JPH0799651A - Image signal reproducing device - Google Patents

Abstract

PURPOSE:To obtain a satisfactory interpolated image quality by obtaining correlations between a picture including an error block and a picture different from it by time and selecting image information to be used for motion vector estimation and interpolation and preventing a picture from being interpolated with a non-correlative picture by the scene change or the like. CONSTITUTION:An error detecting circuit 103 generates a block error flag indicating whether each picture element block can be decoded or not based on the error flag from an error correction ECC circuit 102. A scene change detecting circuit 122 uses image data in field memories 108 and 109 to discriminates whether the scene change occurs or not. In the case of the scene change, the motion vector of image data in the present field to image data of the next field is obtained by a motion vector detecting circuit 110, and the motion vector of the error block is estimated based on this obtained motion vector. This motion vector of the error block is used to interpolate picture data of the error block with picture data of the next field.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image signal reproducing apparatus for reproducing an image signal by decoding an image signal compressed by block coding to reproduce the image signal.

[0002]

2. Description of the Related Art An orthogonal transform coding system is known as a technique for compressing and coding an image signal with high efficiency. this is,
After the image signal is grouped into blocks for each predetermined number of pixels, orthogonal transform such as discrete cosine transform (DCT) is performed,
Quantization, entropy coding, etc. are performed on the coefficient after conversion.

FIG. 4 shows a digital VTR as an image signal reproducing apparatus using such block coding and decoding.
3 is a block diagram showing the configuration of the main part of FIG. Here, the flow of the image signal in FIG. 4 will be described. Input terminal 401
A more digitized image is input and is blocked by the blocking circuit 402 in units of m pixels × n pixels. The image data blocked in units of m × n pixels is
The DCT circuit 403 performs orthogonal transformation, and transforms the spatial domain into the frequency domain. The data converted into the frequency domain is quantized by the quantization circuit 404, and further the variable length coding circuit 4
The desired data transfer rate is obtained by encoding with 05. The encoded data is converted into a signal of a format suitable for recording by the recording / reproducing system 406 and recorded on the tape.

At the time of reproduction, the data reproduced from the tape by the recording / reproduction system 406 is converted into a signal of a format suitable for the signal processing in the subsequent stage, and an error correction circuit (ECC circuit) 4
After the code error is corrected by 07, it is decoded by the variable length code decoding circuit 408. The decoded data is inversely quantized by the inverse quantization circuit 409, then inverse DCT is performed by the inverse DCT circuit 410, converted from frequency domain data to spatial domain data, and written in the frame memory 411.
The image data written in the frame memory 411 is read out in accordance with scanning by a monitor or the like, the error correction impossible data (hereinafter referred to as error data) is interpolated by the interpolation circuit 412, and then output from the output terminal 413 to be monitored. Etc. are displayed.

In such an encoding method, encoding is performed in block units. Therefore, if an uncorrectable error occurs in the encoded data due to error correction during decoding, a decoding error occurs in block units. In addition, since a variable length code is often used, a decoding error propagates over a plurality of blocks, resulting in a large deterioration in image quality.

As means for interpolating and correcting a block containing such error data (hereinafter referred to as error block), conventionally, inter-frame interpolation for replacing an error block with an image of a previous frame, intra-field interpolation for interpolating an error block with an image of the same field Etc. are known. Further, while each of the above methods was selected only with or without the motion of the error block, it is also possible to estimate the motion vector of the error block and perform inter-field interpolation using this motion vector. ing.

[0007]

However, in the interpolation method using the motion vector as described above, normally good interpolation can be performed by using the image data of the previous field, but an image completely different between fields due to a scene change or the like. In this case, since there is no data that can be used for interpolation in the image of the previous field, good image quality cannot be obtained.

In view of the above problems, the present invention obtains a correlation between an image including an error block and an image temporally different from each other, and selects image information to be used for motion vector estimation and interpolation. Therefore, it is an object of the present invention to provide an image signal reproducing apparatus capable of preventing interpolation with an image having no correlation due to the above, and capable of obtaining good interpolation image quality.

[0009]

SUMMARY OF THE INVENTION In order to solve the conventional problems and achieve the above object, the present invention decodes image data in block units for each predetermined number of pixels. An image signal reproducing apparatus for obtaining an image signal, the correlation means for detecting a correlation value between an image including an undecodable block and an image temporally different from each other, and the undecodable based on the detection result of the correlation means. And a construction means for constructing the data of the undecodable block by using the data of the image temporally different from the block.

[0010]

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

FIG. 1 is a block diagram showing the configuration of an image signal reproducing apparatus as a first embodiment of the present invention. Less than,
The processing of the image signal in FIG. 1 will be described. In addition,
In this embodiment, when the error block is interpolated,
Based on the motion vector of the error block, it is assumed that the image of another field is used for interpolation. Further, in the present embodiment, for the sake of explanation, it is assumed that the reproduction signal is reproduced and decoded in frame units.

In FIG. 1, an image signal reproduced from a recording medium such as a magnetic tape is input from an input terminal 101, and E
It is sent to the CC circuit 102. Then, the ECC circuit 102
In the data separation circuit 104
Output to. Further, the ECC circuit 102 outputs an error flag indicating that the error cannot be corrected to the error detection circuit 103 for the data in which the code error cannot be corrected. The error detection circuit 103 generates a block error flag indicating whether or not each pixel block can be decoded based on the error flag from the ECC circuit 102.

The data separation circuit 104 is the ECC circuit 102.
Data is separated from the block address data indicating the position of the block on the screen and the image data, and the memory control circuit 106 and the block decoding circuit 105 are separated.
Output to each. The block decoding circuit 105 decodes image data in pixel block units, and the frame memory 1
It outputs to 07. At this time, the memory control circuit 106 determines the write address of the decoded image data of each pixel block to the frame memory 107 based on the block address data from the data separation circuit 104.

The frame memory 107 is the first of each frame.
Field memory 1 in the order of field and second field
The image data is output to 21. Field memory 108
The image data of the previous field of the image data stored in the field memory 121 is stored in, and the image data stored in the field memory 108 is interpolated. Further, the field memory 109 stores image data one field before the image data stored in the field memory 108, that is, one field before the image to be interpolated.

The scene change detection circuit 122 uses the image data in the field memories 108 and 109 to determine whether or not there is a scene change. As an example of the determination method, there is a method of calculating the absolute difference value of each pixel between fields and determining whether or not the total exceeds a threshold value. Such a method is also acceptable. Delay circuits 123, 124, 12
In No. 5, the delay time is set so that each image signal is delayed until the scene change detection circuit 122 outputs the determination result. The selection switch 126 receives the output signal of the scene change detection circuit 122, and when the scene change has not occurred, the image data of the previous field (the image data in the field memory 109), and when the scene change has occurred, the next field of the next field is detected. Image data (image data in the field memory 121) is selected.

The motion vector detection circuit 110 calculates an inter-field motion vector in pixel block units from the image data of the field to be interpolated and the image data of the field selected by the selection switch 126 (the fields before and after the field to be interpolated). It is detected and output to the interpolation circuit 111. For detecting the motion vector, various detection methods such as a matching method for matching the image data of the current field and the image data of the previous field can be used.

The interpolation circuit 111 estimates the motion vector of the error block by using the motion vector of the block around the error block detected by the motion vector detection circuit 110, and then uses the estimated motion vector of the error block in the field memory. The addresses to be read in 109 and 121 are determined, and the addresses are output to the selection switch 127. The selection switches 127 and 128 interlock with the output signal of the scene change detection circuit 122 to perform selection. When the scene change has not occurred, the field memory 109 side is selected, and when the scene change has occurred, the field memory 121 side is selected. The selected field memory receives the address data from the selection switch 127 and outputs the image data of the corresponding address to the selection switch 128. The selection switch 128 outputs the image data to the interpolation circuit 111, and the interpolation circuit 111 outputs the address data of the block to be interpolated and the interpolation image data to the field memory 108. Field memory 10
Reference numeral 8 receives the address data and the interpolated image data from the interpolation circuit 111, and rewrites the image data by writing the image data in the corresponding address to perform the interpolation.

This situation will be described with reference to FIG. Figure 2
FIG. 2B shows a case where there is no scene change.
Indicates a scene change between the previous field and the current field.

When there is no scene change, FIG.
As shown in (a), the motion vector of the image data of the current field with respect to the image data of the previous field is obtained, the motion vector of the error block is estimated using this motion vector, and interpolation is performed.

Further, when there is a scene change, it is considered that there is almost no correlation between the image of the previous field and the image of the current field, so that interpolation cannot be performed using the image data of the previous field. Therefore, the motion vector of the image data of the current field with respect to the image data of the next field is obtained, and the motion vector of the error block is estimated from this motion vector. Then, using the motion vector of this error block, the image data of the error block is interpolated by the image data of the next field.

When the interpolation processing of the image stored in the field memory 108 is completed in this way, the data of the field memory 109 is output from the output terminal 112 to an external device. Further, the data in the field memory 108 is written in the field memory 109, and the field memory 12
1 receives the next field data from the frame memory 107. By repeating the above operation, the field memory 1
If there is error data in the image stored in 08, interpolation processing is similarly performed.

As described above, a good interpolated image can be always obtained by adaptively selecting the image data used for the motion vector detection and the interpolation from the preceding and following fields by detecting the scene change. This is especially effective when there is a scene change immediately before the field containing the error block.

Next, a second embodiment of the present invention will be described with reference to FIG. It should be noted that members having the same or similar effects as in FIG.

In FIG. 3, the image data separated by the data separation circuit 104 is the block decoding circuit 105.
And to the delay circuit 131 and the scene change detection circuit 132.

The delay circuit 131 includes a data separation circuit 104.
The delay time is set so that the image data before conversion from the frequency domain to the spatial domain is input, and the image data of one frame before the image data output by the data separation circuit 104 is output. Scene change detection circuit 1
32 receives the image data for each block from the data separation circuit 104 and the delay circuit 131, and determines whether or not there is a scene change between frames. As an example of the determination method, there is a method of calculating the absolute difference value of the DC component coefficient of each block between frames and determining whether the total exceeds a threshold value. That is, in this embodiment, DC
Since only the difference absolute value of the components needs to be calculated, the above-mentioned first
The calculation amount is smaller and the circuit scale is smaller than that of the embodiment.

The selection switches 134 and 135 operate in conjunction with each other, and when the scene change determination result is received from the delay circuit 133 having a delay time of 3 fields + motion vector determination time, and no scene change occurs, The image data of the previous field (image data stored in the field memory 109) and the image data of the next field (image data stored in the field memory 121) are selected when a scene change occurs.

The motion vector detection circuit 110, similar to the above-described embodiment, includes image data of fields to be interpolated,
The inter-field motion vector is detected in pixel block units from the image data of the field selected by the selection switch 135 (the field selected from the fields before and after the field to be interpolated) and output to the interpolation circuit 111. The interpolation circuit 111 estimates the motion vector of the error block from the motion vector of the block around the error block, and based on this motion vector, the field memory 1
09, 121 to determine the address to be read,
The address is output to the selection switch 134.

The selection switch 134 selects the field memory 108 side when the scene change has not occurred, and selects the field memory 121 side when the scene change has occurred. The selected field memory receives the address data from the selection switch 134 and outputs the image data of the corresponding address to the selection switch 135. The selection switch 135 outputs the image data to the interpolation circuit 111, and the interpolation circuit 111 outputs the address data of the block to be interpolated and the interpolation image data to the field memory 108.
The field memory 108 receives the address data and the interpolated image data from the interpolation circuit 111, rewrites the image data of the error block, and performs the interpolation.

As described above, by detecting the scene change based on the image data before decoding and adaptively selecting the image data used for the motion vector detection and the interpolation, it is possible to always obtain a good interpolated image.

In the above embodiment, the correlation between the image including the error block and the image one field before is calculated, and based on this correlation, it is detected whether or not there is a scene change, and interpolation is performed. However, the image data used for interpolation may be selected simply based on the detected correlation. That is, the correlation between the image including the error block and the image one field after is detected, and when the correlation is lower than a predetermined level, the error block can be interpolated using the image data one field before.

Further, by detecting both the correlation with one field before and after one field of the error block, and interpolating the error block using the data of the image with the higher correlation, it is possible to be influenced by a scene change or the like. Instead, the image data having the first correlation can always be used for interpolation.

[0032]

As described above, according to the present invention, the correlation between the image containing the undecodable block and the image temporally different from each other is obtained, and the data of the undecodable block is constructed based on the correlation result. There is.

Therefore, when the data of the undecodable block is interpolated using the data of the surrounding blocks, for example,
Of the images before and after the image including the undecodable block, the image data having more correlation can be used for interpolation, and a good reproduced image can be obtained for the image missing in block units.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an image signal reproducing apparatus as a first embodiment of the present invention.

FIG. 2 is a diagram for explaining an interpolation operation in the embodiment of the present invention.

FIG. 3 is a block diagram showing a configuration of an image signal reproducing device as a second embodiment of the present invention.

FIG. 4 is a block diagram showing a configuration of a digital VTR.

[Explanation of symbols]

 110 Motion Vector Detection Circuit 111 Interpolation Circuit 122 Scene Change Detection Circuit

Claims (4)

[Claims] 1. An image signal reproducing apparatus which obtains an image signal by decoding data obtained by block-encoding an image signal for each predetermined number of pixels, the image signal including an undecodable block and its temporal A correlating means for detecting a correlation value with a different image, and a constructing means for constructing data of the undecodable block using data of an image temporally different from the undecodable block based on the detection result of the correlating means. An image signal reproducing apparatus comprising: 2. If the correlation value between an image including the undecodable block and an image temporally preceding the image is smaller than a predetermined value, the constructing means takes longer than the image including the undecodable block. 2. The image signal reproducing apparatus according to claim 1, wherein the data of the undecodable block is constructed by the data of the image that is afterward. 3. The constructing means, if the correlation value between the image including the undecodable block and the image temporally subsequent thereto is larger than a predetermined value, the constructing means takes a time longer than that of the image including the undecodable block. The image signal reproducing apparatus according to claim 1, wherein the data of the undecodable block is constructed by the data of the previous image. 4. The construction means compares the correlation value between an image including the undecodable block and a temporally previous image and a temporally subsequent image, and uses the image data having a larger correlation value to perform the decoding. The image signal reproducing apparatus according to claim 1, wherein the image data of the impossible block is constructed.