JP2008199267A - Video receiver - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a video signal noise amount measuring instrument and a cyclic noise reduction receiver which are capable of precisely and efficiently estimating a noise amount in actual moving videos by using an MPEG noise amount calculation means, an MPEG noise removal means, and a means for detecting correlativity between preceding and succeeding frames. <P>SOLUTION: The noise amount measuring instrument includes a storage means 101 for storing an input signal, an MPEG noise removal means 102, an MPEG noise amount calculation means 103, and the means for detecting correlativity between preceding and succeeding frames per area. In this case, the sum of an MPEG noise amount calculated by the MPEG noise amount calculation means and a noise amount calculated on the basis of differences between highly correlative areas between preceding and succeeding frames of a signal after MPEG noise removal is measured. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to noise in a video signal, and more particularly to measurement of the amount of noise and a cyclic noise reduction apparatus.

  Conventionally, as shown in Patent Document 1, this kind of noise detection in a video signal is performed by determining a signal of a known level as a part of the video and taking a difference from the known level for this signal. The amount was calculated. In other words, it was not an evaluation of the amount of noise included in the video in the true sense, but the amount of noise experienced by the system through which the video passed.

  Alternatively, a difference is taken between frames, and a noise amount is calculated from the difference. That is, if this difference is large, it is determined that the amount of noise is large. However, in this case, even for images with motion, the difference is large, and it is difficult to determine whether the difference is due to noise or motion. It was difficult.

  Alternatively, an area with high correlation was searched for in the previous and subsequent frames, and the amount of noise was calculated from the difference from that area. When this processing is performed on an image in which MPEG noise such as block noise is generated, Since MPEG noise is highly correlated with the previous and subsequent frames, there is a problem that it is not reflected as the amount of noise.

Also, in order to realize a cyclic noise reduction device, a noise reduction effect works even on an image with little noise due to an error in measuring the amount of noise, which makes the image unsightly.
Japanese Patent No. 3590117

  As described above, conventionally, in order to detect the amount of noise in this type of video signal, it is necessary to determine a signal of a known level as a part of the video. That is, it is not an evaluation of the amount of noise included in the video in the true sense by detecting the amount of noise experienced by the system through which the video has passed. In addition, there is a problem that noise having high correlation between frames before and after such as block noise cannot be recognized as noise and is not reflected in the measurement of the noise amount.

  Accordingly, an object of the present invention is to solve the above-mentioned problems, and to reduce the amount of noise in a moving real image using both the MPEG noise amount calculation / removal means and the means for detecting the correlation between the previous and next frames. It is an object of the present invention to provide a video signal noise amount measuring apparatus and a cyclic noise reduction receiver that can be estimated accurately and efficiently.

In order to solve this problem, in the first invention of the present application, a first input terminal for inputting a video signal and a second input terminal for storing the video signal input from the second input terminal are stored. A storage means, a means for detecting MPEG noise from the first video signal inputted from the first input terminal and a second video signal read from the first storage means, and the MPEG noise detection means; MPEG noise amount calculating means for calculating an MPEG noise amount for a certain pixel based on the MPEG noise detection information, and an MPEG noise amount for the certain pixel calculated by the MPEG noise calculating means is integrated over the video area. 1 integrating means, MPEG noise removing means for removing MPEG noise, and the MPEG noise removing means for the first video signal and the second video. A difference calculation means for finding a region having the highest correlation by comparing the first MPEG noise removal signal and the second MPEG noise removal signal from which the MPEG noise has been removed, and calculating a sum of signal differences in the region; And adding the results calculated by the second integrating means for integrating the difference of the signal difference calculated by the correlation amount calculating means over the video region, the first integrating means and the second integrating means. Equipped with
Based on the MPEG noise detection information detected by the MPEG noise detection means from the first video signal and the second video signal, the MPEG noise amount calculation means calculates an MPEG noise amount for a certain pixel, The first integration means calculates the first noise amount in the video area by integrating over the video area, and also removes the MPEG noise for the first video signal and the second video signal, respectively. MPEG noise is removed by the means, and the difference calculating means searches the second MPEG noise removal signal for a region having the highest correlation with the image signal in an arbitrary first region of the first MPEG noise removal signal. (The searched area is set as the second area), and a sum of signal differences between the first area and the second area is calculated, and the second area is calculated. The integration means integrates the signal difference between the areas over the video area to calculate a second noise amount, and then adds the first noise amount and the second noise amount by the adder. Thus, the amount of noise is measured.

In the second invention of the present application, first and second input terminals for inputting a video signal, first storage means for storing a video signal input from the second input terminal, and the first MPEG noise detection information by means for detecting MPEG noise from the first video signal input from one input terminal and the second video signal read from the first storage means and MPEG noise detection means. MPEG noise amount calculating means for calculating an MPEG noise amount for a certain pixel, first integrating means for integrating the MPEG noise amount for a certain pixel calculated by the MPEG noise calculating means over a video area; MPEG noise removing means for removing MPEG noise, and MPEG noise for the first video signal and the second video respectively by the MPEG noise removing means. The first MPEG noise removal signal from which the noise has been removed and the second MPEG noise removal signal are compared to find a region having the highest correlation, and a difference calculating means for obtaining a sum of signal differences in the region; A second integrating means for integrating the difference of the signal difference calculated by the means over the video region, an adder for adding the results calculated by the first integrating means and the second integrating means, and further a cycle Type noise reduction means,
Based on the MPEG noise detection information detected by the MPEG noise detection means from the first video signal and the second video signal, the MPEG noise amount calculation means calculates an MPEG noise amount for a certain pixel, The first integration means calculates the first noise amount in the video area by integrating over the video area, and also removes the MPEG noise for the first video signal and the second video signal, respectively. MPEG noise is removed by the means, and the difference calculating means searches the second MPEG noise removal signal for a region having the highest correlation with the image signal in an arbitrary first region of the first MPEG noise removal signal. (The searched area is set as the second area), and a sum of signal differences between the first area and the second area is calculated, and the second area is calculated. The integration means integrates the signal difference between the areas over the video area to calculate a second noise amount, and then adds the first noise amount and the second noise amount by the adder. Thus, the noise amount is calculated, and the noise reduction effect is adjusted according to the calculated noise amount.

In the third invention of the present application, first and second input terminals for inputting a video signal, first storage means for storing a video signal input from the second input terminal, and the first Means for detecting MPEG noise from a first video signal inputted from one input terminal and a second video signal read from the first storage means, MPEG noise removing means for removing MPEG noise, The first MPEG noise removal signal and the second MPEG noise removal signal from which the MPEG noise has been removed from the first video signal and the second video by the MPEG noise removal means are compared, and the correlation is highest. A difference calculating means for finding a high area and obtaining a sum of signal differences in the area; and a second integrating means for integrating the difference of the signal difference calculated by the correlation amount calculating means over the video area. Comprising a recursive noise reduction means,
MPEG noise is removed from the first video signal and the second video signal by the MPEG noise removal unit, and the difference calculation unit determines an arbitrary first area of the first MPEG noise removal signal. An area having the highest correlation with the image signal is searched from the second MPEG noise removal signal (the searched area is set as the second area), and a signal between the first area and the second area is detected. The sum of the differences is calculated, the second integration means integrates the sum of the signal differences between the regions over the video region to calculate a second noise amount, and the first MPEG noise removal signal is cycled. The noise reduction effect is adjusted according to the second noise amount as an input of the type noise reduction means.

  According to the first invention of the present application, it is possible to measure the amount of noise in the video region, and it is also possible to measure the amount of noise even for moving video signals compared to the case of measuring without considering the correlation. Further, by using the MPEG noise amount calculation / removal means, MPEG noise is also taken into account, so that there is an effect of improving the accuracy of the noise amount.

  In addition, according to the second and third aspects of the present invention, the amount of noise can be accurately measured for the above-described reason, so that the noise reduction effect can be adjusted appropriately, and a video receiver with improved image quality can be realized. It becomes. In particular, according to the third invention of the present application, a signal in which MPEG noise, which is not good for the cyclic noise reduction means, is processed in advance, and the cyclic noise is calculated based on the noise amount calculated by eliminating the MPEG noise component. Therefore, the noise reduction effect is further improved.

  Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS.

(Embodiment 1)
FIG. 1 shows an example of a block diagram of a noise amount measuring apparatus according to the first aspect of the present invention. The noise amount measuring apparatus of the present invention includes a storage unit 101 that delays an input video signal by at least one frame, an MPEG noise detection unit 102 that detects MPEG noise such as block noise and mosquito noise from the input video signal, An MPEG noise amount calculating means connected to the MPEG noise detecting means 102 for calculating an MPEG noise amount, a first integrating means 104 for integrating the noise amount for a predetermined time, and an MPEG noise detecting means 102 for removing the MPEG noise. MPEG noise removing means 105, 106 and the first and second MPEG noise removal signals connected to the MPEG noise removing means 105, 106 and from which MPEG noise has been removed by the MPEG noise removing means 105, 106 are input signals. The signal difference of the corresponding predetermined area of A region sum total minimum value calculation means (hereinafter referred to as “difference calculation means”) to be output; a second integration means for integrating the difference value output from the difference calculation means for a predetermined time; and the first integration described above. And an adder 109 for adding the values of the means and the second integrating means.

  Next, an operation process of the noise amount measuring apparatus having these configurations will be described. In the first invention of the present application, first, the MPEG noise detection means shown in 102 from the first video signal inputted from the input terminal 1 and an arbitrary second video signal at least one frame before the first video signal. Thus, MPEG noise is detected by a method described later. From the detection information by the MPEG noise detecting means, the MPEG noise amount for a certain pixel is calculated by the MPEG noise amount calculating means shown at 103. The calculated MPEG noise amount for a certain pixel is integrated over the video area by the first integration means shown at 104 to obtain the first noise amount.

  Next, a first MPEG noise removal signal obtained by removing MPEG noise components from the first video signal is generated from the detection information by the MPEG noise detection means indicated by 102 by the MPEG noise removal means indicated by 105. Similarly, a second MPEG noise removal signal obtained by removing an MPEG noise component from the second video signal is generated by the MPEG noise removal means 106.

  Next, the difference calculating means shown in 107 is an arbitrary first area centered on an arbitrary pixel having the first MPEG noise removal signal (for example, 5 × 5, hereinafter referred to as “area P”). Is used as a reference image, a region having the highest correlation between the second MPEG noise removal signal and the reference image (region where the signal difference between the regions is minimized) is searched, and the signal difference between the regions is calculated. The difference between the signals in the region having the highest correlation is integrated over the video region by the second integration means indicated by 108 to calculate the second noise amount. More specific processing of the difference calculating unit 107 will be described in detail below.

  Next, the first noise amount output from the first integration circuit and the second noise amount output from the second integration circuit are added by an adder indicated by 109 to obtain the total noise amount. .

  Here, the processing in the above-described MPEG noise detection means 102 will be described in more detail. As a specific MPEG noise detection method by the above-described MPEG noise detection means 102, the following method is conceivable. MPEG noise is roughly classified into two types, block noise and mosquito noise.

  First, a block noise detection method will be described. As a first block noise detection method, a high-pass filter process (not shown) is performed on the first video signal by a high-pass filter (HPF), and then a pixel whose signal level after the process is within a certain threshold is extracted. In addition, if the pixels are arranged in a straight line in the horizontal or vertical direction, a method of discriminating from block noise pixels can be considered. Further, as a second block noise detection method, a high-pass filter process (not shown) is performed on the first video signal by a high-pass filter (HPF), and then the pixel level after the process is within a certain threshold ( Pixel B), and the pixels are arranged in a straight line in the horizontal or vertical direction. Further, the pixel B on the time axis of the signal obtained by performing the high-pass filter process on the second video signal is extracted. If the signal level of the pixel at the same position as the pixel B is similar to that of the pixel B (if the difference is within a certain threshold value), a method of discriminating from the block noise pixel can be considered.

  Next, a method for detecting mosquito noise will be described. As a mosquito noise detection method, after a high-pass filter process (HPF) (not shown) is performed on the first video signal, a pixel whose signal level after the process is within a certain threshold is extracted, If there is a large outline component, a method of using a mosquito noise pixel can be considered.

  A specific MPEG noise amount calculation method by the MPEG noise amount calculation means 103 is to obtain a difference in signal level between the block noise pixel and the mosquito noise pixel detected by the MPEG noise detection means 102 and the surrounding pixels, A method in which the value is used as an MPEG noise amount can be considered.

  As a specific MPEG noise removal method by the MPEG noise removal means, a method of applying an appropriate filter such as a low-pass filter to the block noise pixels and mosquito noise pixels detected by the MPEG noise detection means can be considered.

  Next, the processing of the difference calculation means 107 will be described with reference to FIG. As a specific search method for a highly correlated area in the difference calculating means shown in 107, first, as shown in FIG. 2, the area P and the second MPEG removal signal are the same position as the area P. The difference for each pixel from the second MPEG removal signal at is calculated, and the sum of the differences is obtained. This sum of differences will be referred to as a correlation value.

  Next, one pixel or a plurality of pixels are moved in the region P right and left and up and down, and the correlation value at that position is similarly obtained. Then, by moving the region P within a certain range and obtaining a minimum correlation value, it is possible to search for a region with high correlation. Note that the minimum correlation value is referred to as a minimum correlation value.

  After calculating the minimum correlation value, the minimum correlation value is similarly determined for all the pixels in the video area, and the sum of the minimum correlation values is determined by the second integration means shown at 108. The sum of the minimum correlation values obtained by the second integrating means will be referred to as Vn. Vn can be considered as a noise amount reflecting the noise amount not only for a motionless image but also for a moving image.

  However, the Vn value becomes extremely large for an image that moves too quickly and cannot catch up with a search for a highly correlated area, or an image that changes rapidly due to a scene change or the like. However, a threshold TH1 for determining that the correlation between preset frames is high (correlation is equal to or higher than a predetermined value) and TH2 for determining that the correlation between frames is low (correlation is equal to or lower than a predetermined value). By using (TH1 <TH2), if Vn <TH1, it is judged that the image is less noisy, if TH1 <Vn <TH2, the image is noisy, and if TH2 <Vn, the image is fast moving. Thus, it is possible to perform an appropriate noise measurement by excluding such a noise-moving image from a noise measurement target. By providing such a predetermined threshold value, it is possible to make Vn an effective value for measuring the amount of noise even for a video image including a scene having no correlation between frames.

  Also, the reason for removing the MPEG noise component in the MPEG noise removing means shown in 105 to 106 in the previous stage of the difference calculating means shown in 107 is to search for a highly correlated area in the MPEG noise before removing the block noise component. Then, since the block noise component has a correlation between the previous and subsequent frames, as shown in FIG. 3, a region with high correlation may be detected at an incorrect position due to the influence of the block noise component. There is sex. Therefore, the area difference sum minimum value is calculated after removing the block noise because noise having no frame correlation other than MPEG noise can be accurately extracted.

  When searching for a highly correlated area, the area P is a rectangle of one pixel in the vertical direction and several pixels (for example, five pixels) in the horizontal direction. It is possible to save memory resources by conforming to the method.

  In addition, two-dimensional noise candidate detection means is provided immediately before or immediately after the MPEG noise removal means shown in 105, noise candidates are calculated in advance, and the area difference sum minimum value is calculated only for the area centered on the pixel. It is also possible. This has the effect of reducing the processing burden.

  Further, statistical processing is performed using Vn calculated for each frame as data using a first computing means (not shown) provided after the second integrating means of 108 or after the adder of 109. As a result, the accuracy of noise amount measurement is improved. As statistical processing, a method using a histogram, a method for calculating a median value, and a method for calculating an average value are conceivable.

  Here, a method of using a histogram in statistical processing will be described in detail. In the method using the histogram, the histogram is calculated for a plurality of noise amount data calculated in a certain continuous plurality of frames including the current frame. Vn values existing in the most frequent range are extracted from the histogram. For example, an average value Vn1 is calculated for the extracted Vn value. By using such an average value Vn1 as the noise measurement amount in the current frame, it is possible to stably measure the noise amount even when the Vn value suddenly increases due to a fast-moving image, and the noise amount measurement accuracy is improved. There is an effect that it is greatly improved.

  Next, a method for calculating a median value in statistical processing will be described in detail. The method for calculating the median is to calculate a median value Vnm of a plurality of noise amount data calculated in a certain continuous plurality of frames including the current frame, and use the Vnm as a noise measurement amount in the current frame. . The method of calculating the median value has an effect that a simple hardware design can be realized and the accuracy of the noise measurement value is improved. The average value is calculated by calculating an average value Vna of a plurality of noise amounts calculated in a certain continuous plurality of frames from the current frame, and using the Vna as a noise amount in the current frame. As a result, a simple hardware design can be realized and the accuracy of noise measurement values can be improved.

  Further, before performing statistical processing in the second invention of the present application, by applying a low-pass filter in the frame axis direction to the Vn data calculated for each frame in advance, the variation in Vn is reduced. By performing the processing, there is an effect that the accuracy of noise amount measurement is further improved.

  Thus, by using the MPEG noise amount calculation means, the MPEG noise amount can be reflected as the noise amount. In addition, by using the MPEG noise removing unit and the difference calculating unit, it is possible to accurately calculate the noise amount even for random noise other than MPEG noise, and there is an effect of improving the noise amount measurement accuracy.

  That is, in the noise amount measuring apparatus or the noise amount measuring method according to the present invention, in the video signal from which the MPEG noise amount is removed in addition to the MPEG noise amount such as block noise and mosquito noise with respect to the input video signal. By calculating the amount of noise, it is possible to measure both continuously generated noise and randomly generated random noise, and it is possible to measure the amount of noise in a video area that could not be measured conventionally. To do.

(Embodiment 2)
Next, in the second embodiment, the noise reduction effect is adjusted according to the value of Vn measured by the noise measuring device in the configuration including the cyclic noise reducing device in addition to the noise amount measuring device described above. A video receiver will be described.

  FIG. 4 shows a block diagram of this video receiver. This apparatus includes an input terminal 401 to which a digital video signal is input, a first multiplier 402 for multiplying the input video signal by a coefficient 1-K, and a feedback coefficient K for the video signal delayed by one frame by the frame memory 406. A second multiplier 403 for multiplication, a coefficient generator 404 for generating a coefficient K (where 0 ≦ K ≦ 1), and outputs of the multipliers 402 and 403 are added to supply an addition signal to the frame memory 406. On the other hand, in addition to the adder 405 that outputs the same addition signal to the output terminal 407 and the frame memory 406 that stores the addition signal from the adder 405 for one frame as a new video signal, the description is made in the first embodiment. The noise amount measuring device 408 is provided. The multipliers 402 and 403, the adder 405, and the frame memory 406 constitute a frame recursive filter.

  Here, the configuration of the frame recursive filter will be described. By this frame recursive filter, a signal obtained by multiplying the input video signal by 1-K (which can be said to be a signal obtained by reducing the input video signal by the feedback coefficient K) and a signal obtained by multiplying the video signal delayed by one frame by K are obtained. As a result, the difference signal between frames (in the case of a video signal with strong frame correlation, this difference signal is mainly a noise component) multiplied by a coefficient K is subtracted from the input video signal. The noise is removed.

  Here, assuming that the input video signal is S1, and the one-frame delayed video signal output from the frame memory 406 is S2, (1-K) S1 + KS2 = S1-K (S1-S2) is output to the output terminal 407. Is done. That is, noise removal is achieved by subtracting K (S1-S2) corresponding to the noise component from the current input video signal S1.

  In this apparatus, the coefficient value K is determined by the coefficient generator 404 based on the noise measurement value measured by the noise amount measurement apparatus. For example, it is assumed that the coefficient generator 404 generates the coefficient K with the characteristics as shown in FIG. That is, when the noise measurement value is 0 or more and Nth1 or less, it is considered that the image does not move, the coefficient K is monotonously increased with respect to the increase of the noise measurement value, and when the noise measurement value is Nth1 or more, the image is considered to move quickly The coefficient K is monotonously decreased with respect to the increase of the noise measurement value, and the coefficient K is set to 0 when the noise measurement value is Nth2 or more.

  Then, when the coefficient K is large, it is assumed that the corresponding pixel is a portion where there is a lot of noise with the passage of time, and a highly correlated video signal is added and averaged in time series, thereby reducing noise. Can do. On the other hand, when the coefficient K is small, the input video signal is output as it is, assuming that the corresponding pixel is a part with little noise or a part that changes and moves with time, and the correlation is low. It is possible to prevent the occurrence of an afterimage such that the video signal of the moved part is added and the tail is pulled. Therefore, this makes it possible to realize a cyclic noise reduction device according to the amount of noise in the video signal.

(Embodiment 3)
The third invention of the present application is a video receiver that adjusts a noise reduction effect in accordance with the value of Vn measured by a noise measuring device in a device including the noise amount measuring device and a cyclic noise reducing device. is there.

  FIG. 6 is a block diagram showing the configuration of this video receiver. The video receiver includes a storage unit 601 that delays an input video signal by at least one frame, an MPEG noise detection unit 602 that detects MPEG noise such as block noise and mosquito noise from the input video signal, and MPEG noise detection. MPEG noise removing means 605 and 606 connected to the means 602 for removing MPEG noise, and first and second MPEG signals connected to the MPEG noise removing means 605 and 606 and from which MPEG noise has been removed by the MPEG noise removing means 605 and 606. An area difference total minimum value calculating means (hereinafter referred to as “difference calculating means”) that calculates a difference between signals in a predetermined area corresponding to the input signal by using the noise removal signal as input, and is output from the difference calculating means. Second integration means 608 for integrating the difference value for a predetermined time, and a second product It is constituted by recursive type noise reducing means 610 for performing noise reduction processing signals values and MPEG noise removing unit 605 of the device as an input.

  Here, this apparatus inputs the first MPEG noise removal signal removed by the MPEG noise removal means indicated by 605 to the cyclic noise reduction apparatus indicated by 610, and outputs the output of the second integration means described above as noise. The noise reduction effect is adjusted according to this value.

  Thereby, the MPEG noise component can be reduced by the MPEG noise removing means, and the noise component having no frame correlation other than MPEG (random noise) can be reduced by the cyclic noise reducing means, and the cyclic noise reducing means can be reduced. Since the noise amount serving as a reference for adjusting the noise reduction effect is calculated from the signal from which the MPEG noise component has been removed, random noise can be reduced with high accuracy.

  The present invention described above is not limited to the embodiments described herein, and it is obvious that various modifications and changes can be made within the scope of the invention.

  The apparatus and method for measuring the amount of noise of a video signal according to the present invention enables measurement of the amount of noise in the video region, and can prevent erroneous measurement due to movement compared to the case where measurement is performed without considering correlation. Further, by using the MPEG noise amount calculation means, the MPEG noise amount can be reflected as the noise amount. In addition, the amount of noise can be accurately calculated for random noise other than MPEG noise, and the noise amount measurement accuracy can be improved. As a result, the noise reduction effect can be adjusted appropriately, and there is also an effect that a video receiver with improved image quality can be realized. The noise measurement and cyclic noise reduction device and the video receiver using the same It is useful for such as.

The figure which shows the structure of the noise amount measuring apparatus by 1st invention of this application. The figure which showed the method of searching the area | region with high correlation by 1st invention of this application Explanatory drawing when searching for a highly correlated area for a signal that does not remove MPEG noise components according to the first invention of the present application The figure which shows the structure of the cyclic | annular noise reduction apparatus by 2nd invention of this application. The figure which showed the output characteristic of the coefficient K in the coefficient generation circuit in the cyclic | annular noise reduction apparatus by 2nd invention of this application The figure which shows the structure of the cyclic | annular noise reduction apparatus by this invention 3rd invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Memory | storage means 102 MPEG noise detection means 103 MPEG noise amount calculation means 104 1st integration means 105 MPEG noise removal means 106 MPEG noise removal means 107 Area difference sum total minimum value calculation means 108 Second integration means 109 Adder

Claims (11)

First and second input terminals for inputting a video signal, first storage means for storing a video signal input from the second input terminal, and first input from the first input terminal MPEG noise detection means for detecting MPEG noise from one video signal and a second video signal read from the first storage means, and a predetermined pixel based on detection information at the MPEG noise detection means. An MPEG noise amount calculating means for calculating an MPEG noise amount, a first integrating means for integrating the noise amount calculated by the MPEG noise calculating means in a video area, and a video based on detection information from the MPEG noise detecting means. The MPEG noise removing means for removing the MPEG noise of the signal and the first video signal output from the MPEG noise removing means and the second video are compared. A difference calculating means for calculating a difference between signals in a correlated area; a second integrating means for integrating the signal difference calculated by the difference calculating means in a video area; the first integrating means; and the first integrating means. A noise amount measuring apparatus comprising: an adder for adding integration results of the two integrating means. Based on MPEG noise detection information detected by the MPEG noise detection means from the first video signal and the second video signal, the MPEG noise quantity calculation means calculates an MPEG noise quantity of a predetermined pixel, The first integration means calculates the first noise amount in the video area by integrating over the video area, and also removes the MPEG noise for the first video signal and the second video signal, respectively. MPEG noise is removed by means, and in the difference calculating means, a second region having the highest correlation with an image signal in an arbitrary first region of the first MPEG noise removal signal is determined as the second MPEG noise removal. Search from the signal, calculate the sum of the signal differences between the first region and the second region, and the signal between the regions in the second integration means After calculating the second noise amount by integrating the sum of minutes over the video area, the noise amount is measured by adding the first noise amount and the second noise amount in the previous period by the adder. The noise amount measuring apparatus according to claim 1. The first area is formed by an area of one pixel in the vertical direction and at least two pixels in the horizontal direction, and the first integration means or the second integration means is at least for a plurality of areas in the video area. 3. The noise amount measuring apparatus according to claim 2, wherein integration processing is performed. The integration result of the second integration means is compared with a preset threshold value for determining the presence or absence of correlation between frames, and only when the correlation is determined to be greater than the threshold value and correlated. The noise amount measuring apparatus according to claim 1, wherein the noise amount measuring apparatus outputs the signal to a measuring device. The noise amount measuring apparatus further includes a first calculation unit that calculates an output result of the second integration unit or the adder, and the first calculation unit is output by the second integration unit. An average noise amount between frames is calculated for an integration result or an addition value added by the adder, and compared with a result output from the second integration unit or the adder. Item 5. The noise amount measuring apparatus according to any one of Items 1 to 4. 6. The noise amount measuring apparatus according to claim 5, wherein the first calculating means sets the average noise amount to a noise amount existing in a most frequent range of a histogram in a fixed number of consecutive frames. The noise amount measuring apparatus according to claim 5, wherein the first calculation means sets a median value in a plurality of continuous frames as the average noise amount. The first calculation means performs the low-pass filter on the input integration result in advance in the time axis direction, and then acquires the average noise amount. The noise amount measuring apparatus according to any one of the above. A two-dimensional noise candidate detecting means is provided after the MPEG noise removing means for removing MPEG noise from the first video signal, and a noise candidate is previously detected by the two-dimensional noise candidate detecting means for the first MPEG noise removed signal. And the difference calculation means searches the second MPEG noise removal signal for a fourth region having the highest correlation with the image signal in the arbitrary third region including the noise candidate pixel, and The noise amount measuring apparatus according to claim 1, wherein a sum of signal differences between the third region and the fourth region is calculated. The noise amount measuring apparatus further includes a coefficient generator that outputs a predetermined coefficient, and a multiplier that multiplies the predetermined coefficient output by the coefficient generator,
The multiplier multiplies the integration results output from the first integration unit and the second integration unit by predetermined coefficients, respectively, and outputs the result as a first multiplication result and a second multiplication result. The video receiver according to claim 1, wherein the first multiplication result and the second multiplication result are added by a detector. First and second input terminals for inputting a video signal, first storage means for storing a video signal input from the second input terminal, and first input from the first input terminal A first video signal and a second video signal read from the first storage means, a means for detecting MPEG noise, a MPEG noise removal means for removing MPEG noise, and the MPEG noise removal means for the first noise signal. An area having the highest correlation is found by comparing the first MPEG noise removal signal and the second MPEG noise removal signal from which the MPEG noise has been removed from the video signal and the second video, respectively. A difference calculating means for obtaining a sum of differences; a second integrating means for integrating the difference of the signal difference calculated by the correlation amount calculating means over the video region; and a cyclic noise reducing means. For example,
MPEG noise is removed from the first video signal and the second video signal by the MPEG noise removal unit, and the difference calculation unit determines an arbitrary first area of the first MPEG noise removal signal. A second region having the highest correlation with the image signal in the second MPEG noise removal signal is searched, a sum of signal differences between the first region and the second region is calculated, and the second region is calculated. And integrating the signal difference between the regions over the video region to calculate a second noise amount, and using the first MPEG noise removal signal as an input to the cyclic noise reduction unit. A video receiver, wherein the noise reduction effect is adjusted according to the amount of noise of 2.

Images