIG. 7 is changed according to the detection result. Or, the data amount of two pieces of encoded image data obtained by encoding a field as an image unit is counted, and the count value is compared with the data amount of encoded image data obtained by encoding a frame as an image unit in step 5. By using the method, one is a frame-encoded one and the other is a field-encoded one. I'm done.

As described above, according to the present embodiment, the coded image data group formed from the coded intra-picture data obtained by coding the intra-picture information by variable-length coding is used as the input image data group, and the input image An image unit detection means as a means for generating a control signal for extracting the data amount of each image from the data group, and counting the data amount of a part or all of the input image data group according to the control signal. And means for substituting at least one of the values into the variable group, and comparing the value of the variable group with a preset threshold value, or a value obtained by comparing and calculating a plurality of values of the variable group in advance. By comparing with the set threshold value,
By having a means for determining the presence / absence of a cut between intra-coded data in the image data group, the decoding work of the coded image data, which was necessary in the conventional example shown in FIG. 14, becomes unnecessary. High-speed cut detection is possible.

(Tenth Embodiment) A tenth embodiment of the present invention will be described below with reference to the drawings. This embodiment implements the same function as that of the second embodiment of the present invention by software on a computer. FIG. 18 shows a flowchart of the operation of this embodiment, and the operation will be described with reference to FIGS. 18, 9 and 10.

In the present embodiment, as in the second embodiment of the present invention, the presence / absence of a cut is determined between image data which are intra-coded as in the configuration A in FIG. In FIG. 18, first, in step 1, variable X and variable Y
Are initialized to 0. Both the variable X and the variable Y hold the value of the quantization parameter for each image. Next, in step 2, image-by-image detection is performed.
As an example, the first image unit is the image unit I of the configuration A of FIG.
(10). Next, in step 3 of FIG. 18, the value of the variable X is substituted for the variable Y. In this case, variable Y and variable X
Is assigned 0. Next, in step 4, the quantization parameter of the image unit I (10) of the configuration A of FIG. 9 is detected,
The value of the quantization parameter is assigned to the variable X.

Next, in step 5 of FIG. 18, variable X and variable Y
Takes the absolute value of the difference, and if it is greater than or equal to the threshold value β, step 6
Then, it is determined that there is a cut, and if it is less than the threshold value β, it is determined in step 6 that there is no cut. The value of the quantization parameter for each image currently read is substituted for the variable X,
The value of the quantization parameter of the image unit in the front in the reproduction time is assigned to the variable Y, and if the image unit currently read and the image unit in the front in the reproduction time are exactly the same, the variable The absolute value of the difference between the value of X and the value of variable Y becomes zero. Even if the image unit currently being read is not exactly the same as the preceding image unit in the playback time, the difference between the quantization parameters is generally small if there is no cut between them, and the value of the variable X And the absolute value of the difference between the values of the variable Y is close to 0. But,
If there is a cut between the image unit that is currently being read and the image unit that precedes it in the playback time, the difference between the respective quantization parameters is generally large, so the difference between the value of the variable X and the value of the variable Y is The absolute value becomes large. Therefore, by setting a certain threshold value β and comparing the absolute value of the difference between the value of the variable X and the value of the variable Y with the threshold value β, it is possible to determine the presence or absence of the cut.

Next, in step 7 of FIG. 18, it is determined whether the data has ended. If the data is finished, all processes are terminated. If the data has not ended, the process returns to step 2 and the same processing as above is repeated thereafter. The above is the description of the operation of the present embodiment.

The operation for comparing the variable X and the variable Y in step 5 in FIG. 18 is not limited to the above, and the presence / absence of a cut is determined using the ratio of the variable X and the variable Y. May be. In addition to the variables X and Y, a plurality of variables are provided, and each variable stores a quantization parameter for each image unit, and the time change of the quantization parameter for each image unit is detected to perform quantization. The presence / absence of a cut may be determined from the rate of change of parameters. Further, the presence / absence of a cut may be determined not only between continuous image units as described above but also between non-adjacent images.

Further, there is provided means for detecting whether the encoded image data is encoded in the image unit of the frame or the field, and step 5 of FIG. 7 is executed according to the detection result.
The threshold value of is changed, or the comparison unit in the comparison operation unit 5 is controlled to be a field unit comparison or a frame unit comparison according to the detection result.

As described above, according to the present embodiment, the coded image data group constituted by the coded image data obtained by coding the image information by quantization is used as the input image data group, and the input image data group is set. Image unit detecting means as means for generating a control signal for detecting the quantization parameter of each image, and detecting some or all of the quantization parameters of the input image data group according to the control signal. And means to assign at least one value to the variable group,
Comparing the value of the variable group with a preset threshold value,
Alternatively, means for determining the presence / absence of a cut between intra-image encoded data in the input image data group by comparing a value obtained by comparing and calculating a plurality of values of the variable group with a preset threshold value With this, the work of decoding the encoded image data, which was necessary in the conventional example shown in FIG. 14, is not necessary, and thus faster cut detection is possible.

(Eleventh Embodiment) An eleventh embodiment of the present invention will be described below with reference to the drawings. In this embodiment, the same function as that of the third embodiment of the present invention is realized by software on a computer. FIG. 19 shows a flowchart of the operation of this embodiment, and the operation will be described with reference to FIGS. 19, 9 and 10. It should be noted that this embodiment is similar to the third embodiment of the present invention in that the presence or absence of a cut is determined between image data that has been intra-coded as in the configuration A in FIG.

In FIG. 19, first, in step 1, variables S, T, X and Y are initialized to 0. The variable S and the variable T hold the value of the quantization parameter and the data amount of each image, and the variable X and the variable Y hold the product of the values of the variable S and the variable T. Is. Next, in step 2, image-by-image detection is performed. As an example, the first image unit is the image unit I (10) of the configuration A in FIG. Next, in step 3 of FIG. 19, the value of the variable X is substituted for the variable Y. At this time, 0 is substituted for the variables Y and X. Next, in step 4, the quantization parameter of the image unit I (10) of the configuration A of FIG. 9 is detected, the value of the quantization parameter is substituted into the variable S, and the data amount of the image unit I (10) is counted. Then, the count value is substituted into the variable T.

Next, in step 5 of FIG. 19, the value of the variable S and the value of the variable T are multiplied and substituted into the variable X. Next, in step 6 of FIG. 19, the absolute value of the difference between the variable X and the variable Y is taken, and if it is the threshold value β or more, it is determined in step 7 that there is a cut,
If it is less than the threshold value β, it is determined in step 7 that there is no cut. The multiplication value of the quantization parameter value of the currently read image unit and the data amount is substituted into the variable X, and the variable Y
Means that the multiplication value of the quantization parameter value of the forward image unit and the data amount at the reproduction time is substituted, and the currently read image unit and the forward image unit at the reproduction time are exactly the same. Even if there is no cut between them, the difference between the respective quantization parameters or data amount becomes small, and the absolute value of the difference between the value of the variable X and the value of the variable Y becomes 0 or a value close to 0. Become. However, if there is a cut between the currently read image unit and the preceding image unit in the reproduction time, the difference in the quantization parameter or the data amount becomes large, so the value of the variable X and the value of the variable Y are large. The absolute value of the difference of becomes large. Therefore, by setting a certain threshold value β and comparing the absolute value of the difference between the value of the variable X and the value of the variable Y with the threshold value β, it is possible to determine the presence or absence of the cut. Next, in step 8 of FIG. 19, it is determined whether the data has ended. If the data is finished, all processes are terminated. If the data has not ended, the process returns to step 2 and the same processing as above is repeated thereafter. The above is the description of the operation of the present embodiment.

The operation for comparing the variable X and the variable Y in step 5 in FIG. 18 is not limited to the above, and the presence or absence of the cut is determined using the ratio of the variable X and the variable Y. May be. In addition to the variables X and Y, a plurality of variables are provided, and each variable stores a multiplication value of the quantization parameter and the data amount of each image unit, and the quantization parameter and the data amount of each image unit are stored. It is also possible to detect the change over time and determine the presence or absence of a cut from the change rate of the multiplication value of the quantization parameter and the data amount. Further, the presence / absence of a cut may be determined not only between continuous image units as described above but also between non-adjacent images. Further, there is provided means for detecting whether the coded image data is coded in the image unit of the frame or the field, and the threshold value in step 5 of FIG. 7 is changed according to the detection result, or Depending on the detection result, the comparison unit 5 controls whether the comparison unit is a field unit comparison or a frame unit comparison. Further, in the above, an example in which the quantization parameter and each value of the data amount are multiplied has been shown, but exponentiation may be used instead of multiplication.

As described above, according to this embodiment, a group of coded image data composed of coded data coded by variable length coding or quantization, or variable length coding and quantization is input. An image data group, an image unit detection means as a means for generating a control signal for extracting the data amount of each image or detecting a quantization parameter from the input image data group, and the image signal detection means according to the control signal. Counting the data amount of a part or all of the input image data group, detecting the quantization parameter of a part or all of the input data group according to the control signal, and calculating the count value and the value of the quantization parameter. , Means for substituting at least one of the values into a variable group, and a value obtained by comparing the value of the variable group with a preset threshold value or comparing and calculating a plurality of values of the variable group. By having a means for determining the presence or absence of a cut between the image coded data of the input image data in the group by comparing the order set threshold,
Cut detection can be performed for both input data, that is, encoded data in which the amount of data encoded using the variable-length encoding method varies and encoded data that is encoded using the quantization method.

Further, in the case of coded data using quantization and variable length coding like MPEG, in the case of coded data in which the quantization parameter is constant or almost constant and the data size changes greatly by cutting, Even in the case of encoded data in which the quantization parameter greatly changes due to the cut, the cut can be detected without depending on the content of such data. Further, in the case of coded data in which the change in the data amount due to the cut of the input data is small and the change in the quantization parameter is small, the method of determining the cut by the change of the data amount can be the method of determining the cut by the change of the quantization parameter. However, it is difficult to detect the cut, but by calculating the data amount and the quantization parameter, the change in the data due to the cut can be greatly expressed, and the cut is judged by the change in the data amount only and the change in the quantization parameter only. More accurate cut detection is possible. Since the work of decoding the encoded image data, which is required in the conventional example shown in FIG. 14, is unnecessary, it is possible to detect the cut at a higher speed.

(Embodiment 12) A twelfth embodiment of the present invention will be described below with reference to the drawings. This embodiment realizes the same function as that of the eighth embodiment of the present invention by software on a computer. FIG. 8 shows a flowchart of the operation of this embodiment, and the operation will be described with reference to FIGS. 8 and 9.

Note that this embodiment is similar to the seventh embodiment of the present invention in that the forward predictive-coded image data of the configurations B and C in FIG. 9 and their reference images, and bidirectional predictive encoding. The presence / absence of a cut is determined before and after the image data. The operation will be described by taking the case of using the data of configuration C as an example. First, in step 1 of FIG. 8, 0 is assigned to the variables J and K. Next, in step 2, image units are detected. At this time, the coded image data of the image unit I (12) is detected from the coded image data of the configuration C of FIG. Next, in step 3 of FIG. 8, it is detected that the image unit I (12) of the configuration C of FIG. 9 is an image-coded image unit, and the process jumps to step 6 of FIG. In step 6, it is determined whether the data has ended. Since the data has not ended, the process goes to step 2. In step 2, the image unit B of the configuration C of FIG. 9 is set as the next image unit.
(10) is detected.

Next, in step 3 of FIG. 8, it is detected that the image unit B (10) of the configuration C of FIG. 9 is bidirectionally predictive coded. Next, in step 4 of FIG. 8, the configuration B of FIG.
Each of the secondary divided areas of the image unit B (10) of No. 1 is detected by referring to which image of the front image, the rear image, and the interpolated image of the front image and the rear image is encoded, and the front image is referred to. The number of coded secondary divided areas is substituted into a variable J, and the number of coded secondary divided areas with reference to a rear image is substituted into a variable K. Next, in step 5 of FIG.
If the value of is greater than or equal to a certain threshold value γJ, the image unit B (10)
It is determined that there is no cut between the playback image of the image unit B (10) and the reference image ahead of the playback time of the image unit B (10), and if the value of the variable J is less than the threshold value γJ, the playback image of the image unit B (10) and the image unit It is determined that there is a cut between the reference images ahead in the reproduction time of B (10). If the value of the variable K is greater than or equal to a certain threshold value γK, it is determined that there is no cut between the reproduced image of the image unit B (10) and the reference image behind in the reproduction time of the image unit B (10), and the variable K Value is the threshold γK
If it is less than, it is determined that there is a cut between the reproduced image of the image unit B (10) and the reference image at the rear in the reproduction time of the image unit B (10).

Next, in step 6 of FIG. 8, it is judged whether the data has ended. Since the data has not ended,
The process after step 2 is performed, and the presence / absence of a cut is similarly determined for the subsequent bidirectionally predictive coded data.

In the above, in step 5 of FIG. 8, the presence or absence of the cut is determined using both the variable J and the variable K. However, the presence or absence of the cut is determined using either one of the variable J and the variable K. You may.

Also, in the above, an example of determining the presence / absence of a cut when the image unit is bidirectionally predictive coded has been described.
By using only the variable J, it is possible to determine the presence / absence of a cut when the image unit is forward predictive coded. In that case, by comparing the value of the variable J with the threshold value γJ, it is determined whether or not there is a cut between the reproduced image of the forward predictive encoded data and the forward reference image at the reproduction time.

As described above, according to the present embodiment, the intra-picture coded data obtained by coding the intra-picture information by the variable length coding and the reference picture with the picture located ahead in the reproduction time as the reference picture are used. A coded image data group IP composed of forward predictive coded data obtained by coding a difference value for each secondary divided area from the image of interest by variable length coding, or intra-coded data and forward predictive coded data And the difference value for each secondary divided region between the reference image and the target image is used as a reference image using the front image or the rear image or the interpolated images of both the front image and the rear image in the reproduction time by variable length encoding. A coded image data group IPB composed of the bidirectional predictive coded data is set as an input image data group, and a control signal for extracting the data amount of each image from the input image data group is set. And an image unit detection unit as a unit for generating the image data, and detects whether each image unit of the input image data group is encoded by intra-picture coding, forward prediction coding, or bidirectional prediction coding. And the coded image data in the input image data group is bidirectional predictive coded data, each secondary divided area refers to either the image ahead in the playback time or the image backward in the playback time. Whether or not each of the secondary divided areas has been encoded with reference to the preceding image at the reproduction time when the encoded image data in the input image data group is forward predictive encoded data. Is detected, the number of secondary divided areas coded with reference to a temporally previous image is substituted into a variable J, and the secondary divided area encoded with reference to a temporally posterior image is detected. Substitute the number into the variable K The variable J,
Alternatively, using the variable K, or both the variable J and the variable k, the presence or absence of a cut before and after the bidirectional predictive encoded data, or the presence or absence of a cut between the forward predictive encoded data and its reference image is determined. By including the means, the complete decoding work of the coded image data, which was necessary in the conventional example shown in FIG. 14, is not required, and thus the faster cut detection can be performed.

[0216]

As described above, according to the present invention, a group of coded image data composed of coded image data coded by variable length coding is used as an input image data group, and each of the input image data groups is selected. A control unit that generates a control signal for extracting an image data amount, the input image data group and the control signal from the control unit as inputs, and the input image data group according to the control signal from the control unit A counter that counts a part or all of the amount of data, a storage unit that stores at least one value of the counter, and a value stored in the storage unit and a preset threshold value, or the storage unit Comparison for determining the presence / absence of a cut between image data in the input image data group by comparing a value obtained by comparing and calculating a plurality of values stored in With the structure having a calculation unit, a recording medium was necessary in the conventional example shown in FIG. 14 1
Since the work of decoding the coded image data read out from is unnecessary and the work of counting the data amount for extracting the data amount of each image is performed instead, it is possible to detect the cut in a shorter time.

Further, unlike the conventional example, a large-scale image data memory for storing the decoded image data is not required, and only the value of the data amount need be stored instead. It enables downsizing and cost reduction. Furthermore,
In the conventional example, the comparison operation unit 5 performs the comparison operation process for detecting the cut using a large amount of image data, whereas the comparison operation unit 5 only needs to compare the values of a small amount of data. The time required for the processing in 5 can be greatly reduced. In summary, it is possible to speed up the process, reduce the size of the device, and reduce the cost of the entire device, as compared with the conventional example.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a coded image cut detection device according to a first embodiment of the present invention.

FIG. 2 is a configuration diagram of a coded image cut detection device according to a fourth embodiment of the present invention.

FIG. 3 is a configuration diagram of a coded image cut detection device according to a fifth embodiment of the present invention.

FIG. 4 is a configuration diagram of a coded image cut detection device according to a sixth embodiment of the present invention.

FIG. 5 is a configuration diagram of a coded image cut detection device according to a seventh embodiment of the present invention.

FIG. 6 is a configuration diagram of a coded image cut detection device according to an eighth embodiment of the present invention.

FIG. 7 is a flowchart of a coded image cut detection method according to a ninth embodiment of the present invention.

FIG. 8 is a flowchart of a coded image cut detection method according to a twelfth embodiment of the present invention.

FIG. 9 is a configuration diagram of encoded image data for explaining an encoded image cut detection device of the present invention.

FIG. 10 is a configuration diagram of divided areas in an image.

FIG. 11 is a graph showing an example of a difference value between intra-image encoded data in an encoded image data group.

FIG. 12 is a graph showing an example of values of forward predictive encoded data in an encoded image data group.

FIG. 13 is a graph showing an example of a difference value between forward predictive coded data in a coded image data group.

FIG. 14 is a configuration diagram of a conventional encoded image cut detection device.

FIG. 15 is a configuration diagram of coded image data for explaining a conventional coded image cut detection device.

FIG. 16 is a configuration diagram of a coded image cut detection device according to a second embodiment of the present invention.

FIG. 17 is a configuration diagram of a coded image cut detection device according to a third embodiment of the present invention.

FIG. 18 is a flowchart of a coded image cut detection method according to the tenth embodiment of the present invention.

FIG. 19 is a flowchart of a coded image cut detection method according to an eleventh embodiment of the present invention.

[Explanation of symbols]

 1 recording medium 2 counter 3 control unit 3A image unit detection unit 3B area detection unit 3C image coding type detection unit 3D image address memory 3E read position control unit 3F address counter 4 memory unit 5 comparison calculation unit 6 reference image determination unit 7 cut Judgment unit 8 Decoding unit 9 Quantization parameter detection unit

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Ikuji Shimizu 1006 Kadoma, Kadoma City, Osaka Prefecture Matsushita Electric Industrial Co., Ltd.

Claims (14)

[Claims] 1. An input image data group is a coded image data group composed of coded data obtained by coding each image by variable length coding, with one frame or one field as an image unit. A control unit for generating a control signal for extracting the data amount of each image from the group, and a control signal from the input image data group and the control unit as an input, and according to the control signal from the control unit. A counter that counts a part or all of the data amount of the input image data group, a storage unit that stores at least one value of the counter, a value stored in the storage unit, or a plurality of values stored in the storage unit. A comparison calculation unit that determines the presence or absence of a cut between encoded data in the input image data group by comparing a value obtained by performing a comparison calculation of the And a coded image cut detecting device. 2. One frame or one field is a unit of an image, and an encoded image data group formed from encoded data obtained by encoding each image is an input image data group. A control unit that generates a control signal for detecting the quantization parameter of each image from the inside, the control signal from the input image data group and the control unit as an input, and in accordance with the control signal from the control unit A quantization parameter detection unit that detects a quantization parameter of a part or all of the input image data group, a storage unit that stores at least one value of the quantization parameter, a value stored in the storage unit or the storage Between image coded data in the input image data group by comparing a value obtained by comparing and calculating a plurality of values stored in the unit with a preset threshold value. A coded image cut detection device including a comparison calculation unit that determines the presence or absence of a cut. 3. A coded image composed of coded data in which one frame or one field is used as a unit of image and each image is quantized or variable length coded or quantized and variable length coded. A control unit that generates a control signal for extracting the quantization parameter and the data amount of each image from the input image data group, and the input image data group and the control. A counter that receives a control signal from the control unit and counts a part or all of the data amount of the input image data group according to the control signal from the control unit, and a quantum of all or part of the input image data group. A quantization parameter detecting unit that detects a quantization parameter; a storage unit that calculates the counter value and the quantization parameter value and stores at least one of the values; Of the cut values between the image coded data in the input image data group by comparing a value obtained by comparing and calculating a plurality of values stored in the storage unit with a preset threshold value. A coded image cut detection device, comprising: a comparison calculation unit for determining the presence or absence. 4. The encoded image cut detection device according to claim 1, wherein the control unit includes an image unit detection unit that detects each image unit from the input image data group. 5. An image data unit of an input image data group is arbitrarily divided to form a primary divided area, all the primary divided areas are arbitrarily divided to form a secondary divided area, and the control unit The coded image cut detection device according to any one of claims 1 to 4, further comprising an area detection unit that detects each of the primary divided areas or the divided areas of the secondary divided areas. 6. A group I of encoded image data obtained by variable length encoding or quantization of each image using only information in the image, or encoding by variable length encoding and quantization, or intra-image encoded data. And a difference value for each divided region between the reference image and the image of interest is encoded by variable length encoding or quantization or variable length encoding and quantization using an image located ahead in the reproduction time as a reference image. Encoded image data group I composed of forward predictive encoded data
P, or intra-picture coded data and forward predictive coded data, and a divided area between the reference image and the image of interest with reference to the front image or the rear image or the interpolated images of both the front and rear images at the reproduction time. Variable length coding for each difference value, or
Any one of the coded image data group IPB composed of the quantized or variable length coded and the bidirectional predictive coded data coded by the quantized is used as the input image data group, and the control unit controls the input image An image coding type detection unit for detecting whether each image unit of the data group is coded by intra-picture coding, forward prediction coding, or bidirectional prediction coding is provided, and the comparison calculation unit Intra-image coding in the input image data group by comparing a value stored in a storage unit or a value obtained by comparing and calculating a plurality of values stored in the storage unit with a preset threshold value A structure for determining whether or not there is a cut between data, between forward predictive encoded data and its reference image, or between forward predictive encoded data.
The encoded image cut detection device according to claim 5. 7. The input image data group and the output of the image coding type detection unit, or the output of the input image data group and the image coding type detection unit and the output of the comparison operation unit When the coded image data is bi-directional predictive coded data, it is detected which of the forward image and the backward image in the reproduction time each divided area is coded, and the input When the coded image data in the image data group is forward predictive coded data, it is detected whether each divided area is coded by referring to the preceding image at the reproduction time and referring to the preceding image at the reproduction time. The number J of divided areas coded as a reference image, or the number K of divided areas coded with a backward image in the reproduction time as a reference image, or a reference image determination unit that outputs both J and K. Whether or not there is a cut before or after the bidirectional predictive coded data with the output of the reference image determining section or both the output of the reference image determining section and the output of the comparison calculating section as input, or forward predictive encoded data and its reference The coded image cut detection device according to claim 6, further comprising a cut determination unit that determines whether or not there is a cut between images. 8. A recording medium for recording a group of input image data, wherein a control unit controls a read position of the recording medium, a read position control unit, a storage unit of the read position of the recording medium, or the storage unit. The coded image cut detection apparatus according to claim 4, further comprising both a read position calculation unit and a read position calculation unit of the recording medium. 9. The control unit includes a coding unit detection unit that detects whether each coded image data in the input image data group has been coded in a frame or field image unit. The coded image cut detection device according to claim 8. 10. An intra-picture code in which one frame or one field is used as an image unit, and each picture is coded by variable-length coding or quantization or variable-length coding and quantization using only intra-picture information. The difference value for each divided area between the reference image and the image that is located ahead of the encoded data group I in the reproduction time is variable-length coded or quantized, or variable-length coded and quantized. A coded image data group IP composed of forward predictive coded data encoded by, or intra-coded data and forward predictive coded data, and a forward image or a backward image or a forward and backward image at the reproduction time. Using the interpolated image of both images as the reference image, the difference value for each divided area between the reference image and the image of interest is variable-length coded, or quantized, or both are coded by variable-length coding and quantized. Any one of the coded image data group IPB composed of predictive coded data is used as an input image data group, an image unit detection unit that detects each image unit of the input image data group, and each of the input image data group An image coding type detection unit that detects whether the image unit is coded by intra-picture coding, forward predictive coding, or bidirectional predictive coding, the input image data group, and the image coding When the encoded image data in the input image data group is bidirectional predictive encoded data with the output of the type detection unit as an input, each divided area is either a forward image at the reproduction time or a backward image at the reproduction time. If the encoded image data in the input image data group is forward predictive encoded data, each divided area is detected at the reproduction time. It is detected by referring to one of the images, and the number of divided areas J encoded by referring to the image preceding in time, or encoded by referring to the image posterior in time. A reference image determination unit that outputs the number K of divided regions, or both J and K, and whether or not there is a cut before and after bidirectional predictive encoded data, or forward prediction, using the output of the reference image determination unit as an input. An encoded image cut detection device comprising: a cut determination unit that determines whether or not there is a cut between encoded data and its reference image. 11. An input image data group is an encoded image data group composed of encoded data obtained by encoding each image by variable length encoding, with one frame or one field as an image unit. Generating a control signal for extracting the data amount of each image from the group, and counting a part or all of the data amount of the input image data group according to the control signal, and setting the value to at least one. One of the input image data group by substituting one into a variable group, the value of the variable group or a value obtained by comparing and operating a plurality of values of the variable group, and a preset threshold value. And a step of determining whether or not there is a cut between image coded data in the coded image cut detection method. 12. A coded image data group composed of coded data obtained by coding each image with one frame or one field as an image unit is set as an input image data group. Generating a control signal for detecting the quantization parameter of each image from the inside, and detecting at least one of the quantization parameters of a part or all of the input image data group according to the control signal and setting the value to at least one. In the input image data group by comparing the value of the variable group or a value obtained by comparing and calculating a plurality of values of the variable group with a preset threshold value. And a step of determining whether or not there is a cut between the image coded data of (1). 13. An input image data group is an encoding data group in which one frame or one field is used as an image unit and each image is quantized or variable-length coded or quantized and variable-length coded. And a step of generating a control signal for generating a control signal for detecting a quantization parameter of each image and extracting a data amount from the input image data group, and the input according to the control signal from the control. Counting a part or all of the data amount of the image data group, detecting a part or all of the quantization parameter of the input image data group,
A step of calculating the counter value and the quantization parameter value, substituting the value into at least one variable group, and a value obtained by comparing and calculating the value of the variable group or a plurality of values of the variable group, A method for detecting a coded image cut, comprising the step of determining whether or not there is a cut between the coded image data in the input image data group by comparing with a preset threshold value. 14. An intra-picture code in which one frame or one field is used as an image unit, and each picture is coded by variable-length coding or quantization or variable-length coding and quantization using only intra-picture information. The difference value for each divided area between the reference image and the image that is located ahead of the encoded data group I in the reproduction time is variable-length coded or quantized, or variable-length coded and quantized. A coded image data group IP composed of forward predictive coded data encoded by, or intra-coded data and forward predictive coded data, and a forward image or a backward image or a forward and backward image at the reproduction time. Using the interpolated image of both images as the reference image, the difference value for each divided area between the reference image and the image of interest is variable-length coded, or quantized, or both are coded by variable-length coding and quantized. One of the coded image data group IPB composed of predictive coded data is set as an input image data group, and a step of detecting each image unit of the input image data group, and each image unit of the input image data group are A step of detecting which of intra-picture coding, forward predictive coding, or bidirectional predictive coding is used, and the coded image data in the input image data group is bidirectional predictive coded data. In the case of, it is detected whether each divided area is coded by referring to a front image in the reproduction time or a rear image in the reproduction time, and the encoded image data in the input image data group is forward. In the case of predictive coded data, it is detected whether or not each divided area is coded by referring to the preceding image at the playback time, and the segment coded by referring to the temporally preceding image is detected. The number of divided areas is substituted into a variable J, the number of divided areas coded with reference to a temporally backward image is substituted into a variable K, and the variable J or the variable K or the variable J and A code for determining whether or not there is a cut before or after the bidirectional predictive encoded data, or whether or not there is a cut between the forward predictive encoded data and its reference image using both of the variables K. Image cut detection method.