JPH08317347A - Image information converting device - Google Patents

Abstract

PURPOSE: To convert the video signal of an NTSC system to the video signal of a high-definition system. CONSTITUTION: HD data are turned to SD data through a vertical thinning filter 22 and a horizontal thinning filter 23 and supplied to an SD contour correcting circuit 24. The contour corrected SD data are supplied from the SD contour correcting circuit 24 to area segmenting circuits 26, 28 and 32. At the area segmenting circuit 26, a spatial class is formed through an ADRC circuit 27 and at the area segmenting circuit 28, a motion class is formed through a motion class deciding circuit 29. Those classes are respectively supplied to a class code generating circuit 30. At the class code generating circuit 30, a class code is generated and supplied to a regular equation adder circuit 33. At the normalized equation adder circuit 33, the SD data are supplied through a delay circuit 31 and the area segmenting circuit 32, the HD data are supplied through an HD contour correcting circuit 25, and a normalized equation is generated. When a predictive coefficient is decided at a predictive coefficient deciding circuit 34, it is stored in a memory 35.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image information conversion device suitable for use in, for example, a television receiver, a video tape recorder device, or the like. The present invention relates to an image information conversion device that converts and outputs the image.

[0002]

2. Description of the Related Art Nowadays, due to the increasing audio-visual orientation, it is desired to develop a television receiver capable of obtaining a higher resolution image, and in response to this demand, a so-called high vision has been developed. . The number of scanning lines defined in the so-called NTSC system is 525 in this high-definition system, which is more than doubled to 1125 lines, and the aspect ratio of the display screen is 9: 3 in the NTSC system 3: 4. 16 and wide-angle screen. Therefore, it is possible to obtain a high-resolution and realistic screen.

Here, although it is a high-definition television having such excellent characteristics, it is not possible to display an image even if an NTSC video signal is supplied as it is. This is because the standards are different between the NTSC system and the high-definition system as described above. For this reason, when an image corresponding to an NTSC video signal is to be displayed in high-definition, the video signal rate conversion has conventionally been performed using an image information conversion device as shown in FIG. 9, for example.

Referring to FIG. 9, a conventional image information conversion apparatus includes a horizontal interpolation filter 42 for performing horizontal interpolation processing of an NTSC video signal (SD data) supplied through an input terminal 41, and a horizontal interpolation filter 42. Vertical interpolation filter 4 for performing vertical interpolation processing on the video signal for which interpolation processing has been performed
And 3. Then, a high-definition video signal (HD data) can be obtained from the output terminal 44.

Specifically, the horizontal interpolation filter 42 has a configuration as shown in FIG. 10, and the NTSC system video signal supplied through the input terminal 51 receives the input terminal 5.
1 to m-th multipliers 52-52 _m , respectively. Each multiplier 52 multiplies the video signal by a coefficient and outputs the result. The video signals multiplied by the coefficients are supplied to the _first to mth adders 54 to 54 _n-1 . Between each adder 54 to 54 _m-1 , the time T
Delay registers 53 to 53 _m are provided. The video signal output from the mth multiplier 52 _m is delayed by the time T by the mth delay register 53 _m ,
It is supplied to the _m- 1th adder 54 _m-1 .

[0006] Adder 54 _m-1 of the (m-1) is a video signal delay time of the time T from the delay register 53 _m of the m is applied, the multiplier 52 _m-1 of the (m-1) And the video signal of are added and output. The video signal on which this addition processing has been performed is again subjected to the delay time of time T by the (m−1) th delay register 53 _m−1 , and the same is applied in the (m−2) th adder 54 _m−2 not shown. The addition processing is performed with the video signal from the _m-th multiplier 54 _m-2 (not shown). The horizontal interpolation filter 42 thus supplies the NTSC video signal to the vertical interpolation filter 43 via the output terminal 55.

The vertical interpolation filter 43 has a structure similar to that of the horizontal interpolation filter 42 described above, and interpolates pixels in the vertical direction with respect to the video signal subjected to the horizontal interpolation processing. As a result, vertical pixel interpolation is performed on the NTSC video signal. The high-definition video signal thus converted is supplied to the high-definition receiver. As a result, an image corresponding to the NTSC video signal can be displayed on the high-definition receiver.

However, since the conventional image information converting apparatus merely interpolates in the horizontal and vertical directions based on the video signal of the NTSC system, the resolution is the video signal of the base NTSC system. It didn't change at all. In particular, when a normal moving image is to be converted, it is common to perform vertical interpolation by intra-field processing, but in that case, since the inter-image field correlation is not used, conversion is not performed in the image still part. Loss, NTSC
There is a drawback that the resolution is degraded rather than the video signal of the system.

On the other hand, in Japanese Patent Application No. 6-205934, three-dimensional (spatiotemporal) of the image signal level of the input signal.
The class is divided according to the distribution, and each class has a storage unit that stores the prediction coefficient value acquired by learning in advance.
There is a method of outputting an optimum estimated value by a calculation based on a prediction formula.

According to this method, when creating HD pixels, class division is performed using a plurality of SD pixel data in the vicinity of the HD pixel to be created, and a prediction coefficient value is acquired by learning for each class. The ingenuity is such that the HD pixel value closer to the true value is obtained by utilizing the intra-frame correlation in the image static portion and the intra-field correlation in the moving portion.

For example, for the purpose of creating HD pixels y ₁ and y ₂ as shown in FIG. 3, SD pixels m _{1 to} m ₅ and SD pixels n _{1 to} n ₅ as shown in FIG. 5, respectively. An average value of inter-frame differences between pixels located at the same spatial position is obtained, and threshold value processing is performed to classify the values, thereby classifying mainly the degree of motion. It should be noted that FIGS. 3 to 6 show that # k-2, # k-1,
The vertical pixel array of 4 fields of #k and # k + 1 is shown.

At the same time, as shown in FIG. 4, by subjecting the SD pixels k _{1 to} k ₅ to ADRC processing, class classification is performed with a small number of bits, mainly for the purpose of waveform representation in space. For each of the classes determined by these two types of classification, SD pixels x _{1 to} x ₉ as shown in FIG. 6 are used to form a linear linear expression to obtain a prediction coefficient value by learning. In this method, a class classification that mainly represents the degree of motion and a class classification that mainly represents the waveform in the space are individually performed in a suitable form, so that high conversion performance can be obtained with a relatively small number of classes. There is a feature called.

The estimation calculation of the HD pixel y is performed by the following equation using the prediction coefficient values w _{1 to} w _n obtained by the above procedure. y = w ₁ x ₁ + w ₂ x ₂ + ... + w _n x _n (1) In this example, n = 9.

As described above, the coefficient data for estimating the HD data corresponding to the SD data is obtained by learning in advance for each class and then stored in the ROM table.
By outputting the input SD data and the coefficient read from the ROM table, unlike the case where the input SD data is simply interpolated, it is possible to output data closer to the actual HD data. .

By the way, the learning in the conventional image information converting apparatus is performed by the structure shown in FIG. That is, the HD data input from the input terminal 61 is thinned by the vertical thinning filter circuit 62 so that the frequency in the vertical direction in the field becomes 1/2, and further, by the horizontal thinning filter circuit 63. The SD data can be obtained by performing the thinning-out processing so that the horizontal frequency of 2 becomes 1/2.

Then, the area cutout circuit 64 and the AD
The spatial class is obtained from the RC circuit 65, and the motion class is obtained from the area cutout circuit 66 and the motion class determination circuit 67. A class code is generated from the class code generating circuit 68 according to the space class and the motion class, and the normal equation adding circuit 71 normalizes the SD code and the HD data from the class code, the delay circuit 69, and the area cutting circuit 70. The equation is generated. The prediction coefficient determination circuit 72 determines the prediction coefficient of each class, and the prediction coefficient is stored in the memory 73.

Further, the output signal of the television camera has a lower output as the spatial frequency component is higher due to the influence of the resolution characteristic of the lens, the aperture characteristic of the electron beam of the image pickup tube and the like. A contour corrector (also known as an enhancer) is provided to correct this and send out by emphasizing the contour of the video signal in anticipation of a drop in the high frequency range of the spatial frequency in the display device, creating a sharp image. Has been.

In this contour corrector, the processing shown in FIG. 8 is generally performed. A correction signal is generated from, for example, a G signal of three primary colors of R (red), G (green), and B (blue). The delay circuits 85 and 86 delay the R signal and the B signal by the time required for processing the correction signal. The delay circuit 87 and the correction signal processing circuit 88 basically perform processing of a high-pass filter that cuts out the high frequency component of the G signal, and the high frequency component is used as a correction signal for the attenuators 92 and 93.
The level is adjusted by 94 to adders 89, 90,
At 91, the color signals are superimposed. Note that contour correction is performed in both the horizontal and vertical directions,
In the delay circuit 87, the G signal is delayed by one sample when the contour is corrected in the horizontal direction, and is delayed by one line when the contour is corrected in the vertical direction.

The correction signal processing circuit 88 specifically includes many non-linear processes such as coring and non-linear processes for contour signals. Coring is a process for suppressing a small contour signal by regarding it as a noise component. The non-linear processing for the contour signal is generally a method in which the gain of the small amplitude correction signal is large and the gain of the large amplitude correction signal is small. In addition, the gain of the correction signal may be changed depending on the brightness of the image, and in the case of SD data, the correction signal may be controlled in consideration of NTSC encoding.

That is, in the conventional image signal converting apparatus, the image signal of the HD data, which has been subjected to the contour correction as described above, and the image signal of the SD data, which is created by down-converting the HD data by the ideal filter, are used. I was learning between. Therefore, the obtained SD data is the SD data obtained by down-converting the image signal of the HD data on which the HD contour correction has been performed, and the characteristics of the data near the contour are SD captured by the SD camera.
It was different from the data. Therefore, in the conventional image signal conversion device, ideal conversion can be realized for up-conversion of SD data obtained by down-converting HD data, but for tap conversion of SD data captured by an SD camera. However, there is a drawback that the conversion is not always ideal especially near the contour.

[0021]

Therefore, the present invention has been made in view of the above-mentioned problems, and it is possible to improve the resolution and convert the video signal of the NTSC system into the video signal of the high vision system. An object is to provide such an image information conversion device.

[0022]

According to a first aspect of the present invention, in an image information converting apparatus for converting a digital image signal into a digital image signal having a larger number of pixels, image information supplied from the outside is supplied. An image cropping unit that crops image data at a predetermined position from the image, and a pattern of the level distribution of the image information extracted by the image cropping unit is detected, and based on the pattern, the class to which the image information belongs is determined and A class detecting means for outputting, a class code generating means for determining a class from the result of the class detecting means, and a coefficient of an estimation formula which is information for converting image information supplied from the outside into image information of higher resolution. Depending on the coefficient data storage means in which data is stored for each class and the coefficient data supplied from the coefficient data storage means, A picture information converting apparatus, comprising an image converting means for converting the high-resolution image information.

Further, in the invention described in claim 5, in an image information conversion device adapted to convert a digital image signal into a digital image signal having a larger number of pixels, standard image is converted from image information of higher resolution. A filter means for converting into information, a first contour correction means for performing contour correction on standard image information, a second contour correction means for performing contour correction on image information of higher resolution, and a standard Image cutout means for cutting out image data at a predetermined position from the image information, and a pattern of level distribution of the image information extracted by the image cutout means is detected, and the class to which the image information belongs is determined based on the pattern. A class detecting means for outputting detection information; a class code generating means for determining a class from the result of the class detecting means; standard image information; Coefficient data generating means for generating coefficient data of an estimation formula that is information for converting standard image information from higher resolution image information from the contour correction means to higher resolution image information, and coefficient data are The image information conversion device is characterized by comprising coefficient data storage means stored for each class.

[0024]

The image information converting apparatus according to the present invention is provided with an input SD
The pattern of the level distribution of SD pixels located near the HD pixel to be created is detected from the signal, and the class to which the image information of the area belongs is determined based on the detected pattern and the class detection information is output. . Further, the class representing the degree of motion is determined using the inter-frame difference of SD pixels located near the HD pixel to be created, and class detection information is output. The above two classes are integrated by a class code generation circuit and output as a final class. The coefficient data storage means stores the coefficient information of the linear estimation formula, which is information for converting the image information supplied from the outside to the image information having a resolution higher than this image information, for each class, This coefficient data is output according to the class detection information. Then, the image information conversion means converts the image information supplied from the outside into image information having a higher resolution than the image information supplied from the outside according to the coefficient data supplied from the coefficient data storage means.
As the learning target for obtaining the coefficient data of the linear estimation formula, the SD image data created by simply down-converting the HD image data is not used as in the conventional method, but the image is captured without the contour correction. It is characterized in that the HD image data is down-converted, and the image subjected to the contour correction by the contour correction circuit similar to the contour correction circuit used in the SD camera is used as the SD image data.

[0025]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is an embodiment showing a schematic configuration of signal processing of an image information conversion apparatus. As the image information supplied from the input terminal 1 shown in FIG. 1, a so-called NTSC system video signal is digitized and supplied as SD (Standerd Definition) data. SD pixel and H to be created in this embodiment
The positional relationship of the D pixels is as shown in FIG. That is, the HD pixel to be created is the HD pixel y ₁ existing at a position close to the SD pixel when viewed in the same field.
And y ₂ existing at a position far from the SD pixel. Hereinafter, a mode for estimating an HD pixel existing at a position close to the SD pixel will be referred to as a mode 1, and a mode for estimating an HD pixel existing at a position far from the SD pixel will be referred to as a mode 2.

The area cutout circuit 2 cuts out from the SD image signal supplied from the input terminal 1 pixels necessary for class classification (hereinafter referred to as space class) mainly for a waveform phenomenon in space. In this example, cut out HD pixel y _1, 5 single SD pixels k ₁ to k ₅ located in the vicinity of the y ₂ to create, for example, as shown in FIG. The SD data extracted by the area cutout circuit 2 is supplied to the ADRC circuit 3. In the ADRC circuit 3, for the purpose of patterning the level distribution of the SD data of the area, an operation of compressing the data of each area from 8-bit SD data to 2-bit SD data is performed. As a result, the formed pattern compressed data is transferred to the class code generation circuit 6
Supply to.

Originally, ADRC (Adaptive Dynamic Range)
Coding) is an adaptive quantization developed for high-efficiency coding of a digital VTR. However, since a local pattern of signal level can be efficiently expressed with a short word length, in this embodiment, ADRC is used as a signal. It is used to generate code for pattern classification. The ADRC circuit 3 sets the maximum value MAX and the minimum value in the area according to the following equation (2) where DR is the dynamic range in the area, n is the bit allocation, L is the data level of the pixels in the area, and Q is the requantization code. The value MIN and the value MIN are equally divided by a specified bit length and requantization is performed.

DR = MAX-MIN + 1 Q = [(L-MIN + 0.5) * ²ⁿ / DR] (2) However, [] means rounding down.

In this embodiment, it is assumed that the SD data of each 5 pixels separated by the area cutting circuit 2 is compressed into 2 bits. Let the compressed SD data be q _{1 to} q ₅ , respectively.

On the other hand, the SD image signal supplied from the input terminal 1 is also supplied to the area cutting circuit 4. The area cutout circuit 4 has a function of cutting out pixels required mainly for class classification (motion class) for expressing the degree of motion. In this embodiment, for example, from the supplied SD image signal, ten SD pixels m _{1 to} m ₅ and n _{1 to} n existing at the positions shown in FIG. 5 with respect to the HD pixels y ₁ and y ₂ to be created.
Extract ₅ . The data cut out by the area cutout circuit 4 is supplied to the motion class determination circuit 5. The motion class determination circuit 5 calculates the inter-frame difference of the supplied SD pixel data, and thresholds the average value of the absolute values to calculate a motion parameter that is a motion index. Specifically, the motion class determination circuit 5 uses the equation (3) to calculate the average value pa of the absolute values of the differences of the supplied SD data.
Calculate ram.

[0031]

[Equation 1] However, in this embodiment, n = 5.

The motion class mv-class is calculated using the average value param of the absolute values of the SD data differences by a preset threshold value of the average value param of the absolute values of the SD data differences calculated by the above method. To calculate. For example, here, four motion classes are provided, and the motion class mv-cla
Determine ss as follows:

When param ≦ 2: Motion class mv-class = 0 When param ≦ 4: Motion class mv-class = 1 When param ≦ 8: Motion class mv-class = 2 When param> 8: Motion class mv-class = 3 The motion class determined in this way is supplied to the class code generation circuit 6.

The class code generating circuit 6 receives the pattern compressed data (spatial class) supplied from the ADRC circuit 3 and the motion class mv supplied from the motion class determining circuit 5.
The class is detected by performing the calculation of the following expression (4) based on -class, and the class code class is supplied to the ROM table 9. This class code class is RO
It indicates the read address from the M table 9.

[0035]

[Equation 2] In this example, n is 5 and p is 2.

In the ROM table 9, coefficient data for calculating the HD data corresponding to the SD data is learned for each class by using the linear estimation formula by learning the relationship between the SD data pattern and the HD data. Remembered
This is the information for converting the SD data into the HD (High Definition) data, which is the image information having a higher resolution than this image information, which conforms to the so-called high-definition standard, by the linear estimation formula. In this embodiment, coefficient data is prepared independently for mode 1 and mode 2.
The method of creating the coefficient data stored in the ROM table 9 will be described later. From the ROM table 9, wi (class) which is the coefficient data of the class is read from the address indicated by the class code class.
This coefficient data is supplied to the estimation calculation circuit 10.

On the other hand, the input SD data is also supplied to the area cutout circuit 8. The area cutout circuit 8 delays the input SD data by the processing time of the area cutout circuit 4 and the motion class determination circuit 5.
From in the position as shown in FIG. 6, cut nine SD pixel data x ₁ ~x ₉ used for estimation operation. The output signal of the area cutout circuit 8 is supplied to the estimation calculation circuit 10. The estimation calculation circuit 10 calculates HD data corresponding to the input SD data based on the SD data supplied from the area cutout circuit 8 and the coefficient data supplied from the ROM data table 9.

More specifically, the estimation calculation circuit 10 uses the SD data x ₁ supplied from the area cutting circuit 8 as x ₁
The w ₁ to w ₉ is a coefficient data supplied from ~x ₉ and ROM table 9, using the coefficient for mode 1 for mode 1, with respect to mode 2 using the coefficients for the mode 2, the coefficient data The HD data corresponding to the input SD data is calculated by performing the calculation shown in the following equation (5) on the basis of wi (class).
The created HD data is supplied to the horizontal interpolation filter 11.

Hd ′ = w ₁ x ₁ + w ₂ x ₂ + w ₃ x ₃ + ... + w ₉ x ₉ (5)

The horizontal interpolation filter 11 is the same as the horizontal interpolation filter 43 of FIG. 9, and doubles the number of pixels in the horizontal direction by interpolation processing. Horizontal interpolation filter 1
The output of 1 is output via the output terminal 12. HD data output via the output terminal 12 is, for example, H
It is supplied to a D television receiver, an HD video tape recorder device, or the like.

As described above, the coefficient data for estimating the HD data corresponding to the SD data is obtained for each class in advance by learning and then stored in the ROM table 9, and the SD data and the ROM to be input are input. By performing an operation based on the coefficient data read from the table 9 and forming and outputting HD data corresponding to the input SD data, the actual SD data is different from the one obtained by simply interpolating the input SD data. Data closer to HD data can be output.

Next, a method of creating coefficient data stored in the ROM table 9 will be described with reference to FIG. The method of creating the coefficient data, more specifically, the method of creating HD data and SD data used for learning is different from the conventional method. In order to obtain the above-mentioned coefficient data by learning, first, an SD image having a pixel number ¼ of the HD image corresponding to the already known HD image is formed.

First, HD data is supplied through the input terminal 21, and the HD data is an image taken by turning off the contour correction circuit of the HD camera. For the supplied HD data that has not been subjected to contour correction, the pixels in the vertical direction of the supplied HD data are filtered by the vertical thinning filter 22.
By this, the thinning processing is performed so that the vertical frequency in the field is reduced to 1/2, and the horizontal thinning filter 2
3, the pixels in the horizontal direction of the HD data are thinned out. Further, an SD contour correction circuit 24 similar to the contour correction circuit used in a standard SD camera performs horizontal and vertical contour corrections to create SD data. The SD data used in the conventional learning is created by down-converting the HD data whose contour is corrected by the HD camera, whereas the SD data of this embodiment is the HD data which is not contour-corrected. Since it is down-converted and contour correction is performed by a contour correction circuit similar to that used in the SD camera, compared to the conventional SD data, the characteristics in the vicinity of the contour are particularly S.
It is close to the so-called generally-known SD data that is imaged and contour-corrected by the D camera.

The SD data created in this manner is used for the area cutting circuit 26, the area cutting circuit 28, and the delay circuit 3.
1 is supplied. On the other hand, the HD supplied to the input terminal 21
The data (HD data that has not undergone contour correction) is contour-corrected by an HD contour correction circuit 25 similar to the contour correction circuit used in a standard HD camera, and then supplied to a normal equation addition circuit. . The area cutout circuit 26 cuts out necessary pixels from the supplied SD image signal in order to perform the spatial class classification. Specifically, the area cutout circuit 26 has the same function as the area cutout circuit 2 described above. The cut SD data is supplied to the ADRC circuit 27.

The ADRC circuit 27 detects a one-dimensional or two-dimensional level distribution pattern of the SD data supplied for each area and, as described above, all or a part of the data in each area. , For example, 8-bit S
Pattern compression data is formed by performing an operation for compressing D data into 2-bit SD data, and this pattern compression data is supplied to the class code generation circuit 30. The ADRC circuit 27 is the same as the ADRC circuit 3 described above.

On the other hand, the SD image signal supplied to the area cutout circuit 28 is subjected to data cutout necessary for motion class classification. Specifically, the area cutout circuit 2
Reference numeral 8 has the same function as that of the area cutout circuit 4 described above. The SD data cut out by the area cutout circuit 28 is supplied to the motion class determination circuit 29. The motion class determination circuit 29 specifically has the same function as the motion class determination circuit 5 described above. The motion class determined by the motion class determination circuit 29 is supplied to the class code generation circuit 30. The class code generation circuit 30 is the same as the class code generation circuit 6 described above. The class code generation circuit 30 uses the pattern compressed data (spatial class) supplied from the ADRC circuit 27 and the motion class supplied from the motion class determination circuit 29. Based on this, the class is detected by performing the calculation of the above-mentioned formula (4), and the class code indicating the class is output. The class code generation circuit 30 outputs the class code to the normal equation addition circuit 33.

On the other hand, the output signal of the SD contour correction circuit 24 is supplied to the area cutout circuit 32 via the delay circuit 31. The delay circuit 31 delays the input SD data for the processing time of the area cutout circuit 28 and the motion class determination circuit 29, and the SD data is delayed by the area cutout circuit 3.
2 and the SD pixel data used for the estimation calculation is cut out. The area cutout circuit 32 is specifically the same as the area cutout circuit 8 described above, and functions to cut out the SD pixels necessary for the linear estimation formula according to the motion class. The output of the area cutout circuit 32 is supplied to the normal equation addition circuit 33.

Here, in order to explain the normal equation adding circuit 33, learning of a conversion formula from a plurality of SD pixels to HD pixels and signal conversion using the prediction formula will be described. For the sake of explanation, a case will be described below in which pixels are generalized and prediction is performed using n pixels. The re-quantized data obtained by performing p-bit ADRC on the SD pixel levels x ₁ , ..., X _n are q ₁ ,.
, Q _n . At this time, the class class of this area is defined by the above equation (5). As described above, when the SD pixel level is x ₁ , ..., X _n and the HD pixel level is y, the coefficients w ₁ ,.
An n-tap linear estimation formula by w _n is set. This is shown in equation (6). Before learning, wi is an undetermined coefficient.

Y = w ₁ x ₁ + w ₂ x ₂ + ... + w _n x _n (6)

Learning is performed on a plurality of signal data for each class. When the number of data is m, the following equation (7) is set according to the equation (6).

Y _k = w ₁ x _k1 + w ₂ x _k2 + ... + w _n x _kn (7) (k = 1, 2, ..., m)

When m> n, w ₁ , ..., W _n are
Since it cannot be uniquely determined, the element of the error vector e is defined by the equation (8), and the coefficient that minimizes the threshold (9) is obtained. This is the so-called least squares method.

E _k = y _k − {w ₁ x _k1 + w ₂ x _k2 + ... + w _n x _kn } (8) (k = 1, 2, ..., m)

[0054]

(Equation 3)

Here, the partial differential coefficient by w _{i in} equation (9) is obtained. It makes each w _i such that equation (10) becomes zero.
You should ask.

[0056]

[Equation 4]

Below, as in equations (11) and (12),
Defining X _ji Y _i , equation (10) can be rewritten as equation (13) using a matrix.

[0058]

(Equation 5)

[0059]

(Equation 6)

[0060]

(Equation 7)

This equation is generally called a normal equation. The normal equation addition circuit 33 supplies the class code supplied from the class code generation circuit 30 and the SD data x ₁ , ..., X _n supplied from the area cutout circuit 32.
This equation is added using the HD pixel level y corresponding to the SD data supplied from the HD correction circuit 25.

After inputting all the training data, the normal equation adding circuit 33 outputs the normal equation data to the prediction coefficient determining circuit 34. The prediction coefficient determination circuit 34 solves w _i using a general matrix solution method such as a sweeping method of a normal equation and calculates a prediction coefficient.
The prediction coefficient determination circuit 34 writes the calculated prediction coefficient in the memory 35. As a result of the training as described above, the memory 35 stores a prediction coefficient for estimating the HD data y of interest for each class, which is statistically closest to the true value. The table stored in the memory 35 is the ROM table 9 used in the image signal conversion apparatus of the present invention, as described above. Through the above processing, learning of coefficient data for creating HD data from SD data by a linear estimation formula is completed.

As is clear from the above description, in this embodiment, the learning target, S, used in the conventional learning is used.
The D data was created by down-converting the HD data whose contour has been corrected by the HD camera, whereas the SD data of this embodiment down-converts the HD data which has not been contour-corrected and is used by the SD camera. Since contour correction is performed by a contour correction circuit similar to that described above, it is generally called that the characteristics near the contour are imaged and contour-corrected by an SD camera as compared with conventional SD data. However, it is close to the SD data. Therefore, in the conventional image information conversion apparatus, the conversion performance was deteriorated particularly in the vicinity of the contour in the up conversion of the SD data created by the down conversion. However, by using the learning method of this embodiment, Deterioration can be suppressed, and even in the up-conversion of SD data captured by an SD camera, high conversion performance can be obtained overall.

In the description of this embodiment, in order to calculate the motion parameter, the inter-frame difference of the pixel data is calculated, and the average value of the absolute values is thresholded. This is an example, and the time difference may be divided by the spatial difference to obtain normalized data, and the data may be subjected to threshold processing.

Further, in the description of this embodiment, in order to obtain the motion class, the average value of the absolute values of the differences is preset as the threshold value, but the histogram of the absolute values of the differences is divided into n equal parts. It is also possible to set the value.

In the description of this embodiment, as the information compression means for patterning the spatial waveform with a small number of bits,
Although ADRC is provided, this is an example, and any information compression means can be provided as long as it can be expressed by a class having a small number of signal waveform patterns. For example, DPCM or VQ (vector quantization). You may use compression means, such as.

Further, in the description of this embodiment, for the sake of simplicity, the horizontal interpolation filter 11 is used for the horizontal up-conversion, but instead, a ROM for the horizontal up-conversion is prepared and the horizontal up-conversion is performed. It is of course possible to adopt a method using the estimation formula also in the up conversion.

[0068]

According to the present invention, the HD image data captured with the contour correction turned off is down-converted by the ideal filter, and the contour is corrected by the contour correction circuit similar to the contour correction circuit used in the SD camera. SD corrected
By using the image data as the learning target, it is possible to target the SD data closer to the camera input image than the conventional learning target SD data, and up-convert the SD image data captured by the SD camera, especially in the vicinity of the contour correction. The conversion performance can be improved.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of an image information conversion apparatus according to the present invention.

FIG. 2 is a schematic diagram for explaining a case of creating a correction data table.

FIG. 3 is a diagram for explaining a positional relationship between SD data and HD data.

FIG. 4 is a diagram for explaining data used for space class classification.

FIG. 5 is a diagram for explaining data used for motion class classification.

FIG. 6 is a diagram for explaining pixels used for estimation calculation.

FIG. 7 is a circuit diagram of an example of a conventional image information conversion device.

FIG. 8 is a circuit diagram showing an example of a waist of a conventional image information conversion device.

FIG. 9 is a block diagram for explaining when a correction data table is created in a conventional image conversion apparatus.

FIG. 10 is a block diagram for explaining a contour correction device.

[Explanation of symbols]

 22 Vertical decimation filter 23 Horizontal decimation filter 24 SD contour correction circuit 25 HD contour correction circuit 26, 28, 32 Region cutout circuit 27 ADRC circuit 29 Motion class determination circuit 30 Class code generation circuit 31 Delay circuit 33 Normal equation addition circuit 34 Prediction coefficient Decision circuit 35 memory

Claims (9)

[Claims] 1. An image information conversion device adapted to convert a digital image signal into a digital image signal having a larger number of pixels, and image cutting means for cutting image data at a predetermined position from image information supplied from the outside. , Detecting the pattern of the level distribution of the image information extracted by the image clipping means, based on the pattern,
The class detection means for determining the class to which the image information belongs and outputting the class detection information, the class code generation means for determining the class from the result of the class detection means, and the image information supplied from the outside Depending on the coefficient data storage means in which coefficient data of the estimation formula, which is information for converting into image information of resolution, is stored for each class, and the coefficient data supplied from the coefficient data storage means, An image information conversion device, comprising: an image conversion unit for converting and outputting image information of high resolution. 2. The image information conversion apparatus according to claim 1, wherein the image cutout unit cuts out image data at a predetermined position to obtain a space class from the image information supplied from the outside. Used by the clipping means, a second image clipping means for clipping image data at a predetermined position to obtain a motion class from the image information supplied from the outside, and the image conversion means from the image information supplied from the outside. To cut out image data at a predetermined position for
The image code conversion means, wherein the class code generation means determines a class from the spatial class and the motion class. 3. The image information conversion apparatus according to claim 1, wherein the coefficient data storage means converts the image information of a higher resolution at a position closer to the image information supplied from the outside. Of the image data supplied from the outside and the second coefficient data used when converting the image information of the higher resolution at a position far from the image information supplied from the outside are stored in the image information supplied from the outside. Based on the first
And one of the second coefficient data is selectively read out, and in the image conversion means, image information having a higher resolution than the above based on a linear linear equation of the read coefficient data and the image information supplied from the outside. An image information conversion device, characterized in that 4. The image information conversion device according to claim 2, wherein image data at a predetermined position is cut out from the first image cutout means, and the cutout image data is
A space class detecting unit that detects the space class by performing ADRC encoding, and image data at a predetermined position is cut out from the second image cutting unit, and the cut image data is
And a motion class detecting means for detecting the motion class by performing ADRC encoding, wherein the ADRC encoding is the maximum value of a plurality of data included in a two-dimensional or three-dimensional block and the plurality of data. Has been modified to have a relative level relationship based on the value defining the dynamic range, and the means for detecting the minimum value of the above, the means for detecting the dynamic range of the block from the above maximum value and the minimum value. An image information conversion apparatus comprising: means for forming modified input data; and means for quantizing the modified input data with a bit number equal to or less than the original quantization bit number. 5. An image information conversion apparatus for converting a digital image signal into a digital image signal having a larger number of pixels, and a filter means for converting image information of higher resolution into standard image information, First contour correction means for performing contour correction on standard image information, second contour correction means for performing contour correction on image information with higher resolution, and predetermined from the standard image information. An image cropping unit that crops the image data at the position of, and a pattern of the level distribution of the image information extracted by the image cropping unit is detected, and based on the pattern,
Class detection means for determining the class to which the image information belongs and outputting class detection information, class code generation means for determining the class from the result of the class detection means, the standard image information and the second contour Coefficient data generation means for generating coefficient data of an estimation formula, which is information for converting the standard image information from the higher resolution image information from the correction means to the higher resolution image information, and An image information conversion device comprising: coefficient data storage means in which coefficient data is stored for each class. 6. The image information conversion apparatus according to claim 5, wherein in the filter means, the image information in the horizontal direction is halved with respect to the image information of the higher resolution, and further the image information in the vertical direction. The image information conversion device is characterized by comprising image filter means for converting the standard image information to 1/2. 7. The image information conversion apparatus according to claim 5, wherein the image cutout unit cuts out image data at a predetermined position to obtain a space class from the standard image information. A second image cutout means for cutting out image data at a predetermined position to obtain a motion class from the standard image information; and a predetermined position for use in the image conversion means from the standard image information. An image information conversion apparatus, comprising: a third image clipping means for clipping image data, wherein the class code generating means determines a class from the spatial class and the motion class. 8. The image information conversion apparatus according to claim 5, wherein the coefficient data storage means converts the first image information having a higher resolution at a position closer to the standard image information. Data and second coefficient data for converting image information of a higher resolution at a position far from the standard image information are stored, and the first coefficient data is stored based on the standard image information. And the second
2. An image information conversion device, characterized in that one of the coefficient data is selectively read out. 9. The image information conversion apparatus according to claim 7, wherein image data at a predetermined position is cut out from the first image cutout means, and the cutout image data is
A space class detecting unit that detects the space class by performing ADRC encoding, and image data at a predetermined position is cut out from the second image cutting unit, and the cut image data is
And a motion class detecting means for detecting the motion class by performing ADRC encoding, wherein the ADRC encoding is the maximum value of a plurality of data included in a two-dimensional or three-dimensional block and the plurality of data. Has been modified to have a relative level relationship based on the value defining the dynamic range, and the means for detecting the minimum value of the above, the means for detecting the dynamic range of the block from the above maximum value and the minimum value. An image information conversion apparatus comprising: means for forming modified input data; and means for quantizing the modified input data with a bit number equal to or less than the original quantization bit number.