JP2010166349A - Noise reduction apparatus and program therefor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a noise reduction apparatus that prevents the deterioration of picture quality in a digital video signal having large movement of an object. <P>SOLUTION: The noise reduction apparatus 1 reduces noise that is superimposed to a digital video signal, and it includes: an input terminal T<SB>IN1</SB>to which the digital video signal is input; an input terminal T<SB>IN2</SB>to which a level value indicating noise level is input; a median filter 11 to select a median of target pixels in a plurality of frames; a pixel value detection means 12 to detect a pixel value of the target pixel of a certain frame; a threshold setting means 13 to set a threshold corresponding to the level value; a pixel setting value computation means 14 to compute the specified pixel setting value; a noise reducing means 15 to set the pixel setting value as an output signal; and an output terminal T<SB>OUT</SB>to output an output signal. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a noise reduction device that reduces noise superimposed on a digital video signal and a program thereof.

  In general, in addition to the original video itself, noise added via a transmission path or the like is superimposed on the analog video signal. Conventionally, as a technique for reducing such noise, there is a technique called a noise reducer, and various methods have been proposed so far (see, for example, Patent Document 1).

  Here, for example, in order to operate the noise reducer described in Patent Document 1, it is necessary to detect the level of the noise component in the screen (noise level) as a reference for reducing noise. . For example, a noise reducer (noise reduction circuit) described in Patent Literature 1 detects a noise level from a signal in a blanking period of an analog video signal. Since the signal is flat (the voltage level is constant) during this blanking period, if even a slight signal component is superimposed during this period, it can be regarded as noise. Therefore, conventionally, noise reduction processing corresponding to the amount of noise is performed by regarding the noise superimposed on the signal in the blanking period as the noise superimposed on the entire video signal.

  On the other hand, digital video signals are discrete values of video signal data, so they are less likely to deteriorate than analog video signals on the transmission line. Even when transmission noise is superimposed, noise is corrected by error correction or the like. Can be reduced.

Japanese Patent No. 399026 (Japanese Patent Laid-Open No. 7-250264)

  However, in the invention according to Patent Document 1, there is a problem that in a digital video signal in which the luminance changes greatly between frames due to the movement of the subject, image quality deterioration such as blurring of the subject with large motion occurs. . For example, as shown in FIG. 21, in the invention according to Patent Document 1, since noise reduction processing is performed on all pixels, a portion with a large movement in this image, for example, various portions such as the left arm and right foot of the subject H Causes image degradation.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a noise reduction apparatus and a program therefor that suppress image quality deterioration in a digital video signal with a large subject movement.

  In order to solve the above problem, a noise reduction device according to claim 1 is a noise reduction device that reduces noise superimposed on a digital video signal, and includes a video signal input unit, a median filter, and a pixel value detection unit. And a pixel set value calculation means and a noise reduction means.

  According to this configuration, the noise reduction device inputs a digital video signal from the outside by the video signal input means. In addition, the noise reduction device presets a pixel at the same screen position as a target pixel in a median filter, a frame constituting the digital video signal input by the video signal input means, and a frame before and after this frame. The median value in a plurality of target pixels is selected. And a noise reduction apparatus detects the pixel value of the object pixel of a flame | frame by a pixel value detection means.

  Further, the noise reduction device determines whether or not the absolute value of the difference between the median values selected by the median filter from the pixel values detected by the pixel value detection unit is larger than a preset threshold value, and the absolute value is When it is not determined that the value is larger than the threshold value, a predetermined pixel setting value is calculated. This threshold value is preset according to the level value of the noise level superimposed on the digital video signal. For example, when the level value is high, the threshold value is set to a large value, and when the level value is low, the threshold value is set. Set to a smaller value.

  Further, the noise reduction device, when the pixel setting value calculation unit determines by the noise reduction unit that the absolute value is larger than the threshold value, the pixel value is used as an output signal, and the absolute value is not determined to be larger than the threshold value. In the case, the pixel set value is output as an output signal. Here, when it is determined that the absolute value is larger than the threshold value, that is, when the target pixel is set in a portion with a large movement such as an edge where the pixel value changes suddenly, the noise reduction device changes the pixel value. Output signal. As a result, the noise reduction apparatus performs a noise reduction process in a portion where the motion is small in the digital video signal, and reduces a noise reduction process in a portion where the motion is large.

  In order to solve the above-described problem, a noise reduction device according to claim 2 is a noise reduction device that reduces noise superimposed on a digital video signal, and includes a video signal input unit, a noise level detection unit, The threshold setting unit, the median filter, the pixel value detection unit, the pixel set value calculation unit, and the noise reduction unit are provided.

  According to this configuration, the noise reduction device inputs a digital video signal from the outside by the video signal input means. In the noise reduction device, the noise level detection means detects the noise level superimposed on the digital video signal input by the video signal input means as a level value indicating the magnitude of the noise level. Then, the noise reduction device sets the threshold value based on the level value detected by the noise level detection means by the threshold value setting means. Furthermore, the noise reduction device uses a median filter to preset a pixel at the same screen position as a target pixel in a frame constituting a digital video signal input by the video signal input means and in a frame before and after this frame. The median value in a plurality of target pixels is selected. Furthermore, the noise reduction device detects the pixel value of the target pixel of the frame by the pixel value detection means.

  The noise reduction device determines whether the absolute value of the difference between the median values selected by the median filter from the pixel values detected by the pixel value detection unit is larger than a preset threshold value, and the absolute value is a threshold value. If it is not determined that the value is larger than the predetermined value, a predetermined pixel setting value is calculated. This threshold value is preset according to the level value of the noise level superimposed on the digital video signal. For example, when the level value is high, the threshold value is set to a large value, and when the level value is low, the threshold value is set. Set to a smaller value.

  Further, the noise reduction device, when the pixel setting value calculation unit determines by the noise reduction unit that the absolute value is larger than the threshold value, the pixel value is used as an output signal, and the absolute value is not determined to be larger than the threshold value. In the case, the pixel set value is output as an output signal. Here, when it is determined that the absolute value is larger than the threshold value, that is, when the target pixel is set in a portion with a large movement such as an edge where the pixel value changes suddenly, the noise reduction device changes the pixel value. Output signal. As a result, the noise reduction apparatus performs a noise reduction process in a portion where there is little movement in the digital video signal, and reduces a noise reduction process in a region where the movement is large.

  The noise reduction device according to claim 3 is the noise reduction device according to claim 1 or 2, wherein the pixel setting value calculation means determines that the median value when the absolute value is not determined to be larger than the threshold value. Is output as a pixel set value.

  According to such a configuration, if the absolute value is not determined to be greater than the threshold value, the noise reduction device is considered to contain small noise in the digital video signal, so the median value is output as the pixel setting value, Reduce this small noise

  The noise reduction device according to claim 4 is the noise reduction device according to claim 1 or 2, wherein a correspondence relationship between the pixel value detected by the pixel value detection unit and the shift amount is set in advance, and this correspondence A shift amount calculating unit that calculates a shift amount according to the pixel value detected by the pixel value detecting unit based on the relationship is further provided, and the pixel set value calculating unit is not determined that the absolute value is larger than the threshold value Determines whether the pixel value is greater than the median value, and if it is determined that the pixel value is greater than the median value, calculates the pixel setting value by subtracting the shift amount from the pixel value, If the pixel value is not determined to be larger than the median value, the pixel setting value is calculated by adding the pixel value and the shift amount.

  According to such a configuration, the noise reduction device is considered to include small noise in the target pixel when the absolute value is not determined to be larger than the threshold value. An appropriate pixel setting value corresponding to the value is calculated, and this noise is reduced.

  A noise reduction device according to claim 5 is the noise reduction device according to any one of claims 2 to 4, wherein the noise level detection means includes a block division means, an orthogonal transformation means, a high frequency It comprises a component detection means, a high frequency component accumulation means, and a noise level determination means.

  According to such a configuration, the noise level detecting means divides the digital video signal into blocks having a predetermined size for each screen by the block dividing means. The block obtained by dividing the screen includes at least a flat portion with little change in pixel value in the screen. Further, the noise level detecting means orthogonally transforms the blocks divided by the block dividing means by the orthogonal transform means for each block. As a result, the pixel value of the block is converted into the coefficient of the frequency component.

  The noise level detecting means averages the coefficients of the high frequency components excluding the DC component of the block transformed by the orthogonal transform means and the vicinity of the predetermined DC component in the block by the high frequency component detecting means, It is detected as the level of the high frequency component.

  The noise level detection means accumulates the number of blocks corresponding to the level of the high frequency component detected by the high frequency component detection means for each level by the high frequency component accumulation means. Note that the high-frequency component of a block composed of a flat portion with little change in pixel value is almost “0” if there is no noise, so the high-frequency component of the flat block can be regarded as noise.

  The noise level detection means determines the noise level as an average value corresponding to a predetermined number of blocks from the low level value of the high frequency component level accumulated by the high frequency component accumulation means and outputs the noise level. To do. The low level value may be the minimum value of the high-frequency component level, or a value that is larger than the minimum value by a predetermined level may be used. Thereby, the high-frequency component of the flat block can be detected with high accuracy, and the noise level can be determined.

  Furthermore, the noise reduction device according to claim 6 is the noise reduction device according to any one of claims 2 to 4, wherein the noise level detection means includes a line extraction means and an inter-line noise level detection means. And a level accumulating unit and a noise level determining unit.

  According to this configuration, the noise level detection unit sequentially extracts the target line that is a line constituting the screen and the adjacent line adjacent to the target line as a correlation line group from the digital video signal by the line extraction unit. The correlation line group (target line and adjacent line) may be a horizontal line or a vertical line with respect to the screen.

  The noise level detection means averages the pixel value differences of the corresponding pixels between the target line and the adjacent line in the correlation line group extracted by the line extraction means by the inter-line noise level detection means. Here, the corresponding pixel means a pixel at the same horizontal position when the line is selected horizontally, and a pixel at the same vertical position when the line is selected vertically.

  Note that the difference between the pixel values of adjacent lines has a smaller image component and a larger uncorrelated noise component. In this case, the uncorrelated noise is usually about twice (3 dB) higher than the image component. Therefore, the inter-line noise level detection means detects the average value of the pixel value difference between the target line and the adjacent line as the inter-line noise level in the correlation line group of the target line. Here, the adjacent line is one of the target lines on the screen (up or down for a horizontal line, right or left for a vertical line), or both. May be.

  The noise level detection means accumulates the number of lines corresponding to the inter-line noise level detected by the inter-line noise level detection means for each inter-line noise level. Of the accumulated inter-line noise levels, the smaller the level is, the more noise can be regarded as noise generated by uncorrelated noise.

  The noise level detection means determines and outputs an average value corresponding to a predetermined number of lines as a noise level from the low level value of the inter-line noise level accumulated by the level accumulation means by the noise level determination means. . The low level value may be the minimum value of the noise level between lines, or a value that is larger than the minimum value by a predetermined level may be used. As a result, the uncorrelated noise component can be detected with high accuracy, and the noise level can be determined.

  In order to solve the above-described problem, a noise reduction program according to a seventh aspect includes a computer, a median filter, a pixel value detection unit, a pixel set value calculation unit, in order to reduce noise superimposed on a digital video signal. And functioning as noise reduction means.

  According to such a configuration, the noise reduction program preliminarily sets a pixel at the same screen position as a target pixel in the frame constituting the input digital video signal and the frames before and after this frame by the median filter, A median value in a plurality of target pixels is selected. And a noise reduction program detects the pixel value of the object pixel of a flame | frame by a pixel value detection means.

  In addition, the noise reduction program determines whether or not the absolute value of the difference between the median values selected by the median filter from the pixel values detected by the pixel value detection means is greater than a preset threshold value by the pixel setting value calculation means. In addition, when it is not determined that the absolute value is larger than the threshold value, a predetermined pixel setting value is calculated. This threshold value is preset according to the level value of the noise level superimposed on the digital video signal. For example, when the level value is high, the threshold value is set to a large value, and when the level value is low, the threshold value is set. Set to a smaller value.

  Further, the noise reduction program, when the pixel setting value calculation unit determines that the absolute value is larger than the threshold value by the noise reduction unit, uses the pixel value as an output signal, and does not determine that the absolute value is larger than the threshold value. In the case, the pixel set value is output as an output signal. Here, when it is determined that the absolute value is larger than the threshold value, it is considered that the target pixel is set to a region where the pixel value of an edge or the like suddenly changes, that is, a large movement, for example. The program uses the pixel value as an output signal. As a result, the noise reduction program performs noise reduction processing in a region with little motion while reducing noise in a region with large motion in the digital video signal.

According to the present invention, the following excellent effects can be obtained.
According to the first, second, and seventh aspects of the present invention, noise reduction processing is performed in a portion with little motion in the digital video signal, and noise reduction processing is performed less in a portion with large motion. It is possible to suppress image quality deterioration in the signal.

In the invention according to claim 3, small noise included in the digital video signal can be reduced, and the image quality can be further improved.
In the invention according to claim 4, since an appropriate pixel setting value is calculated and small noise included in the digital video signal is reduced based on the pixel setting value, the image quality is further improved and the noise is excessively reduced. There is nothing.

  According to the fifth and sixth aspects of the present invention, it is possible to detect the noise level superimposed on the digital video signal even if the digital video signal has no blanking period, and to accurately detect noise from the digital video signal. Can be removed.

It is a block diagram which shows the structure of the noise reduction apparatus which concerns on 1st Embodiment of this invention. It is a figure explaining selection of a median value and detection of a pixel value in the present invention. It is a figure explaining the correspondence of the pixel value and shift amount in this invention. It is a flowchart which shows operation | movement of the noise reduction apparatus of FIG. It is a figure which shows the example of the image which the noise reduction apparatus of FIG. 1 performed the noise reduction process. It is a block block diagram which shows the structure of the noise reduction apparatus which concerns on 2nd Embodiment of this invention. It is a block block diagram which shows the structure of a noise level detection apparatus (1st example). It is a schematic diagram which shows the area | region which selects the block in a screen, Comprising: (a) is a figure which shows the case where it selects with the clearance gap between blocks, (b) selects the whole screen as a block. It is a figure which shows the relationship between the pixel value of a block, and the frequency component by orthogonal transformation. It is typical explanatory drawing for demonstrating the method to detect the level of a high frequency component from the frequency component of a block. It is a figure which shows the example of a pattern for demonstrating the area | region of the frequency component used as the object in order to detect the level of a high frequency component. It is the graph which represented the cumulative value of the block for every level of a high frequency component as a histogram. It is a figure which shows the example which extracts a block over a some field or frame. It is a flowchart which shows operation | movement of a noise level detection apparatus (1st example). It is a block block diagram which shows the structure of a noise level detection apparatus (2nd example). It is explanatory drawing for demonstrating the extraction method of the correlation line group of 2 lines in a line extraction means. It is typical explanatory drawing for demonstrating the method of detecting the noise level between lines from the pixel value of an adjacent line. It is the graph which represented the total value of the number of lines for every noise level between lines as a histogram. It is explanatory drawing for demonstrating the extraction method of the correlation line group of 3 lines in a line extraction means. It is a flowchart which shows operation | movement of a noise level detection apparatus (2nd example). It is a figure which shows the example of the image degradation in a prior art.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings as appropriate. In each embodiment, means having the same function and the same member are denoted by the same reference numerals, and description thereof is omitted.

(First embodiment)
[Configuration of noise reduction device]
FIG. 1 is a block diagram showing a configuration of a noise reduction apparatus according to the first embodiment of the present invention. The noise reduction device 1 is for reducing noise from a digital video signal on which noise is superimposed (hereinafter sometimes simply referred to as a video signal). Here, the noise reduction device 1 is connected to a noise level detection device 2.

  The noise level detection device (noise level detection means) 2 detects the noise level superimposed on the video signal as a level value indicating the magnitude of the noise level. Here, it is assumed that the noise level detection device 2 detects the level value of the noise level superimposed on the digital video signal in units of frames. The noise level detection device 2 will be described in detail later. The noise level detecting device 2 corresponds to the noise level detecting means described in the claims.

Hereinafter, the configuration of the noise reduction device 1 will be described in detail.
As shown in FIG. 1, the noise reduction device 1 reduces noise superimposed on a digital video signal. The noise reduction device 1 has an input terminal (video signal input means) T _IN1 , an input terminal T _IN2, and a median filter 11. A pixel value detection unit 12, a threshold setting unit 13, a pixel set value calculation unit 14, a noise reduction unit 15, and an output terminal (video signal output unit) T _OUT .

The input terminal (video signal input means) _TIN1 is for inputting a digital video signal. The input terminal T _IN1 outputs the input digital video signal to the median filter 11 and the pixel value detection means 12. The input terminal _TIN1 corresponds to the video signal input means described in the claims.

The median filter 11 preliminarily sets a pixel at the same screen position as a target pixel in a frame constituting the digital video signal input at the input terminal _TIN1 and a frame before and after this frame, and a plurality of target pixels Select the median (median value).

The pixel value detection means 12 detects the pixel value of the target pixel of the frame constituting the digital video signal input at the input terminal _TIN1 . The pixel value is, for example, the luminance value of the target pixel.

<Selection of median value and detection of pixel value>
Hereinafter, the selection of the median value by the median filter 11 and the detection of the pixel value by the pixel value detection means 12 will be described in detail with reference to FIG. 2 (see FIG. 1 as appropriate). FIG. 2 is a diagram for explaining median value selection and pixel value detection in the present invention.

  Here, FIG. 2 shows three consecutive frames, frame # N−1, frame #N, and frame # N + 1. In this example, frame # N−1 and frame # N + 1 are the frames before and after frame #N. The target pixel p is set at the same screen position in each of the frame # N−1, the frame #N, and the frame # N + 1. Note that the screen position of the target pixel p is not particularly limited, and can be set to any screen position such as the edge or center of the screen.

  First, the median filter 11 acquires a pixel value (for example, 100) in the target pixel p of the frame #N. In addition, the median filter 11 acquires a pixel value (for example, 105) in the target pixel p of the frame # N-1. Then, the median filter 11 acquires a pixel value (for example, 150) in the target pixel p of the frame # N + 1.

  The median filter 11 selects a median value from the three pixel values in the target pixel p of the frame # N−1, the frame #N, and the frame # N + 1. In this example, since the pixel values of the frame # N−1, the frame #N, and the frame # N + 1 are 100, 105, and 150, respectively, the median filter 11 selects 105 as the median value from these, and this median value is selected. To the pixel set value calculation means 14. In addition, although the example which selects the median value from 3 frames was demonstrated in FIG. 2, the median filter 11 may select a median value from an odd number of 5 or more frames.

  Next, the pixel value detection unit 12 detects the pixel value (for example, 100) of the target pixel p in frame #N. Further, the pixel value detection unit 12 outputs the detected pixel value to the pixel set value calculation unit 14. Then, a pixel setting value calculation unit 14 described later calculates a pixel setting value using a preset threshold value, a median value output from the median filter 11, and a pixel value detected by the pixel value detection unit 12. .

Hereinafter, returning to FIG. 1, the description of the configuration of the noise reduction apparatus 1 will be continued.
The input terminal (noise level input means) _TIN2 is used to input the level value of the noise level detected by the noise level detection device 2. Then, the input terminal T _IN2 outputs the input level value to the threshold setting means 13.

The threshold value setting means 13 sets a threshold value to be described later based on the level value input from the input terminal _TIN2 . In the present embodiment, the threshold setting means 13 stores a table or function defining an association in which the threshold value increases as the level value increases and the threshold value decreases as the level value decreases. It is stored in advance in storage means such as “Only Memory”. Then, the threshold setting unit 13 may refer to this table or function, set a threshold based on the input level value, and output the threshold to the pixel setting value calculation unit 14.

  The pixel setting value calculation means 14 determines whether or not the absolute value of the difference between the median values selected by the median filter 11 from the pixel values detected by the pixel value detection means 12 is larger than the threshold value output by the threshold value setting means 13. To do. In the present embodiment, the pixel setting value calculation unit 14 performs the above-described determination using the following formula (1). In the equation (1), x is a pixel value detected by the pixel value detecting unit 12, β is a median value selected by the median filter 11, and α is a threshold value output by the threshold setting unit 13. In the expression (1), “||” represents an absolute value.

  Here, when it is determined by the expression (1) that the absolute value is larger than the threshold value, the pixel setting value calculating unit 14 reduces the noise and the determination result indicating that the absolute value is larger than the threshold value and the pixel value. Output to means 15. On the other hand, when the absolute value is not determined to be larger than the threshold value by the expression (1), the pixel setting value calculating unit 14 calculates the pixel setting value by a method described later. Then, the pixel setting value calculation unit 14 outputs a determination result indicating that the absolute value is not greater than the threshold value and the calculated pixel setting value to the noise reduction unit 15.

<First Example of Calculation of Pixel Setting Value>
Hereinafter, the calculation of the pixel setting value will be specifically described.
As a first example, the pixel setting value calculation means 14 outputs the calculation of the following equation (2), that is, the median value selected by the median filter 11 as the pixel setting value. This first example is a very simple calculation, and the noise reduction device 1 can perform noise reduction processing at higher speed. In Equation (2), x ′ is a pixel setting value.

<Second Example of Calculation of Pixel Setting Value>
As a second example, the pixel set value calculation means 14 outputs the calculation result of the following formula (3) as a set value. In the expression (3), sign is a function that returns “1” if the argument is positive, “−1” if the argument is negative, and “0” if the argument is “0”. . In the equation (3), sft is a shift amount described later.

  Thus, in the above-described equation (3), the pixel setting value calculation unit 14 determines whether or not the pixel value is larger than the median value. Then, when it is determined that the pixel value is larger than the median value, the pixel set value calculation unit 14 calculates the pixel set value by subtracting the shift amount from the pixel value. On the other hand, when it is not determined that the pixel value is larger than the median value, the pixel setting value calculation unit 14 adds the pixel value and the shift amount to calculate the pixel setting value. When the pixel value is equal to the median value, the pixel value and the pixel setting value are equal in the calculation result of Expression (3). Therefore, the pixel setting value calculation unit 14 may not calculate the pixel setting value. good.

Since the shift amount needs to be calculated in order to use equation (3), the pixel set value calculation unit 14 includes a shift amount calculation unit 14a.
The shift amount calculation unit 14a has a correspondence relationship between the pixel value detected by the pixel value detection unit 12 and the shift amount in advance, and the shift amount corresponding to the pixel value detected by the pixel value detection unit 11 based on the correspondence relationship. Is calculated.

  Here, as a correspondence relationship between the pixel value and the shift amount, as shown in FIG. 3A, a shift amount having a different value for each pixel section with respect to a pixel section obtained by dividing continuous pixel values at a constant interval. Are associated with each other. Here, for example, the shift amount is 10 in the pixel interval where the pixel value is 0 or more and less than 16, the shift amount is 20 in the pixel interval where the pixel value is 16 or more and less than 64, and the shift amount is in the pixel interval where the pixel value is 64 or more and less than 128. Is 30 and the pixel value of 128 or more and less than 255 is such that the shift amount is 40, such that the shift amount increases as the pixel value of the pixel interval increases.

  In addition, as a correspondence relationship between the pixel value and the shift amount, there is a function that returns a larger shift amount as the pixel value increases. FIG. 3B is a plot of the output of this function. According to this function, it can be seen that the shift amount increases as the pixel value increases. That is, the shift amount calculation unit 14a calculates the shift amount using the correspondence relationship shown in FIG. 3A or the correspondence relationship (function) shown in FIG. Note that FIG. 3A and FIG. 3B are examples of correspondence relationships, and it goes without saying that the present invention is not limited to these.

Hereinafter, returning to FIG. 1, the description of the configuration of the noise reduction apparatus 1 will be continued.
The noise reduction unit 15 uses the pixel value as an output signal when the pixel setting value calculation unit 14 determines that the absolute value is greater than the threshold value, and sets the pixel setting when the absolute value is not determined to be greater than the threshold value. A value is output as an output signal. In the present embodiment, when the determination result indicating that the absolute value is larger than the threshold value is input from the pixel setting value calculation unit 14, the noise reduction unit 15 uses the pixel value as an output signal. On the other hand, when the determination result indicating that the absolute value is not larger than the threshold value is input from the pixel setting value calculation unit 14, the noise reduction unit 15 uses the calculated pixel setting value as an output signal.

The output terminal (video signal output means) T _OUT outputs an output signal from the noise reduction means 15.

[Operation of noise reduction device]
Hereinafter, the operation of the noise reduction device of FIG. 1 will be described with reference to FIG. 4 (see FIG. 1 as appropriate). FIG. 4 is a flowchart showing the operation of the noise reduction apparatus of FIG. In FIG. 4, it is assumed that a digital video signal is input to the noise reduction device 1 in advance and a threshold value is set.

First, the noise reduction apparatus 1 uses a median filter 11 in advance as a target pixel with pixels at the same screen position in a frame constituting a digital video signal input at the input terminal _TIN1 and in the frames before and after this frame. Set and select the median value for a plurality of target pixels (step S1).

Following the processing of step S1, the noise reduction apparatus 1 detects the pixel value of the target pixel of the frame constituting the digital video signal input at the input terminal _TIN1 by the pixel value detection means 12 (step S2). In addition, the process of step S1 and the process of step S2 may perform the process of step S1 after the process of step S2 regardless of the order.

  Subsequent to the processing in step S2, the noise reduction apparatus 1 uses the pixel set value calculation unit 14 to calculate the median filter selected by the median filter 11 from the pixel values detected by the pixel value detection unit 12 according to the above-described equation (1). It is determined whether or not the absolute value of the difference is greater than a threshold value (step S3).

  When it is not determined that the absolute value is larger than the threshold value (Yes in step S3), the noise reduction device 1 uses the pixel setting value calculation unit 14 to perform, for example, the method of the first example or the second example described above. The pixel setting value is calculated (step S4). In the noise reduction device 1, the noise reduction unit 15 outputs the pixel setting value calculated by the pixel setting value calculation unit 14 as an output signal (step S5).

  On the other hand, when it is determined that the absolute value is larger than the threshold value (No in step S3), the noise reduction device 1 outputs the pixel value as an output signal by the noise reduction unit 15 (step S6). Note that, when the noise reduction device 1 performs the above-described noise reduction processing for a certain target pixel, the noise reduction device 1 may perform the above-described noise reduction processing by resetting the screen position of the target pixel to a different location. Furthermore, the noise reduction apparatus 1 may repeat the resetting of the target pixel and perform noise reduction processing on all the pixels of the frame.

  As described above, according to the noise reduction device 1 according to the present embodiment, noise reduction processing is performed in a portion with little motion, and noise reduction processing is performed less in a portion with large motion, so that the movement of the subject is large. It is possible to suppress image quality deterioration in a digital video signal. For example, even when a portion with a large movement such as the left arm or right foot is included in the digital video signal, as shown in FIG. 5, the noise reduction device 1 rarely performs noise reduction processing on the portion with a large movement. Compared with FIG. 21, the occurrence of image degradation can be suppressed.

  Note that the noise reduction apparatus according to the present invention can also be realized by a program that causes hardware resources such as a general computer arithmetic unit and a storage device to operate cooperatively as each unit of the noise reduction apparatus 1 of FIG. This program can be distributed on a CD-ROM or the like and can be downloaded via the Internet.

The configuration of the noise reduction device according to the present invention is not limited to FIG. For example, when an operator or the like manually calculates and sets the threshold value, the input terminal _TIN2 and the threshold value setting unit 13 can be removed from the noise reduction apparatus 1 of FIG. On the other hand, it is good also as a structure provided with the noise level detection apparatus 2 inside the noise reduction apparatus 1 of FIG. Hereinafter, this configuration will be described as a second embodiment.

(Second Embodiment)
[Noise reduction device]
With reference to FIG. 6, the structure of the noise reduction apparatus which concerns on 2nd Embodiment of this invention is demonstrated. FIG. 6 is a block diagram showing the configuration of the noise reduction apparatus according to the second embodiment of the present invention. Here, the noise reduction apparatus 1C includes an input terminal (video signal input means) _TIN1 , a median filter 11, a pixel value detection means 12, a threshold value setting means 13, a pixel set value calculation means 14, and noise reduction means. 15, a noise level detection device 2, and an output terminal (video signal output means) T _OUT .

The noise level detection device 2 inputs a video signal via an input terminal (video signal input means) _TIN1 , detects the noise level, and then outputs the noise level (level value) to the threshold setting means 13.
The operation of the noise reduction device 1C is basically the same as the operation of the noise reduction device 1 (see FIG. 1) described in FIG.

  Accordingly, the noise reduction device 1C can generate a video signal in which noise is reduced from the video signal with a configuration of one input and one output.

[Configuration of Noise Level Detection Device (Means)]
Next, the configuration of the noise level detection device 2 and the noise level detection device 2 described in FIGS. 1 and 6 will be described in detail.
The noise level detection device 2 detects a noise level from a video signal on which noise is superimposed. Here, as an example of the noise level detection device (means) 2, two configurations will be described as examples.

In the first example, the noise level detection device 2 searches for a flat portion in which at least a change in pixel value is small for each predetermined block in a frame (screen) constituting an image, and is superimposed on that portion. The detected signal component is detected as a noise level.
In the second example, the noise level detection device 2 detects the noise level superimposed in the video signal based on the correlation of the lines in the frame (screen) constituting the video.

(First example: configuration of noise level detection device [block type noise level detection])
First, the configuration of the noise level detection apparatus 2 according to the first example will be described with reference to FIG. FIG. 7 is a block diagram showing the configuration of the noise level detection apparatus according to the first example.
Usually, a video screen (image) has a flat portion with little change in pixel value, for example, an infinite region such as the sky, a region where the camera is not in focus, or the like. These areas are areas where the frequency components of the image have few high frequency components. On the other hand, noise is a high-frequency component with fine particles. That is, in the screen, the high frequency component in the region with a low high frequency component can be regarded as being superimposed by noise.

Therefore, the noise level detection device 2 detects the noise level from the high frequency component in the region where the high frequency component is small in the screen.
Here, the noise level detection device 2 includes blocking means 21, orthogonal transform means 22, high frequency component detection means 23, high frequency component accumulation means 24, and noise level determination means 25.

  The blocking unit 21 converts the video signal into data for each block having a predetermined size. The blocking unit 21 receives a video signal from the outside, and outputs the divided blocks to the orthogonal transform unit 22. Here, the blocking unit 21 includes a block dividing unit 21a and a block selecting unit 21b.

  The block dividing means 21a divides the video signal into blocks of a predetermined size for each screen, that is, for each field or frame constituting the screen. Although the size of this block is not particularly limited, for example, the block is divided into sizes of about 4 × 4 pixels to 128 × 128 pixels.

  The block selection means 21b is for partially selecting the blocks divided by the block dividing means 21a so that the ratio is equal to or higher than a predetermined ratio with respect to the entire screen in the screen. The block selected by the block selection unit 21 b is output to the orthogonal transformation unit 22. For example, as shown in FIG. 8A, the block selection unit 21b provides a gap for each block B in the screen D and covers an area of 50% or more (more preferably 70% or more) of the entire screen. A plurality of blocks B, B,. As a result, a block including a region that is a candidate for a flat portion with a small change in pixel value in the screen is selected.

  Here, the block selection means 21b partially selects the blocks so that the ratio is equal to or higher than a predetermined ratio from the screen. However, as shown in FIG. It is good also as selecting. In this case, the block selecting unit 21b is omitted from the configuration, and all the blocks divided by the block dividing unit 21a are output to the orthogonal transform unit 22.

The orthogonal transform unit 22 performs orthogonal transform on the block output from the blocking unit 21 for each block. This orthogonal transform means 22 transforms each block into a frequency component by performing orthogonal transform on each block. This frequency component is output to the high frequency component detection means 23. For example, the orthogonal transform unit 22 transforms a block into a frequency component by a discrete cosine transform (DCT) which is a kind of orthogonal transform. Accordingly, as shown in FIG. 9, for example, the pixel values (p _{11 to} p ₁₄ , p _{21 to} p ₂₄ , p _{31 to} p ₃₄ , p _{41 to} p ₄₄ ) of the block B of 4 × 4 pixels are represented by pixels. It is converted into a value (DCT coefficient; x _{11 to} x ₁₄ , x _{21 to} x ₂₄ , x _{31 to} x ₃₄ , x _{41 to} x ₄₄ ) indicating the same number of frequency components as the number. The DCT coefficients in the figure, the upper left coefficient x ₁₁ represents a DC component, lower right, it becomes higher frequency toward a lower left direction.

  In addition, as long as this orthogonal transformation transforms the discrete signal of a block into a frequency domain, you may use transformations other than DCT transformation. For example, discrete sine transform (DST), fast Fourier transform (FFT), Hadamard transform, or the like can be used. Thus, by converting the discrete signal of a block into the frequency domain, it becomes possible to analyze which frequency component each block has.

  The high frequency component detection means 23 averages the coefficients of the high frequency components excluding the DC component of the block converted by the orthogonal transform means 22 and the predetermined vicinity of the DC component in the block, and obtains the level of the high frequency component of the block. It is to detect.

  In the orthogonally transformed frequency component, the low frequency component including the direct current component is a component including a large amount of the frequency component of the image signal. Therefore, here, the high frequency component detection means 23 sets the coefficient of the low frequency component to “0” and averages the coefficient of the high frequency component in the other region within the block, thereby obtaining the level of the high frequency component of the block. Will be considered. This averaging can be calculated by the square root of the mean square sum of the coefficients of each high-frequency component, the absolute value average, or the like.

  Here, with reference to FIG. 10 (refer to FIG. 7 as appropriate), a method in which the high frequency component detection means 23 detects the level of the high frequency component from the frequency component of the block will be described. FIG. 10 is a schematic explanatory diagram for explaining a method of detecting the level of the high frequency component from the frequency component of the block. FIG. 10 (a) shows a frequency coefficient (corresponding to FIG. 9 (b)) of a certain block obtained by the orthogonal transform means 22, and FIG. 10 (b) shows a state in which only a high frequency component is extracted from the frequency component. Show. Here, in order to simplify the description, the block size is assumed to be 4 × 4 pixels.

As shown in FIG. 10A, it is assumed that a certain block is orthogonally transformed by the orthogonal transformation means 22 and the same number of coefficients as the number of pixels of the block (here, 4 × 4) are obtained.
At this time, the high frequency component detecting means 23 regards the DC component and the coefficient of the horizontal frequency component including the DC component and the coefficient of the vertical frequency component as “0” as the coefficients in the vicinity of the DC component (FIG. 10B). )), And extract other coefficients as high-frequency component coefficients. That is, the coefficients x ₁₁ , x ₁₂ , x ₁₃ , x ₁₄ , x ₂₁ , x ₃₁ , x ₄₁ are regarded as “0”, and the other coefficients x ₂₂ , x ₂₃ , x ₂₄ , x ₃₂ , x ₃₃ , x _{34 are considered.} , X ₄₂ , x ₄₃ , x ₄₄ are extracted as high frequency component coefficients.

  Then, the high frequency component detection means 23 averages each coefficient extracted as a high frequency component coefficient within the block, thereby obtaining the level of the high frequency component of the block. Here, when averaging is performed using the square root of the mean square sum, the high frequency component detecting means 23 calculates the level L of the high frequency component according to the following equation (4).

  Alternatively, in the case of averaging by absolute value averaging, the high frequency component detecting means 23 calculates the level L of the high frequency component by the following equation (5).

  In the above equations (4) and (5), the coefficient 16/9 is a normalization coefficient, and is a coefficient for normalizing nine high-frequency component coefficients to 16 coefficients of the entire block. It is. By normalizing in this way, the level of the high frequency component for the entire block can be obtained.

Here, as the DC component and the coefficient in the vicinity of the DC component, the horizontal and vertical frequency components including the DC component are regarded as “0”, and the level of the high frequency component of the block is calculated, but the coefficient regarded as “0” Various patterns can be used as long as they contain a DC component. For example, as shown in pattern A of FIG. 11A, the high frequency component detection means 23 positions the coefficient of the minimum vertical frequency component and the maximum horizontal frequency component and the position of the coefficient of the maximum vertical frequency component and the minimum horizontal frequency component. Are divided into two regions R ₁ and R ₂ , the coefficient of the region R _{1 including} the DC component is regarded as “0”, and the level of the high frequency component is calculated by the coefficient of the other region R ₂ To do.

Further, for example, as shown in FIG. 11B, the diagonal line shown in FIG. 11A may be curved to include more high-frequency components and divided into two regions R ₁ and R _2. . As a result, more high-frequency components in the image can be extracted. Also, for example, as shown in FIG. 11C, a region R ₁ partially including a coefficient on the DC component side is compared to a region on the high frequency component side excluding the horizontal frequency component including the DC component and the vertical frequency component. it may be classified blocks in the region R ₂ otherwise. As a result, among the high frequency components in the image, it is possible to remove the components close to the DC component and extract the high frequency components.
Returning to FIG. 7, the description of the configuration of the noise level detection apparatus will be continued.

  The high frequency component accumulating means 24 accumulates the number of blocks corresponding to the level of the high frequency component detected by the high frequency component detecting means 23 for each level. The high-frequency component accumulating unit 24 accumulates (histograms) the number of blocks corresponding to the level of the high-frequency component for each field or frame unit constituting the screen (see FIG. 12). As a result, it is analyzed which level of which high-frequency component is distributed in the screen. The accumulated value for each level accumulated by the high frequency component accumulating unit 24 is output to the noise level determining unit 25.

The noise level determination unit 25 determines a low level value of the high frequency component level accumulated by the high frequency component accumulation unit 24 or an average value corresponding to a predetermined number of blocks from the low level value as a noise level. is there. As described with reference to FIG. 10, the coefficients detected by the high frequency component detection means 23 as the high frequency components (for example, the coefficients x ₂₂ , x ₂₃ , x ₂₄ , x ₃₂ , x ₃₃ , x ₃₄ , x ₄₂ , x ₄₃ , x ₄₄ ) is almost “0” if it is a flat part with little change in the pixel value of the screen, so if noise is superimposed on the flat part, It can be said that the high frequency component is noise. Therefore, the noise level determination unit 25 is a low level value of the level of the high frequency component or a predetermined number (for example, 2 to 5) of the block having the high frequency component accumulated by the high frequency component accumulation unit 24. The average value (weighted average value) according to the number of blocks is determined as the noise level.

  Here, the minimum value of the high frequency component level accumulated by the high frequency component accumulating means 24 is used as the low level value of the high frequency component level. Note that the noise level of an image captured by a CCD or the like may differ depending on the brightness of the screen. When the minimum value of the high-frequency component level is used, the noise level may be determined at a low level. Therefore, the low level value is not necessarily limited to the minimum value, and a value larger than the minimum value within a predetermined range from the minimum value may be used.

  Here, the noise level determination method of the noise level determination means 25 will be described with reference to FIG. FIG. 12 is a graph showing a cumulative value (frequency) of blocks for each level of high-frequency components as a histogram. Note that this graph is a diagram for visually explaining the accumulated value of the block, and does not mean that the high-frequency component accumulating unit 24 generates the graph.

Usually, the high frequency component includes the high frequency component of the image and noise at the same time. At this time, in the graph of FIG. 12, the histogram on the right from the center of the graph having a large high-frequency component is considered to be the high-frequency component of the video (including noise). On the other hand, several histograms at the left end of the graph are considered to be almost noise-only components that do not include high-frequency components of video.
Therefore, the noise level determination means 25 calculates an average value from cumulative values corresponding to several histograms on the left end of the graph and determines it as a noise level.

For example, as shown in FIG. 12, when determining the noise level from three histograms, the noise level determination means 25 selects L ₁ , L ₂ , L ₃ with low levels of high frequency components, and the respective frequencies. An average value (weighted average value) is obtained from C ₁ , C ₂ , and C ₃ by the following equation (6), and set as a noise level N _L.

Here, the noise level determination means 25 calculates the noise level by the weighted average of the levels of a plurality of high frequency components, but the noise level may be determined by the arithmetic average of the levels of the high frequency components without considering the frequency. Good. Further, the noise level determination means 25 may simply set the minimum high frequency component level (L _{1 in} the case of FIG. 12) as the noise level.

  And the noise level determination means 25 outputs the determined noise level (level value) to the noise reduction apparatus 1 (FIG. 1). When the noise level detection device 2 is incorporated in the noise reduction device 1, the noise level determination unit 25 uses the determined noise level (level value) as the threshold setting unit 13 of the noise reduction device 1C (FIG. 6). Output to.

  Here, the example in which the noise level detection device 2 detects the noise level of the video signal in units of screens (in the spatial direction) has been described. However, as shown in FIG. 13, a plurality of screens (a plurality of fields or frames) are detected. ) (Space + time direction). That is, the noise level detecting device 2 accumulates blocks corresponding to the level of the high frequency component for a predetermined time (a time corresponding to a plurality of fields and a plurality of frames) in the high frequency component accumulating unit 24, and the noise level determining unit 25. Thus, the noise level may be determined based on the accumulated result.

  Further, here, the noise level detection device 2 uses the low level value of the level or the low level value in the histogram of the block for each level of the high frequency component accumulated by the high frequency component accumulation unit 24 by the noise level determination unit 25. An average value (such as a weighted average) according to the predetermined number of blocks is determined as the noise level. However, the noise level determination means 25 is simply a low level value (minimum value) of a high frequency component level detected by the high frequency component detection means 23 or a predetermined number of times from the low level value, regardless of the number of blocks. The average value of the level values may be determined as the noise level. In this case, the noise level detection device 2 can be configured by omitting the high-frequency component accumulating means 24 from the configuration.

(First Example: Operation of Noise Level Detection Device)
Next, the operation of the noise level detection apparatus 2 according to the first example will be described with reference to FIG. FIG. 14 is a flowchart showing the operation of the noise level detection apparatus according to the first example.

First, the noise level detection apparatus 2 divides the video signal into blocks of a predetermined size (for example, 4 × 4 pixels to 128 × 128 pixels) by the block dividing unit 21a of the blocking unit 21 (step S11). .
Furthermore, the noise level detection apparatus 2 uses the block selection unit 21b of the blocking unit 21 so that the block divided in step S11 becomes a predetermined ratio (for example, 70%) or more with respect to the entire screen in the screen. Thus, a partial selection is made (step S12). The block selected in this way includes a block including a flat portion in which a change in pixel value is small in the screen.

  And the noise level detection apparatus 2 converts each block into a frequency component by orthogonal transformation for every block selected by step S12 by the orthogonal transformation means 22 (step S13). Thereafter, the noise level detection device 2 uses the high frequency component detection means 23 to average the coefficients of the high frequency components excluding the direct current component and the predetermined vicinity of the direct current component in the frequency component obtained in step S13. The high frequency component level of the block is detected (step S14). As a result, the noise level detection device 2 detects both the high-frequency component level of the video on which the noise is superimposed and the high-frequency component level generated only by the noise.

  Then, the noise level detection apparatus 2 accumulates the number of blocks corresponding to the level of the high frequency component detected in step S14 by the high frequency component accumulating unit 24 for each level (step S15). Thereby, the noise level detection apparatus 2 can analyze the distribution of the level of the high frequency component of each block. At this time, it can be said that the block having a low level of the high frequency component generates the high frequency component only by noise.

Therefore, the noise level detection device 2 uses the noise level determination unit 25 to obtain a minimum value (low level value) of the high-frequency component level accumulated in step S15 or a predetermined number (for example, 2 to 5) from the minimum value. The average value corresponding to the number of blocks is determined as the noise level in the screen (step S16).
Thereby, the noise level detection apparatus 2 can detect the noise level superimposed on the video signal from the video itself.

(Second Example: Configuration of Noise Level Detection Device [Line Correlation Noise Detection])
Next, the configuration of the noise level detection apparatus 2 according to the second example will be described with reference to FIG. FIG. 15 is a block diagram showing the configuration of the noise level detection apparatus according to the second example.
Normally, in a video screen (image), the difference between pixel values of adjacent lines has a small image component and a large uncorrelated noise component. Therefore, the smaller the pixel value difference is, the more the difference can be regarded as noise generated by uncorrelated noise.

  Therefore, the noise level detection device 2 detects the noise level from the difference between the pixel values of adjacent lines in the screen. Here, the noise level detection device 2 includes a line extraction unit 26, an inter-line noise level detection unit 27, a level accumulation unit 28, and a noise level determination unit 29.

  The line extraction unit 26 correlates a target line, which is a line constituting a screen from a video signal, and an adjacent line adjacent to the target line along a scanning line with a correlation line group (more specifically, for each correlation line group). Pixel). Here, it is assumed that the line extraction unit 26 partially extracts lines so that the ratio is equal to or higher than a predetermined ratio with respect to the entire screen in the screen. The pixel value of each pixel in the line extracted by the line extraction unit 26 is output to the inter-line noise level detection unit 27.

Here, with reference to FIG. 16 (refer to FIG. 15 as appropriate), the lines (correlation line group) extracted by the line extraction means 26 will be described. FIG. 16 is an explanatory diagram for explaining a method of extracting a correlation line group of two lines in the line extraction unit.
As shown in FIG. 16 (a), the line extraction means 26 is an Nth line (target line) and an (N + 1) th line (adjacent line) in the main scanning unit of the screen D (two-dimensional image). Two adjacent lines are sequentially extracted as a correlation line group. At this time, the line extraction unit 26 provides a gap for each adjacent correlation line group, and extracts a plurality of correlation line groups so as to cover an area of 50% or more (more preferably 70% or more) of the entire screen D. . Thereby, the line extraction unit 26 can extract a correlation line group in which a change in pixel values between lines is small in the screen D.

16A shows an example in which the line extracting unit 26 extracts a line corresponding to the horizontal width of the screen as a line. However, as shown in FIG. It is good also as extracting a part of short line partially.
Furthermore, as shown in FIG. 16C, the line extraction unit 26, on a screen (frame) continuous in the time direction, is a line at the same horizontal position, and the N _tth line of the screen at time t, The _{Nt +} 1th line on the screen at time (t + 1) may be extracted as a correlation line group. Also in this case, as in FIG. 16B, the line need not necessarily use the horizontal width of the screen, and may be partially extracted.

In addition, here, an example is shown in which lines are extracted in the horizontal direction along the scanning lines, but lines may be extracted in the vertical direction.
Here, the line extraction means 26 partially extracts the correlation line group from the screen so that the ratio is equal to or higher than the predetermined ratio. However, the line extraction unit 26 may extract the correlation line group of the entire screen D. Good. Returning to FIG. 15, the description of the configuration of the noise level detection device 2 will be continued.

  The inter-line noise level detection means 27 averages the pixel value differences of the corresponding pixels (here, the pixels at the same horizontal position) for each adjacent correlation line group in the line extracted by the line extraction means 26. The noise level is detected as the inter-line noise level in the correlation line group of the target line. This averaging can be calculated by the square root of the mean square sum of the differences in pixel values in each correlation line group, the absolute value average, or the like.

  Here, with reference to FIG. 17 (refer to FIG. 15 as appropriate), a method in which the inter-line noise level detection means 27 detects the inter-line noise level from the pixel values of adjacent lines will be described. FIG. 17 is a schematic explanatory diagram for explaining a method of detecting an inter-line noise level from pixel values of adjacent lines.

As shown in FIG. 17, it is assumed that a pixel of a certain Nth line and a pixel of the (N + 1) th line adjacent thereto are extracted by the line extraction unit 26. Each line has M pixels in the horizontal direction. Here, the pixel values of the pixels of the Nth line are p _{N, 1} , p _{N, 2} , p _{N, 3} ,..., P _{N, M,} and the pixel values of the pixels of the (N + 1) th line are pN _{+ 1,1} , pN _{+ 1,2} , pN _{+ 1,3} ,..., pN _{+ 1, M.}

  The inter-line noise level detection means 27 averages the pixel value difference of each pixel at the same horizontal position between the N-th line and the (N + 1) -th line, so that the inter-line noise level in the correlation line group Use noise level. Here, when averaging is performed using the square root of the mean square sum, the line-to-line noise level detection means 27 calculates the line-to-line noise level L using the following equation (7).

  Alternatively, in the case of averaging by absolute value averaging, the inter-line noise level detection means 27 calculates the inter-line noise level L by the following equation (8).

Here, the method for detecting the inter-line noise level on the same screen (frame) has been described. However, as shown in FIG. 16C, the inter-line noise level may be detected between adjacent frames. Good. In this case, in the above-described equations (4) and (5), the pixel values p _{N, i} , p _{N + 1, i} only become the pixel values p _{Nt, i} , p _{Nt + 1, i} corresponding to the time t, respectively. Similarly, the noise level between lines can be calculated.
Returning to FIG. 15, the description of the configuration of the noise level detection apparatus will be continued.

  The level accumulating means 28 accumulates the number of lines corresponding to the inter-line noise level detected by the inter-line noise level detecting means 27 for each level. The level accumulating means 28 accumulates the number of lines corresponding to the inter-line noise level for each level in a screen unit (see FIG. 18). Thus, it is analyzed how much noise level between lines is distributed in the screen. The accumulated value for each level accumulated by the level accumulating unit 28 is output to the noise level determining unit 29.

  The noise level determination unit 29 determines a low level value of the inter-line noise level accumulated by the level accumulation unit 28 or an average value corresponding to a predetermined number of lines from the low level value as a noise level. . The inter-line noise level detected by the inter-line noise level detecting means 27 has a smaller image component between adjacent lines, so that the smaller the value, the closer to the noise level. Therefore, the noise level determination means 29 is a predetermined number (for example, 2 to 5) from the minimum value or minimum value of the inter-line noise level in the total number of lines for each inter-line noise level accumulated by the level accumulating means 28. The average value (weighted average value) corresponding to the number of lines is determined as the noise level.

  Here, the minimum value of the inter-line noise level accumulated by the level accumulating means 28 is used as the low level value of the inter-line noise level. In addition, since the minimum value of the noise level between lines may be detected by chance regardless of noise depending on the image, the low level value is not necessarily limited to the minimum value. A value larger than the minimum value may be used within a predetermined range.

  Here, the noise level determination method of the noise level determination means 29 will be described with reference to FIG. FIG. 18 is a graph showing a cumulative value (frequency) of the number of lines for each inter-line noise level as a histogram. This graph is a diagram for visually explaining the cumulative value of the number of lines, and does not mean that the level cumulative unit 28 generates the graph.

The pixel value difference between the lines includes the image component difference and the noise component at the same time. At this time, in the graph of FIG. 18, it is considered that the right histogram from the center of the graph having a large inter-line noise level includes a large difference in image components of video (including noise). On the other hand, several histograms at the left end of the graph are considered to be almost noise-only components that do not include differences in image components of the video.
Therefore, the noise level determination means 29 calculates an average value from the cumulative values corresponding to several histograms on the left end of the graph and determines it as the noise level.

For example, as shown in FIG. 18, when determining the noise level from three histograms, the noise level determination means 29 selects L ₁ , L ₂ , L ₃ with low inter-line noise levels, and the respective frequencies. An average value (weighted average value) is obtained from C ₁ , C ₂ , and C ₃ by the following equation (9), and set as a noise level N _L.

Here, the noise level determination means 29 calculates the noise level by weighted average of a plurality of inter-line noise levels. However, the noise level may be determined by arithmetic average of the inter-line noise levels without considering the frequency. Good. Further, the noise level determination means 29 may simply set the minimum inter-line noise level (L _{1 in} the case of FIG. 18) as the noise level.

  And the noise level determination means 29 outputs the determined noise level (level value) to the noise reduction apparatus 1 (FIG. 1). When the noise level detection device 2 is incorporated in the noise reduction device 1, the noise level determination unit 29 uses the determined noise level (level value) as the threshold setting unit 13 of the noise reduction device 1C (FIG. 6). Output to.

  Here, the noise level detection device 2 sequentially extracts a total of two lines, that is, a target line and one adjacent line as a correlation line group by the line extraction unit 26, and the noise level detection unit 27 between the lines extracts two lines. The noise level between lines was detected from the pixel value of the line. However, the correlation line group is not limited to two lines. For example, the inter-line noise level may be detected by a total of three lines including the target line and two adjacent lines.

  Here, with reference to FIG. 19 (refer to FIG. 15 as appropriate), a method of detecting the inter-line noise level with three lines will be described. FIG. 19 is an explanatory diagram for explaining a method of extracting a three-line correlation line group in the line extraction unit.

  As shown in FIG. 19A, for example, the line extraction means 26 is the main scanning unit of the screen D (two-dimensional image), the Nth line (target line), (N−1) and (N + 1). Three adjacent lines of the first line (adjacent line) are sequentially extracted as a correlation line group. At this time, the line extraction unit 26 provides a gap for each adjacent correlation line group, and extracts a plurality of correlation line groups so as to cover an area of 50% or more (more preferably 70% or more) of the entire screen D. .

Further, as shown in FIG. 19B, a part of lines shorter than the horizontal width of the screen may be partially extracted as three lines.
Further, as shown in FIG. 19 (c), the line extraction unit 26, on a screen (frame) continuous in the time direction, is a line at the same horizontal position, and is the N _tth line (target) at the time t. Line) and N _t-1 and N _{t + 1} second lines (adjacent lines) on the screen at times (t−1) and (t + 1) may be extracted as a correlation line group. Also in this case, as in FIG. 16B, the line need not necessarily use the horizontal width of the screen, and may be partially extracted.

  As described above, when the line extraction unit 26 sequentially extracts three correlation line groups, the noise level detection device 2 uses the inter-line noise level detection unit 27 to select one of the target lines for each corresponding pixel of the correlation line group. The difference between the pixel value of the target line and 1/2 of the pixel value of the line adjacent to the first line (first adjacent line) and the line adjacent to the other (second adjacent line) and the pixel value of the target line is obtained. A value obtained by averaging is used as the noise level between lines.

That is, the inter-line noise level detection means 27 multiplies the pixel values of the corresponding pixels of the three lines by (−1/2, 1, −1/2) or (1/2, −1, 1/2). Then, the difference between the target line and the adjacent line is obtained. Then, the inter-line noise level detection means 27 calculates the inter-line noise level in the correlation line group from the difference by the above-described equation (7) or (7).
In this way, by detecting the noise level with three lines, the image components are dispersed and the noise level can be detected with higher accuracy than when the noise level is detected with two lines.

  Also, here, in the histogram of the number of lines for each inter-line noise level accumulated by the level accumulation means 28 by the noise level determination means 29, the noise level detection device 2 uses the low level value of the level or the low level value. An average value (such as a weighted average) corresponding to the predetermined number of lines is determined as the noise level. However, the noise level determination means 29 is merely a low level value (minimum value) or a predetermined number from the low level value among the noise levels between lines detected by the noise level detection means 27 between lines regardless of the number of lines. The average value of the minute level values may be determined as the noise level. In this case, the noise level detection device 2 can be configured by omitting the level accumulating means 28 from the configuration.

(Second Example: Operation of Noise Level Detection Device)
Next, the operation of the noise level detection apparatus 2 according to the second example will be described with reference to FIG. 20 (refer to FIG. 15 as appropriate for the configuration). FIG. 20 is a flowchart showing the operation of the noise level detection apparatus according to the second example.

  First, the noise level detection apparatus 2 extracts a line (correlation line group) from the video signal along the scanning line by the line extraction unit 26 (step S21). The line extracting unit 26 may extract lines of the entire screen, or may partially extract lines so that a predetermined ratio (for example, 70%) or more with respect to the entire screen in the screen. It is good also as selecting.

  And the noise level detection apparatus 2 averages the difference of the pixel value in the same horizontal position for every adjacent line (correlation line group) in the line selected by step S21 by the noise level detection means 27 between lines, It is detected as a noise level between lines in the correlation line group (step S22). The inter-line noise level is a level in which a difference between image components and a noise component between video lines are mixed.

  Then, the noise level detecting device 2 accumulates the number of lines corresponding to the noise level between lines detected in step S22 by the level accumulating unit 28 for each level in units of screen (step S23). Thereby, the noise level detection device 2 can analyze the distribution of the noise level between lines in the screen. At this time, it can be said that the noise level between lines with a low level is a component of only noise.

Therefore, the noise level detection device 2 uses the noise level determination unit 29 to obtain a minimum value (low level value) of the inter-line noise level accumulated in step S23 or a predetermined number (for example, 2 to 5) from the minimum value. The average value corresponding to the number of lines is determined as the noise level in the screen (step S24).
Thereby, the noise level detection apparatus 2 can detect the noise level superimposed on the video signal from the video itself.

DESCRIPTION OF SYMBOLS 1 Noise reduction apparatus 1C Noise reduction apparatus 11 Median filter 12 Pixel value detection means 13 Threshold setting means 14 Pixel setting value calculation means 15 Noise reduction means 15
2 Noise level detection device (noise level detection means)
DESCRIPTION OF SYMBOLS 21 Blocking means 22 Orthogonal transformation means 23 High frequency component detection means 24 High frequency component accumulation means 25 Noise level determination means 26 Line extraction means 27 Inter-line noise level detection means 28 Level accumulation means 29 Noise level determination means 3 Delay device T _IN1 video input Means T _IN2 noise level input means T _OUT video signal output means

Claims (7)

A noise reduction device for reducing noise superimposed on a digital video signal,
Video signal input means for inputting the digital video signal;
In a frame constituting the digital video signal input by the video signal input means and a frame before and after the frame, a pixel at the same screen position is set in advance as a target pixel, and a median value in the plurality of target pixels is set. A median filter to select,
Pixel value detecting means for detecting a pixel value of the target pixel of the frame;
It is determined whether or not the absolute value of the difference between the median values selected by the median filter from the pixel values detected by the pixel value detection means is greater than a preset threshold value, and the absolute value is greater than the threshold value. A pixel setting value calculating means for calculating a predetermined pixel setting value if it is not determined to be large;
In the pixel setting value calculation means, when the absolute value is determined to be larger than the threshold, the pixel value is used as an output signal, and when the absolute value is not determined to be larger than the threshold, the pixel setting Noise reduction means for outputting a value as an output signal;
A noise reduction device comprising: A noise reduction device for reducing noise superimposed on a digital video signal,
Video signal input means for inputting the digital video signal;
Noise level detection means for detecting a noise level superimposed on the digital video signal input by the video signal input means as a level value indicating the magnitude of the noise level;
Threshold setting means for setting a threshold based on the level value detected by the noise level detecting means;
In a frame constituting the digital video signal input by the video signal input means and a frame before and after the frame, a pixel at the same screen position is set in advance as a target pixel, and a median value in the plurality of target pixels is set. A median filter to select,
Pixel value detecting means for detecting a pixel value of the target pixel of the frame;
It is determined whether the absolute value of the difference between the median values selected by the median filter from the pixel values detected by the pixel value detection unit is larger than the threshold value set by the threshold value setting unit, and the absolute value is A pixel setting value calculating means for calculating a predetermined pixel setting value when it is not determined that the threshold value is greater than the threshold;
In the pixel setting value calculation means, when the absolute value is determined to be larger than the threshold, the pixel value is used as an output signal, and when the absolute value is not determined to be larger than the threshold, the pixel setting Noise reduction means for outputting a value as an output signal;
A noise reduction device comprising:   The pixel setting value calculation unit outputs the median value as the pixel setting value when the absolute value is not determined to be larger than the threshold value. Noise reduction device. A shift amount calculation is performed in which a correspondence relationship between the pixel value detected by the pixel value detection unit and the shift amount is set in advance, and a shift amount corresponding to the pixel value detected by the pixel value detection unit is calculated based on the correspondence relationship. Further comprising means,
When the absolute value is not determined to be greater than the threshold value, the pixel setting value calculation unit determines whether the pixel value is greater than the median value, and the pixel value is the median value. The pixel setting value is calculated by subtracting the shift amount from the pixel value, and when the pixel value is not determined to be greater than the median value, The noise reduction device according to claim 1, wherein the pixel setting value is calculated by adding a shift amount. The noise level detection means includes
Block dividing means for dividing the digital video signal into blocks of a predetermined size for each screen;
Orthogonal transform means for orthogonally transforming the blocks divided by the block dividing means for each block;
High-frequency component detection means for averaging the coefficient of the high-frequency component excluding the DC component of the block transformed by the orthogonal transform means and the predetermined vicinity of the DC component in the block and detecting it as the level of the high-frequency component of the block; ,
High-frequency component accumulating means for accumulating the number of blocks corresponding to the level of the high-frequency component detected by the high-frequency component detecting means for each level;
Noise level determination means for determining an average value corresponding to a predetermined number of blocks from the low level value of the level of the high frequency component accumulated by the high frequency component accumulation means as the level value;
The noise reduction device according to any one of claims 2 to 4, further comprising: The noise level detection means includes
From the digital video signal, line extraction means for sequentially extracting adjacent lines adjacent to the target line as a line constituting the screen as a correlation line group;
In the correlation line group extracted by the line extraction means, the difference between the pixel values of the corresponding pixels of the target line and the adjacent line is averaged and detected as an inter-line noise level in the correlation line group of the target line. Noise level detection means between lines;
Level accumulation means for accumulating the number of lines of the target line corresponding to the noise level between lines detected by the noise level between lines for each noise level between the lines;
Noise level determination means for determining, as the level value, an average value corresponding to a predetermined number of lines from a low level value of the inter-line noise level accumulated by the level accumulation means;
The noise reduction device according to any one of claims 2 to 4, further comprising: In order to reduce the noise superimposed on the digital video signal,
A median filter that presets pixels at the same screen position as the target pixel in a frame constituting the input digital video signal and the preceding and following frames before and after the frame, and selects median values in the plurality of target pixels;
Pixel value detection means for detecting a pixel value of the target pixel of the frame;
It is determined whether or not the absolute value of the difference between the median values selected by the median filter from the pixel values detected by the pixel value detection means is greater than a preset threshold value, and the absolute value is greater than the threshold value. A pixel setting value calculating means for calculating a predetermined pixel setting value if it is not determined to be large;
In the pixel setting value calculation means, when the absolute value is determined to be larger than the threshold, the pixel value is used as an output signal, and when the absolute value is not determined to be larger than the threshold, the pixel setting Noise reduction means for outputting a value as an output signal,
Noise reduction program characterized by functioning as

Images