JP3826495B2 - Image signal processing apparatus and method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image signal processing apparatus and method for performing processing for improving the resolution of an image signal.
[0002]
[Prior art]
Conventionally, as an image signal, for example, a high definition (HD) having, for example, 1125 scanning lines, together with an SD signal corresponding to a standard definition (SD) television receiver having, for example, 525 scanning lines. HD signals corresponding to television receivers (hereinafter referred to as HDTV) are provided.
[0003]
By the way, an SD signal may be given to the HDTV. In such a case, signal conversion for converting an SD signal into an HD signal, so-called up-conversion is performed.
[0004]
[Problems to be solved by the invention]
At the time of this up-conversion, a process for increasing pixels by linear interpolation has been performed. However, it has been impossible to improve the resolution by this up-conversion by linear interpolation.
[0005]
The present invention has been made in view of the above-described problems, and an object of the present invention is to provide an image signal processing apparatus and method for improving resolution at the time of up-conversion.
[0006]
[Means for Solving the Problems]
  In order to solve the above problems,BookAn image processing apparatus according to the invention includes:Blocking means for blocking the first image signal in the time axis direction, and the blocked firstFor image signalforFirst polynomialFor nth-order polynomialFittingPerform, This first polynomialN-th order polynomialSignal analysis means for deriving a first parameter to be identified;The first parameter acquired by applying the first polynomial to the first image signal, and a second resolution higher than the first image signal for the same object as the first image signal. A statistical process is performed on the second parameter for specifying the n + m degree polynomial obtained by fitting an n + m degree polynomial, which is a second polynomial, to the image signal by (n + m) degree curved surface fitting, and a correspondence relationship is extracted. Set byCorresponding to the first parameterthe aboveSpecified by the parameter table from which the second parameter is read and the second parameter,the aboveSecond polynomialN + m degree polynomialA signal generating means for generating a signal atFirstSignal combining means for combining the image signal and the signal generated by the signal generating means into an image signal;It is characterized by that.
[0007]
  Also,BookAn image processing method according to the invention includes:A blocking step of blocking the first image signal in the time axis direction, and the blocked first stepFor image signalforFirst polynomialFor nth-order polynomialFittingPerform, This first polynomialN-th order polynomialA signal analysis step for deriving a first parameter to be identified;The first parameter acquired by applying the first polynomial to the first image signal, and a second resolution higher than the first image signal for the same object as the first image signal. A statistical process is performed on the second parameter for specifying the n + m degree polynomial obtained by fitting an n + m degree polynomial, which is a second polynomial, to the image signal by (n + m) degree curved surface fitting, and a correspondence relationship is extracted. From the parameter table set bySecond parameter corresponding to the first parameterTheIt is specified by the reading process to read and the second parameter.the aboveSecond polynomialN + m degree polynomialA signal generation step of generating a signal atFirstA signal synthesis step of synthesizing the image signal and the signal generated in the signal generation step into an image signal;It is characterized by that.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an image signal processing apparatus and method according to the present invention will be described in detail with reference to the drawings.
[0011]
As shown in FIG. 1, the example of the image signal processing apparatus includes a blocking unit 1 that blocks an image signal in the time axis direction. This blocking unit 1 divides a standard resolution SD signal having a pixel number and image quality equivalent to a standard television receiver having, for example, 525 scanning lines into blocks. For example, the blocking unit 1 divides a two-dimensional pixel of an input SD signal into a block.
[0012]
The image signal processing apparatus includes a signal fitting unit 2 that is a signal analysis unit that performs signal fitting on a blocked image signal, and a parameter table 3 from which parameters are read according to the result of the signal fitting unit 2. And a signal generation unit 4 that generates a signal using a polynomial specified by the parameters acquired from the parameter table 3.
[0013]
The signal fitting unit 2 is supplied with the blocked signal from the blocking unit 1, and performs signal fitting on the blocked signal. Here, the n-th order polynomial that is the first polynomial is fitted by so-called n-th order curved surface fitting, and a parameter that specifies the n-th order polynomial is set as the first parameter. As the first parameter, for example, the coefficient of the n-th order polynomial can be used. Here, n is an integer of 1 or more.
[0014]
The parameter table 3 is a table that holds a correspondence relationship between a first parameter that specifies an n-th order polynomial that is a first polynomial and a second parameter that specifies an n + m-order polynomial that is a second polynomial. . Here, since m is an integer of 1 or more, the second polynomial has a higher degree than the first polynomial. As the second parameter, for example, a coefficient of the n + m degree polynomial can be used. The parameter table 3 can be realized by addressing a ROM table.
[0015]
For example, as shown in FIG. 2, the parameter table 3 includes first parameters A1, A2, A3,... That specify an nth order polynomial that is a first polynomial, and an n + mth order polynomial that is a second polynomial. Are held in correspondence with the second parameters B1, B2, B3,. In the figure, a correspondence relationship in which A4 in the first parameter is linked to B4 in the second parameter is shown.
[0016]
As will be described later, the parameter table captures the same target, the first parameter applied to the SD signal, and the number of pixels corresponding to, for example, a high-resolution television receiver having 1125 scanning lines And a correspondence relationship with the second parameter applied to the high-resolution HD signal having the image quality is set in advance.
[0017]
The signal generator 4 shown in FIG. 1 receives a second parameter that specifies an n + m-th order polynomial that is a second polynomial, and generates a signal based on the n + m polynomial. That is, the signal generator 4 reads the image based on the n + m degree polynomial that is read from the parameter table 3 and specified by the second parameter corresponding to the first parameter derived by the signal analyzer 2. Generate a signal.
[0018]
Further, the image signal processing apparatus includes a signal synthesis unit 5 that synthesizes image signals. The signal synthesis unit 5 synthesizes the image signal obtained from the blocking unit 1 and the signal generated by the signal generation unit 4 and outputs the image signal obtained by the synthesis as an HD signal.
[0019]
Here, the signal synthesis includes, for example, multiplying or pasting the image signal given from the blocking unit 1 and the signal obtained from the signal generating unit 4 by weighting. Note that the pasting means so-called mapping in which the signal is pasted on the image signal. In addition, the signal may be delayed for timing adjustment during signal synthesis.
[0020]
Subsequently, an example of an image signal processing method will be described. A series of procedures relating to this image signal processing method is shown in the flowchart of FIG.
[0021]
In the first step S1, the input SD signal is blocked. For example, the SD signal is set to one block every predetermined period. And it progresses to step S2 following this.
[0022]
In step S2, signal fitting is performed on the image signal blocked in step S1. That is, an n-th order polynomial that is a first polynomial is applied to the blocked signal, and for example, the coefficient of the n-th order polynomial is determined as the first parameter. However, n is an integer of 1 or more. Then, the process proceeds to the next step S3.
[0023]
In step S3, the parameter table is referred to. That is, the second parameter corresponding to the first parameter obtained in step S2 is read from the parameter table.
[0024]
Here, the parameter table holds a correspondence relationship between the first parameter that specifies the n-th order polynomial that is the first polynomial and the second parameter that specifies the n + m-order polynomial that is the second polynomial. It is a table to do. However, since m is an integer of 1 or more, the second polynomial has a higher degree than the first polynomial. As the second parameter, for example, a coefficient of the n + m degree polynomial can be used. When the reading of the second parameter from the parameter table is completed, the process proceeds to the next step S4.
[0025]
In step S4, a signal is generated based on the n + m-th order polynomial that is the second polynomial specified by the second parameter obtained in step S3. And it progresses to step S5 following this.
[0026]
In step S5, the image signal blocked in step S1 and the signal generated in step S4 are combined, and the image signal obtained by the combination is output as an HD signal. Here, the signal synthesis includes, for example, weighting multiplication or pasting the signal obtained from the signal generation unit 4 and the signal provided from the blocking unit 1.
[0027]
As described above, the image signal processing apparatus and method generates a signal from an n + m-order polynomial that is a second polynomial having a higher order than the n-order polynomial that is the first polynomial applied to the SD signal. Then, an image signal obtained by synthesizing the image signal and the original SD signal is output as an HD signal. Since the order of the polynomial corresponding to the image signal is improved from the nth order of the first polynomial to the n + mth order of the second polynomial, an image signal with improved resolution can be obtained. By processing this image signal, it is possible to generate an HD signal from the SD signal that can be expressed with higher resolution and finer resolution than the SD signal.
[0028]
Next, a method for creating the parameter table 3 by learning will be described. As described above, the parameter table 3 shows the correspondence relationship between the first parameter that specifies the n-th order polynomial that is the first polynomial and the second parameter that specifies the n + m-order polynomial that is the second polynomial. It is to hold.
[0029]
As shown in FIG. 4, a series of procedures for creating the parameter table 3 is obtained by imaging the same target with the SD camera 11 that outputs the SD signal and the HD camera 14 that outputs the HD signal. The first signal fitting unit 13 and the second signal fitting unit 15 derive the first parameter and the second parameter for the SD signal and the HD signal, respectively. The first parameter and the second parameter are derived from the first parameter fitting unit 13 and the second signal fitting unit 15, respectively. Is recorded in the frequency table 17 and the correspondence between the first parameter and the second parameter is analyzed by the determination unit 18, and a corresponding table of the result is recorded in the memory 19.
[0030]
That is, the SD signal which is an image signal obtained by the SD camera 11 is blocked in the time axis direction by the first blocking unit 12. Examples of the blocking in the first blocking unit 12 include dividing the two-dimensional pixels of the input image signal into blocks. The blocked SD signal is fitted by fitting to an n-th order polynomial which is a first polynomial in the first signal fitting unit 13, and a first parameter specifying the n-order polynomial is derived. As the first parameter, for example, the coefficient of the n-th order polynomial can be used.
[0031]
On the other hand, the HD signal obtained by the HD camera 14 that captures the same target as the SD camera 11 is blocked by the second blocking unit 15. As the blocking in the second blocking unit 15, for example, the blocking can be performed in synchronization with the first blocking unit 12.
[0032]
This blocked HD signal is applied to a second polynomial n + m degree polynomial by so-called (n + m) degree curved surface fitting in the second signal fitting unit 14, and a second parameter specifying this n + m degree polynomial is provided. Derived. As the second parameter, for example, the coefficient of the n + m degree polynomial can be used.
[0033]
Here, the SD camera 11 and the HD camera 14 capture the same target to obtain the SD signal and the HD signal, respectively, but the HD signal captured by the HD camera 14 is processed by a low-pass filter or subsampling. As a result, a signal equivalent to the SD signal obtained by imaging the same object is obtained from this HD signal.
[0034]
A pair of the first parameter and the second parameter obtained by the first signal fitting unit 13 and the second signal fitting unit 16 is recorded in the frequency table 17. Recording is continued in the frequency table 17 until, for example, a predetermined amount of the first parameter and second parameter pairs is accumulated.
[0035]
The determination unit 18 makes a determination based on a pair of the first parameter and the second parameter accumulated in the frequency table 17. For example, statistical processing is performed to extract an effective correspondence. Then, the correspondence relationship between the first parameter and the second parameter is determined.
[0036]
The memory 19 stores a correspondence relationship between the first parameter and the second parameter determined by the determination unit 18. The correspondence between the first parameter and the second parameter stored in the memory 19 is read out and set in the parameter table 3 when necessary. The correspondence between the first parameter and the second parameter stored in the memory 19 is also used in the image signal processing method.
[0037]
As described above, the correspondence between the first parameter and the second parameter is the same as that of the SD signal and the first parameter that specifies the n-th order polynomial that is the first polynomial derived from the SD signal. A pair with a second parameter for specifying an n + m degree polynomial that is a second polynomial derived from the HD signal obtained by imaging the object is recorded in the frequency table 17, and the record in the frequency table 17 is analyzed by the determination unit 18. It can be obtained by a series of procedures related to learning.
[0038]
By such learning, the correspondence between the first parameter corresponding to the SD signal and the second parameter corresponding to the HD signal is obtained, and the HD signal is generated from the SD signal by using this correspondence. It becomes possible.
[0039]
Next, another example of the image signal processing apparatus will be described.
As shown in FIG. 5, the image signal processing apparatus includes a blocking unit 1 that blocks an image signal in the time axis direction. This blocking unit 1 blocks the input SD signal in the time axis direction, similarly to the blocking unit 1 included in the image signal processing apparatus described above. For example, the blocking unit 1 may block an input image signal every predetermined time.
[0040]
The image signal processing apparatus includes a class classification unit 6 that is a signal analysis unit that classifies a blocked image signal according to a waveform pattern, and a parameter table 3 from which parameters are read according to the result of the class classification unit 2. And a signal generation unit 4 that generates an image signal corresponding to the parameters obtained in the parameter table 3.
[0041]
The class classification unit 6 is supplied with the blocked signal from the blocking unit 1 and classifies the blocked signal 2 according to the waveform pattern.
[0042]
Here, examples of the class classification include waveform classification based on vector quantization, compression coding, and adaptive dynamic range coding (ADRC).
[0043]
This class classification by ADRC is an adaptive requantization method originally developed for high-efficiency coding for VTRs, and can efficiently express a local pattern of signal level with a short word length.
[0044]
Here, ARDC will be described by taking 1-bit ARDC as an example. An example of the ARDC circuit is shown in FIG. As shown in FIG. 7, the ADRC circuit includes pixels P1 to P3 in the first row from the left to the right in the drawing, pixels P4 to P6 in the second row, and pixels (3 × 3) in the third row. 3) ADRC is applied to the block B in which a total of nine pixels are arranged.
[0045]
That is, in the ADRC circuit, regarding the data converted into the block order from the input terminal 101, the detection circuit 102 detects the maximum value MAX and the minimum value MIN for each block. MAX and MIN are supplied to the subtraction circuit 103, and a dynamic range DR is generated at the output. The input data and MIN are supplied to the subtraction circuit 104, and the minimum value is removed from the subtraction circuit 104, thereby generating normalized pixel data.
[0046]
The dynamic range DR is supplied to the subtraction circuit 105, the normalized pixel data is divided by the dynamic range DR, and the output data of the division circuit 105 is supplied to the comparison circuit 106. In the comparison circuit 106, it is determined whether the percent calculation power of the eight pixels other than the central pixel P5 is larger or smaller with reference to 0.5. Depending on this result, data DT of “0” or “1” is generated. This comparison output DT is taken out to the output terminal 107. If class division is performed using this 1-bit ADRC, a (3 × 3) block B is represented by a 9-bit class code.
[0047]
The parameter table 3 is a table that holds a correspondence relationship between a class given by the class classification unit 6 and a parameter for specifying a k-th order polynomial. Here, k is an integer of 1 or more. As the parameter, for example, the coefficient of the k-th order polynomial can be used. Here, the parameter table 3 can be realized, for example, by addressing a ROM table.
[0048]
For example, the parameter table 3 will be described with reference to FIG. 2. The classes A1, A2, A3,... Given from the class classification unit 6 and the parameters B1, B2, B3 for specifying the k-th order polynomial ,... Are retained. In the figure, a correspondence relationship in which the class A4 and the parameter B4 are linked is shown.
[0049]
As will be described later, the parameter table 3 in FIG. 5 is a parameter that specifies a class derived by classifying an SD signal and a k-order polynomial in which the SD signal is applied to an HD signal obtained by imaging the same target. Is set in advance by analyzing the correspondence relationship between
[0050]
The signal generation unit 4 generates a signal from a k polynomial specified by a parameter given from the parameter table 3. That is, the signal generation unit 4 generates a signal based on the k-th order polynomial specified by the parameter read from the parameter table 3 corresponding to the class derived by the class classification unit 2.
[0051]
Further, the image signal processing apparatus includes a signal synthesis unit 5 that synthesizes image signals. The signal synthesis unit 5 synthesizes the image signal obtained from the blocking unit 1 and the signal generated by the signal generation unit 4 and outputs the image signal obtained by the synthesis as an HD signal.
[0052]
Here, the signal synthesis includes, for example, weighting multiplication and pasting of the signal given from the blocking unit 1 and the signal obtained from the signal generating unit 4. Note that the pasting means mapping. In addition, the signal may be delayed for timing adjustment when the signals are combined.
[0053]
Next, another example of the image signal processing method will be described. A series of procedures relating to this image signal processing method is shown in the flowchart of FIG.
[0054]
In the first step S11, the input SD signal is blocked. As this blocking, for example, the SD signal is made one block every predetermined time. And it progresses to step S12 following this.
[0055]
In step S12, the image signals blocked in step S11 are classified into classes based on the waveform pattern. As this classification, for example, ADRC can be used. Since the classification has been described in detail above, the description thereof is omitted here. Then, the process proceeds to the next step S13.
[0056]
In step S13, the parameter table is referred to. That is, the parameter corresponding to the class obtained in step S12 is read from the parameter table.
[0057]
Here, the parameter table is a table that holds a correspondence relationship between a class obtained by class classification based on a waveform pattern of a blocked image signal and a parameter that specifies a k-th order polynomial. As the parameter, for example, the coefficient of the k-th order polynomial can be used. When reading of the parameters from the parameter table is completed, the process proceeds to the next step S14.
[0058]
In step S14, a signal is generated based on the k-th order polynomial specified by the parameter obtained in step S13. And it progresses to step S15 following this.
[0059]
In step S15, the image signal blocked in step S11 and the signal generated in step S14 are combined, and the image signal obtained by the combination is output as an HD signal. Here, the signal synthesis includes, for example, weighted multiplication of the signal obtained from the signal generation unit and the signal provided from the blocking unit.
[0060]
As described above, the image signal processing apparatus and method generates a signal based on a kth order polynomial corresponding to a class derived by class classification based on a waveform pattern of the SD signal, and this signal and the blocked SD An image signal obtained by combining the signals is output as an HD signal. Therefore, in this image signal processing apparatus and method, the HD signal can be obtained from the input SD signal by appropriately selecting the order k of the k-order polynomial. In other words, the image signal processing apparatus and method can generate an HD signal having higher resolution than the SD signal and capable of finer expression from the SD signal.
[0061]
Next, another example of a learning method for creating the parameter table 3 by learning will be described. As described above, the parameter table 3 holds a correspondence relationship between a class based on the classification of the input SD signal and a parameter for specifying the k-th order polynomial.
[0062]
As shown in FIG. 9, a series of procedures for creating the parameter table 3 includes a first block of SD signals and HD signals obtained by imaging the same target with the SD camera 11 and the HD camera 14, respectively. After classifying by the converting unit 12 and the second blocking unit 15, the class classification unit 21 and the signal fitting unit 22 derive classes and parameters, respectively, and record these classes and parameters in the frequency table 17. The correspondence relationship recorded in the frequency table 17 is analyzed by the determination unit 18 so that a correspondence table as a result of the analysis is recorded in the memory 19.
[0063]
That is, the SD signal which is an image signal obtained by the SD camera 11 is blocked in the time axis direction by the first blocking unit 12. As the blocking, for example, the input image signal is blocked every predetermined time. The image signal blocked by the first blocking unit 12 is classified by the class classification unit 21 according to the waveform pattern. Since this classification has been described in detail above, a description thereof is omitted here.
[0064]
On the other hand, an HD signal that is an image signal obtained by the HD camera 14 that captures the same object as the SD camera 11 is blocked by the second blocking unit 15. As this blocking, for example, it is possible to block in synchronization with the first blocking unit 12. A k-th order polynomial is fitted to the blocked image signal by so-called k-th order curved surface fitting in the signal fitting unit 22, and a parameter specifying this k-th order polynomial is derived. As this parameter, for example, the coefficient of the k-th order polynomial can be used.
[0065]
Here, the SD camera 11 and the HD camera 14 capture the same target to obtain the SD signal and the HD signal, respectively, but the HD signal captured by the HD camera 14 is processed by a low-pass filter or subsampling. As a result, a signal equivalent to the SD signal obtained by imaging the same object is obtained from the HD signal.
[0066]
The class / parameter pairs obtained by the class classification unit 21 and the signal fitting unit 22 are recorded in the frequency table 17. Recording is continued in the frequency table 17 until, for example, a predetermined amount of the class and parameter pairs are accumulated.
[0067]
The determination unit 18 analyzes the class and parameter pairs stored in the frequency table 17 and makes a determination based on the analysis. For example, the determination unit 18 performs a statistical process on the class / parameter pairs recorded in the frequency table 17 and extracts an effective correspondence. Then, the correspondence between the class and the parameter is determined.
[0068]
The memory 19 stores the correspondence between the class and parameters determined by the determination unit 18. The correspondence relationship between the class and parameters stored in the memory 19 is read out and set in the parameter table 3 when necessary. The correspondence between the class and parameters stored in the memory 19 is also used in the image signal processing method.
[0069]
As described above, the correspondence relationship between the class and the parameter is as follows: the class derived from the waveform pattern of the SD signal, and the parameter specifying the kth order polynomial derived from the HD signal obtained by imaging the same object as the SD signal. Thus, the frequency table 17 can be created by learning.
[0070]
By such learning, the correspondence relationship between the class corresponding to the SD signal and the parameter corresponding to the HD signal is obtained, and the HD signal can be generated from the SD signal by using this correspondence relationship.
[0071]
In this specific example, the conversion from the SD signal to the HD signal has been described. However, the present invention is not limited to this, and can be widely used for the purpose of improving the image quality by increasing the resolution of the image signal.
[0072]
【The invention's effect】
As described above, an example of the image signal processing apparatus uses an n + m-th order that specifies the second parameter corresponding to the first parameter from the first parameter that specifies the n-order polynomial that is applied to the input SD signal. A signal is generated based on a polynomial, and an HD signal is obtained by synthesizing this signal and the input SD signal. Therefore, this image signal processing apparatus can provide a high resolution HD signal from the input SD signal.
[0073]
An example of the image signal processing method is the same as the example of the image signal processing apparatus described above. From the first parameter that specifies the nth order polynomial applied to the input SD signal, the second parameter corresponding to this first parameter is used. An HD signal is obtained by creating a signal based on the n + m degree polynomial specified by the parameter and synthesizing this signal and the input SD signal. Therefore, this image signal processing method can provide a high-resolution HD signal from the input SD signal.
[0074]
Furthermore, another example of the image signal processing apparatus generates a signal based on a class obtained by class classification of the waveform pattern of the input SD signal and a k-th order polynomial specified by a parameter corresponding to the class, The HD signal is obtained by combining the signal and the input SD signal. Therefore, this image signal processing apparatus can provide a high resolution HD signal from the input SD signal.
[0075]
Another example of the image signal processing method specifies the class obtained by classifying the waveform pattern of the input SD signal and the parameter corresponding to this class, as in the other example of the image signal processing apparatus. An HD signal is obtained by generating a signal based on the k-th order polynomial and synthesizing this signal and the input SD signal. Therefore, this image signal processing method can provide a high-resolution HD signal from the input SD signal.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a schematic configuration of an example of an image signal processing apparatus.
FIG. 2 is a diagram for explaining a parameter table of the image signal processing apparatus.
FIG. 3 is a block diagram showing a series of steps of an example of an image signal processing method.
FIG. 4 is a block diagram schematically showing an example of the parameter table learning method;
FIG. 5 is a block diagram showing a schematic structure of another example of an image signal processing apparatus.
FIG. 6 is a circuit diagram of an adaptive dynamic range coding (ADRA) circuit.
FIG. 7 is a diagram illustrating one block obtained by dividing a pixel of an image signal into blocks.
FIG. 8 is a flowchart showing a series of procedures of another example of the image signal processing method.
FIG. 9 is a block diagram schematically showing another example of a parameter table learning method of the image signal processing apparatus.
[Explanation of symbols]
1 Blocking unit, 2 signal fitting unit, 3 parameter table, 4 signal generating unit, 5 signal synthesis unit, 17 frequency table

Claims (6)

Blocking means for blocking the first image signal in the time axis direction;
Performs fitting signal by n-th order Surface Fitting for the n-th order polynomial is a first polynomial for the first image signal this which is blocked, a first parameter identifying a polynomial of degree n is a first polynomial Signal analysis means for deriving
The first parameter acquired by applying the first polynomial to the first image signal, and a second resolution higher than the first image signal for the same object as the first image signal. A statistical process is performed on the second parameter for specifying the n + m degree polynomial obtained by fitting an n + m degree polynomial, which is a second polynomial, to the image signal by (n + m) degree curved surface fitting, and a correspondence relationship is extracted. a parameter table is set, the said second parameter corresponding to the first parameter is read by,
Identified by the second parameter, and signal generating means for generating a signal at the n + m order polynomial is the second polynomial,
An image signal processing apparatus comprising: a signal synthesis unit that synthesizes the first image signal and the signal generated by the signal generation unit into an image signal. The first parameter is a coefficient of an n-th order polynomial that is the first polynomial, and the second parameter is a coefficient of an n + m-order polynomial that is the second polynomial. The image signal processing apparatus according to 1. In the statistical processing, a pair of a first parameter that specifies an n-th order polynomial that is the first polynomial and a second parameter that specifies an n + m-order polynomial that is the second polynomial is recorded in a frequency table. The image signal processing apparatus according to claim 1, wherein the record of the frequency table is analyzed. A blocking step of blocking the first image signal in the time axis direction;
Performs fitting signal by n-th order Surface Fitting for the n-th order polynomial is a first polynomial for the first image signal this which is blocked, a first parameter identifying a polynomial of degree n is a first polynomial A signal analysis process for deriving
The first parameter acquired by applying the first polynomial to the first image signal, and a second resolution higher than the first image signal for the same object as the first image signal. A statistical process is performed on the second parameter for specifying the n + m degree polynomial obtained by fitting an n + m degree polynomial, which is a second polynomial, to the image signal by (n + m) degree curved surface fitting, and a correspondence relationship is extracted. A step of reading out the second parameter corresponding to the first parameter from the parameter table set by
Said specified by the second parameter, a signal generating step of generating a signal at the n + m order polynomial is the second polynomial,
An image signal processing method comprising: a signal synthesis step of synthesizing the first image signal and the signal generated in the signal generation step into an image signal. The first parameter is a coefficient of an n-th order polynomial that is the first polynomial, and the second parameter is a coefficient of an n + m-order polynomial that is the second polynomial. 5. The image signal processing method according to 4. In the statistical processing, a pair of a first parameter that specifies an n-th order polynomial that is the first polynomial and a second parameter that specifies an n + m-order polynomial that is the second polynomial is recorded in a frequency table. 5. The image signal processing method according to claim 4, wherein the recording of the frequency table is analyzed.

Images