JP5013932B2 - Noise reduction device, noise reduction method, and electronic device - Google Patents

Description

  The present invention relates to a noise reduction device and a noise reduction method for reducing noise of a moving image composed of a plurality of frames, and in particular, reduces noise by using two-dimensional noise reduction processing and three-dimensional noise reduction processing together. The present invention relates to a noise reduction device and a noise reduction method. The present invention also relates to an electronic apparatus provided with the noise reduction device.

  In recent years, with the development of various digital technologies, imaging devices such as digital cameras and digital videos that take digital images with solid-state imaging devices such as CCD (Charge Coupled Device) and CMOS (Complimentary Metal Oxide Semiconductor) sensors, and digital images Display devices such as liquid crystal displays and plasma televisions for displaying are becoming widespread. Various noise reduction techniques for reducing noise of moving images composed of a plurality of frames in such an imaging apparatus and display apparatus have been proposed.

  As a typical example of this noise reduction technique, a two-dimensional noise reduction process for reducing noise by applying a spatial low-pass filter and a signal of two temporally continuous frames are added. There is a three-dimensional noise reduction process in which a temporal low-pass filter is applied to reduce noise.

  Here, a noise reduction apparatus that uses both a two-dimensional noise reduction process and a three-dimensional noise reduction process that have been known for a long time will be described with reference to the drawings. FIG. 16 is a block diagram showing a schematic configuration of a noise reduction device 90 that uses a two-dimensional noise reduction process and a three-dimensional noise reduction process that have been known for a long time. The noise reduction device 90 performs a three-dimensional noise reduction process on the input signal, and outputs the obtained three-dimensional noise-reduced signal to the weighted addition unit 93; and the input signal A two-dimensional noise reduction processing unit 92 that performs a two-dimensional noise reduction process and outputs the obtained two-dimensional noise-reduced signal to the weighted addition unit 93; and a three-dimensional noise reduction input from the three-dimensional noise reduction processing unit 91 A weighted addition unit 93 that performs weighted addition of the signal and the two-dimensional noise-reduced signal input from the two-dimensional noise reduction processing unit 92 with a predetermined weighting coefficient, and outputs the obtained noise-reduced signal to the subsequent stage; have.

  Here, the predetermined weighting coefficient used in the weighting and adding unit 93 is generally determined by the amount of motion of each pixel, that is, the difference in signal value between two consecutive frames of each pixel. In such a case, weighting is performed so that the influence of the three-dimensional noise reduction processing unit 91 is increased on a pixel having a small amount of motion, and weighting is performed on a pixel having a large amount of motion. Is made.

  Therefore, in the noise reduction device 90, the three-dimensional noise reduction processing is mainly performed by the three-dimensional noise reduction processing unit 91 for pixels with a small amount of motion, and the two-dimensional noise reduction processing unit 92 is mainly performed for pixels with a large amount of motion. The two-dimensional noise reduction processing is performed, and the weak point of the three-dimensional noise reduction processing that does not function effectively for pixels with a large amount of motion is compensated by using the two-dimensional noise reduction processing together.

Further, in Patent Document 1, as another example of a noise reduction device that uses both two-dimensional noise reduction processing and three-dimensional noise reduction processing, each pixel is classified into a still image portion or a moving image portion according to the amount of motion. In addition, a noise reduction apparatus is described that performs a three-dimensional noise reduction process for pixels classified in the still image portion and performs a two-dimensional noise reduction process for pixels classified in the moving image portion.
Japanese Patent No. 2782766

  However, when a large noise occurs in the input signal input to the noise reduction device 90, the amount of motion of the pixel including the noise is calculated to be considerably larger than the original. Even a pixel that is originally suitable for a three-dimensional noise process with a very small amount of motion is mainly subjected to a two-dimensional noise reduction process by the two-dimensional noise reduction processing unit 92, and thus an insufficient noise reduction effect and image blurring. Is a problem.

  In addition, even when a large noise occurs in the input signal input to the noise reduction device of Patent Document 1, since the amount of motion of the pixel including the noise is calculated to be considerably larger than the original, the pixel including the noise is calculated. May be classified as a moving image portion even though it is originally a still image portion. When a pixel that is originally a still image portion is classified as a moving image portion, a two-dimensional noise reduction process is performed on the pixel instead of a three-dimensional noise reduction process that should be originally performed. Insufficient noise reduction effect and image blur are problems.

  In view of the above problems, in the present invention, a noise reduction apparatus and a noise reduction method using both a two-dimensional noise reduction process and a three-dimensional noise reduction process, and calculating a motion amount calculation error of each pixel due to the influence of noise. An object of the present invention is to provide a noise reduction device and a noise reduction method that can be reduced. Moreover, an object of this invention is to provide the electronic device provided with this noise reduction apparatus.

  In the present invention, the noise reduction device is a noise reduction device that receives an image signal from the outside and outputs a signal after noise reduction, and performs a two-dimensional noise reduction process on the image signal input from the outside. A two-dimensional noise reduction processing unit for obtaining a two-dimensional noise-reduced signal; a difference value between the two-dimensional noise-reduced signal input from the two-dimensional noise reduction processing unit and the noise-reduced signal one frame before A motion detection unit that obtains a motion amount of each pixel in the time axis direction by calculating the image signal, and the motion in the time axis direction of each pixel input from the motion detection unit with respect to the image signal input from the outside And a three-dimensional noise reduction processing unit that performs a three-dimensional noise reduction process in which the gain is adjusted according to the amount and obtains a signal after the three-dimensional noise reduction.

  According to the present invention, in the noise reduction device having the above configuration, the two-dimensional noise reduction signal input from the two-dimensional noise reduction processing unit and the three-dimensional noise input from the three-dimensional noise reduction processing unit. A weighted addition unit that obtains the noise-reduced signal by weighting and adding the reduced signal with a predetermined weighting factor, and according to the motion amount in the time axis direction of each pixel input from the motion detection unit It is desirable to have a weighting coefficient detection unit that calculates a predetermined weighting coefficient.

  In the present invention, in the noise reduction device having the above-described configuration, the two-dimensional noise reduction processing performed by the two-dimensional noise reduction processing unit according to the signal level of each pixel of the image signal input from the outside. It is desirable to have a signal level gain adjustment unit that adjusts the gain of the signal.

  In the present invention, in the noise reduction device having the above-described configuration, the two-dimensional noise reduction processing unit performs according to the amount of motion in the time axis direction of the pixel of interest before one time point input from the motion detection unit. It is desirable to have a motion gain adjustment unit that adjusts the gain of the two-dimensional noise reduction process.

  According to the present invention, in the noise reduction device having the above configuration, the two-dimensional noise reduction processing unit includes the two-dimensional noise including processing by an edge preserving noise reduction filter for the image signal input from the outside. It is desirable to perform a reduction process.

  According to the present invention, in the noise reduction device having the above-described configuration, the two-dimensional noise reduction processing unit includes a two-dimensional noise reduction process including processing by an impulse noise reduction filter for the image signal input from the outside. It is desirable to apply.

  According to the present invention, in the noise reduction device having the above configuration, the three-dimensional noise reduction processing unit calculates a difference value between the image signal input from the outside and the noise-reduced signal one frame before a predetermined value. The three-dimensional noise reduction processing is performed using a noise detection value obtained by acting on a function, and the predetermined noise is determined according to the content of impulse noise of each pixel of the image signal input from the outside. It is desirable to have a function changing unit that changes the function of.

  In the present invention, the noise reduction method is a noise reduction method in which an image signal is input from the outside and a signal after noise reduction is output, and a two-dimensional noise reduction process is performed on the image signal input from the outside. A two-dimensional noise reduction processing step for obtaining a two-dimensional noise-reduced signal, and a difference value between the two-dimensional noise-reduced signal obtained in the two-dimensional noise reduction processing step and the noise-reduced signal one frame before A motion detection step of obtaining a motion amount in the time axis direction of each pixel by calculating the time, and the motion in the time axis direction of each pixel obtained in the motion detection step with respect to the image signal input from outside And a three-dimensional noise reduction processing step of performing a three-dimensional noise reduction process in which a gain is adjusted according to the amount to obtain a signal after the reduction of the three-dimensional noise.

  According to the present invention, in the noise reduction method configured as described above, the two-dimensional noise reduced signal obtained in the two-dimensional noise reduction processing step and the three-dimensional noise obtained in the three-dimensional noise reduction processing step. The weighted addition step of obtaining the noise-reduced signal by weighting and adding the reduced signal with a predetermined weighting factor, and the amount of motion in the time axis direction of each pixel obtained in the motion detection step It is desirable to include a weighting factor detection step for calculating a predetermined weighting factor.

  According to the present invention, in the noise reduction method configured as described above, the two-dimensional noise reduction processing performed in the two-dimensional noise reduction processing step according to the signal level of each pixel of the image signal input from the outside. It is desirable to have a signal level gain adjustment step for adjusting the gain of the signal level.

  According to the present invention, in the noise reduction method configured as described above, the two-dimensional noise reduction processing step is performed according to the amount of motion in the time axis direction of the target pixel one time before obtained in the motion detection step. It is desirable to have a motion amount gain adjustment step for adjusting the gain of the two-dimensional noise reduction processing.

  According to the present invention, in the noise reduction method configured as described above, in the two-dimensional noise reduction processing step, the two-dimensional processing includes processing by an edge-preserving noise reduction filter for the image signal input from the outside. It is desirable to perform noise reduction processing.

  According to the present invention, in the noise reduction method having the above configuration, in the two-dimensional noise reduction processing step, the two-dimensional noise reduction includes processing by an impulse noise reduction filter for the image signal input from the outside. It is desirable that processing be performed.

  According to the present invention, in the noise reduction method configured as described above, the three-dimensional noise reduction processing step calculates a difference value between the image signal input from the outside and the noise-reduced signal one frame before a predetermined value. A step of performing the three-dimensional noise reduction process using a noise detection value obtained by acting on a function, wherein the predetermined noise is determined in accordance with a degree of impulse noise contained in each pixel of the image signal input from the outside. It is desirable to have a function change step that changes the function.

  In the present invention, the electronic device is an electronic device provided with a noise reduction device that reduces a noise of the moving image while a moving image including a plurality of frames is given by external input or imaging. The reduction device is a noise reduction device having the above-described configuration.

  In the noise reduction device and the noise reduction method of the present invention, a two-dimensional noise-reduced signal obtained by performing a two-dimensional noise reduction process on an image signal input from the outside, and a noise after one frame before noise reduction In order to obtain the amount of motion in the time axis direction of each pixel by calculating a difference value from the signal, each difference value between the image signal input from the outside and the noise-reduced signal of one frame before is calculated. Compared with the case of obtaining the amount of motion of the pixel in the time axis direction, the calculation error of the amount of motion of each pixel in the time axis direction due to the influence of noise can be reduced. Therefore, a sufficient noise reduction effect can be obtained by the three-dimensional noise reduction process in which the gain is adjusted according to the amount of movement of each pixel in the time axis direction.

  In the noise reduction device and the noise reduction method of the present invention, the predetermined weight calculated according to the amount of motion of each pixel in the time axis direction is applied to the signal after the two-dimensional noise reduction and the signal after the three-dimensional noise reduction. In order to obtain a noise-reduced signal by weighted addition using a coefficient, a sufficient noise reduction effect can be obtained by a two-dimensional noise reduction process even for a pixel with a large amount of motion that does not sufficiently obtain a noise reduction effect by the three-dimensional noise reduction process. Will be obtained.

  In the noise reduction apparatus and noise reduction method of the present invention, the gain of the two-dimensional noise reduction processing is adjusted according to the signal level of each pixel of the image signal input from the outside. Sometimes, a sufficient noise reduction effect is obtained by increasing the gain of the two-dimensional noise reduction processing.

  In the noise reduction device and the noise reduction method of the present invention, since the gain of the two-dimensional noise reduction process is adjusted according to the amount of motion in the time axis direction of the pixel of interest before one point in time, the pixel of interest before one point in time The two-dimensional noise reduction processing with an appropriate gain can be performed using the amount of motion in the time axis direction, and a sufficient noise reduction effect can be obtained.

  Further, in the noise reduction device and noise reduction method of the present invention, random noise is effectively applied because the two-dimensional noise reduction processing including processing by the edge preserving type noise reduction filter is performed on the image signal input from the outside. Therefore, a sufficient noise reduction effect can be obtained.

  Further, in the noise reduction device and noise reduction method of the present invention, since the two-dimensional noise reduction processing including processing by the impulse noise reduction filter is performed on the image signal input from the outside, the smoothing filter sufficiently removes it. Impulse noise that cannot be performed can be effectively removed, and a sufficient noise reduction effect can be obtained.

  Further, in the noise reduction device and the noise reduction method of the present invention, different three-dimensional noise reduction processing is performed according to the impulse noise content of each pixel of the image signal input from the outside. An appropriate three-dimensional noise reduction process can be performed according to the content degree, and a sufficient noise reduction effect can be obtained.

  Embodiments of the present invention will be described with reference to the drawings. In the following description, an imaging apparatus such as a digital camera or digital video provided with a noise reduction apparatus (hereinafter, included in the “image processing unit”) that performs the noise reduction method of the present invention will be described as an example. As will be described later, a display device that performs digital processing of an image, such as a liquid crystal display or a plasma television, may be used as long as it has a similar noise reduction device.

(Configuration of imaging device)
First, the internal configuration of the imaging apparatus will be described with reference to the drawings. FIG. 1 is a block diagram illustrating an internal configuration of the imaging apparatus.

  The imaging apparatus of FIG. 1 is a solid-state imaging device (image sensor) 1 such as a CCD or CMOS sensor that converts light incident from an object into an electrical signal; and an image signal that is an analog signal output from the image sensor 1 is digitally converted. AFE (Analog FrontEnd) 2 that converts the signal into a signal; a microphone 3 that converts the sound input from the outside into an electric signal; and performs various image processing including noise reduction processing on the image signal that becomes a digital signal from the AFE 2 An image processing unit 4; an audio processing unit 5 that converts an audio signal that is an analog signal from the microphone 3 into a digital signal; an MPEG for the image signal from the image processing unit 4 and the audio signal from the audio processing unit 5 (Moving Picture Experts Group) A compression processing unit 6 that performs compression encoding processing such as a compression method; and a compressed encoded signal that has been compression encoded by the compression processing unit 6 A driver unit 7 to be recorded in the memory 20; a decompression processing unit 8 that decompresses and decodes the compressed encoded signal read from the external memory 20 by the driver unit 7; and an image signal obtained by decoding by the decompression processing unit 8. A display 9 that displays an image; an audio output circuit unit 10 that converts an audio signal from the decompression processing unit 8 into an analog signal; and a speaker 11 that reproduces and outputs audio based on the audio signal from the audio output circuit unit 10; A TG (Timing Generator) 12 that outputs a timing control signal for matching the operation timing of each block; a CPU (Central Processing Unit) 13 that controls the drive operation of the entire imaging apparatus; A memory 14 for storing a program and temporarily storing data when the program is executed; an operation unit 15 for inputting an instruction from a user; and a memory 1 It comprises; and a bus line 16 for exchanging data with each block; CPU 13 and the bus line 17 for exchanging data with each block.

  In this imaging apparatus, when the operation unit 15 instructs to perform an imaging operation, an image signal that is an analog signal obtained by the photoelectric conversion operation of the image sensor 1 is output to the AFE 2. At this time, the image sensor 1 receives the timing control signal from the TG 12 to perform horizontal scanning and vertical scanning, and output an image signal serving as data for each pixel. In the AFE 2, when an image signal that is an analog signal is converted into a digital signal and input to the image processing unit 4, various image processing such as noise reduction processing is performed. Details of the image processing unit 4 will be described later.

  Then, the image signal subjected to the image processing by the image processing unit 4 is given to the compression processing unit 6. At this time, an audio signal which is an analog signal obtained by inputting the sound into the microphone 3 is converted into a digital signal by the audio processing unit 5 and given to the compression processing unit 6. As a result, the compression processing unit 6 compresses and encodes the image signal and the audio signal, which are digital signals, based on the MPEG compression encoding method, gives the image signal and the audio signal to the driver unit 7 and records them in the external memory 20. At this time, the compressed signal recorded in the external memory 20 is read out by the driver unit 7 and applied to the expansion processing unit 8 to be subjected to expansion processing to obtain an image signal. This image signal is given to the display 9 and a subject image currently being photographed through the image sensor 1 is displayed.

  In the above description, the operation at the time of moving image shooting has been described. However, even when a still image shooting is instructed, an audio signal is not acquired by the microphone 3 and a compressed signal of only an image signal is recorded in the external memory 20. However, the basic operation is the same as that during moving image shooting. In the case of this still image shooting, not only the compressed signal for the still image shot by the operation unit 15 is recorded in the external memory 20, but also the compressed signal for the current image shot by the image sensor 1 is stored in the external memory. 20 is temporarily recorded. As a result, the compression signal for the currently photographed image is decompressed by the decompression processing unit 8 so that the current image photographed by the image sensor 1 is displayed on the display 9 and can be confirmed by the user.

  When performing the imaging operation in this way, the TG 12 gives a timing control signal to the AFE 2, the video processing unit 4, the audio processing unit 5, the compression processing unit 6, and the expansion processing unit 8, and one frame by the image sensor 1. An operation synchronized with each imaging operation is performed. At the time of still image shooting, a timing control signal is given from the TG 12 to the image sensor 1, the AFE 2, the video processing unit 4, and the compression processing unit 6 based on the shutter operation by the operation unit 15. The operation timing of each unit is synchronized.

  When an instruction to reproduce a moving image or image recorded in the external memory 20 is given through the operation unit 15, the compressed signal recorded in the external memory 20 is read out by the driver unit 7 and is decompressed by the decompression processing unit 8. Given to. Then, the decompression processing unit 8 decompresses and decodes the image signal and the audio signal based on the MPEG compression encoding method. Then, the image signal is given to the display 9 to reproduce the image, and the audio signal is given to the speaker 11 via the audio output circuit unit 10 to reproduce the audio. Thereby, the moving image based on the compressed signal recorded in the external memory 20 is reproduced together with the sound. Further, when the compressed signal consists only of the image signal, only the image is reproduced on the display 9.

(First Embodiment of Noise Reduction Device)
Next, a first embodiment of the noise reduction device included in the image processing unit 4 will be described. FIG. 2 is a block diagram showing a schematic configuration of the noise reduction device 30 according to the first embodiment of the present invention. The noise reduction device 30 receives a signal from the outside and the frame memory 35 and outputs a signal to the multiplier 36; a signal is input from the outside, and the signal is sent to the motion detection unit 33 and the weighting addition unit 38. A two-dimensional noise reduction processing unit 32 to output; a motion detection unit 33 that receives a signal from the two-dimensional noise reduction processing unit 32 and the frame memory 35 and outputs a signal to the feedback coefficient and synthesis coefficient detection unit 34; and a motion detection unit 33, a feedback coefficient and synthesis coefficient detector 34 that outputs a signal to the multiplier 36 and the weighted adder 38; and a signal that is input from the weighted adder 38 to the noise detector 31 and the motion detector 33. A frame memory 35 for outputting a signal; a signal is input from the noise detection unit 31 and the feedback coefficient and synthesis coefficient detection unit 34; A subtractor 37 that receives a signal from the outside and the multiplier 36 and outputs a signal to the weighted addition unit 38; a two-dimensional noise reduction processing unit 32, a feedback coefficient and synthesis coefficient detection unit 34; A weighted addition unit 38 that receives a signal from the subtractor 37 and outputs the signal to the subsequent stage and the frame memory 35.

  Here, an outline of the operation of the noise reduction device 30 will be described. First, the current frame signal from the outside is input to the noise detector 31, the two-dimensional noise reduction processor 32, and the subtractor 37. Further, a signal one frame before the current frame from the frame memory 35 is input to the noise detection unit 31 and the motion detection unit 33. Then, the noise detection unit 31 calculates the noise detection value of each pixel from the signal of the current frame input from the outside and the signal of the previous frame input from the frame memory 35, and outputs this to the multiplier 36. To do. The two-dimensional noise reduction processing unit 32 removes high-frequency components by applying a spatial low-pass filter to the current frame signal input from the outside, and moves the obtained two-dimensional noise-reduced signal. It outputs to the detection part 33 and the weighting addition part 38. FIG.

  Next, the motion detection unit 33 calculates the motion amount of each pixel from the two-dimensional noise-reduced signal input from the two-dimensional noise reduction processing unit 32 and the signal one frame before input from the frame memory 35. This is calculated and output to the feedback coefficient and synthesis coefficient detector 34. Then, the feedback coefficient and synthesis coefficient detector 34 calculates the feedback coefficient k1 and synthesis coefficient k2 of each pixel from the amount of motion of each pixel input from the motion detector 33, and combines the feedback coefficient k1 with the multiplier 36. The coefficient k2 is output to the weighted addition unit 38.

  Then, the multiplier 36 multiplies the noise detection value of each pixel input from the noise detection unit 31 by the feedback coefficient k1 of each pixel input from the feedback coefficient and synthesis coefficient detection unit 34, and obtains The subtracted noise value of each pixel is output to the subtractor 37. Then, the subtractor 37 subtracts the subtraction noise value of each pixel input from the multiplier 36 from the signal of the current frame input from the outside, and the obtained three-dimensional noise reduced signal is sent to the weighted addition unit 38. Output.

  Then, the weighted addition unit 38 receives the three-dimensional noise-reduced signal (E) input from the subtractor 37 and the two-dimensional noise-reduced signal (A) input from the two-dimensional noise reduction processing unit 32. Are weighted and added by the synthesis coefficient k2 of each pixel input from the feedback coefficient and synthesis coefficient detection unit 34 (that is, E × k2 + A × (1−k2) is calculated for each pixel), and the obtained noise is reduced. The signal is output to the subsequent stage and the frame memory 35.

Next, details of the noise detection unit 31 will be described. The noise detection unit 31 detects the noise detection value for the difference signal value x of each pixel obtained by subtracting the signal of the previous frame input from the frame memory 35 from the signal of the current frame input from the outside. f (x) is calculated and output to the multiplier 36. Here, as the function f (x), for example, the following equation 1 and those shown in FIG. 3 are used.

  The function f (x) shown in Equation 1 and FIG. 3 suppresses the noise detection value f (x) when the absolute value of the difference signal value x of each pixel exceeds the threshold th, and the motion f A malfunction that erroneously detects a part as noise can be prevented.

  Next, details of the two-dimensional noise reduction processing unit 32 will be described. FIG. 4 is a block diagram illustrating a schematic configuration of the two-dimensional noise reduction processing unit 32. The two-dimensional noise reduction processing unit 32 is a filter unit 321 that receives a signal from outside and outputs a signal to the noise reduction unit 323; and a filter weight calculation unit that receives a signal from outside and outputs a signal to the noise reduction unit 323 A noise reduction unit 323 that receives a signal from the outside, the filter unit 321, and the filter weight calculation unit 322 and outputs the signal to the motion detection unit 33.

  Here, an outline of the operation of the two-dimensional noise reduction processing unit 32 will be described. First, an external current frame signal is input to the filter unit 321, the filter weight calculation unit 322, and the noise reduction unit 323. Then, the filter unit 321 removes high-frequency components by applying a spatial low-pass filter to the current frame signal input from the outside, and outputs the obtained filtered signal to the noise reduction unit 323. Also, the filter weight calculation unit 322 calculates a filter weight coefficient for each pixel from the signal level of the signal of the current frame input from the outside, and outputs this to the noise reduction unit 323.

  Next, the noise reduction unit 323 receives the filtered signal (NR) input from the filter unit 321 and the current frame signal (ORG) input from the outside from the filter weight calculation unit 322. Weighted addition is performed by the filter weight coefficient (GAIN) of each pixel (that is, NR × GAIN + ORG × (1−GAIN) is calculated for each pixel), and the obtained two-dimensional noise-reduced signal is output to the motion detection unit 33. .

  Next, an edge-preserving noise reduction filter that is an example of a spatial low-pass filter used in the filter unit 321 will be described with reference to FIG. In the following, the coordinates of the pixel of interest are represented by (0, 0), the coordinates of the pixel located at the x pixel on the right and the y pixel above the pixel of interest are represented by (x, y). That is, the coordinates of the pixel at the position of two pixels to the right and the upper two pixels from the target pixel are represented as (2, 2), and the coordinates of the two pixels to the left and the lower pixel from the target pixel are expressed as (2, 2). It will be expressed as (-2, -2).

  The edge-preserving noise reduction filter is a filter that uses a signal value of 5 × 5 pixels centered on the pixel of interest. Then, (a) to (i) in FIG. 5 are regarded as 5 × 5 pixels centered on the target pixel, and the signal value dispersion of the pixels corresponding to the hatched portions in (a) to (i) of FIG. Calculate, select one with the smallest variance, and calculate the average value of the signal values of the pixels corresponding to the one shaded portion of the selected (a) to (i) as the signal after filtering the target pixel Value.

  Specifically, for coordinates (0, 0) of the target pixel, coordinates (−1, 2), (0, 2), (1, 2), (−1, 1), (0, 1) , (1, 1), (0, 0) of 7 pixel signal values (corresponding to (a) in FIG. 5), coordinates (1, 2), (2, 2), (0, 1) , (1, 1), (2, 1), (0, 0), and (1, 0) of 7 pixel signal values (corresponding to (b) in FIG. 5), coordinates (1, 1) , (2, 1), (0, 0), (1, 0), (2, 0), (1, -1), and (2, -1) signal value variance (FIG. 5). (Corresponding to (c)), coordinates (0, 0), (1, 0), (0, -1), (1, -1), (2, -1), (1, -2), ( 2 and -2) of 7 pixel signal values (corresponding to (d) in FIG. 5), coordinates (0, 0), (-1, -1), (0, -1), (1, -1), (-1 -2), (0, -2), (1, -2) signal value variance of 7 pixels (corresponding to (e) of FIG. 5), coordinates (-1, 0), (0, 0) , (−2, −1), (−1, −1), (0, −1), (−2, −2), and (−1, −2) signal value dispersion of 7 pixels (see FIG. 5 (corresponding to (f)), coordinates (−2, 1), (−1, 1), (−2, 0), (−1, 0), (0, 0), (−2, −1) ), (−1, −1) signal dispersion of 7 pixels (corresponding to (g) in FIG. 5), coordinates (−2, 2), (−1, 2), (−2, 1) , (−1, 1), (0, 1), (−1, 0), (0, 0), the dispersion of signal values of 7 pixels (corresponding to (h) in FIG. 5), coordinates (−1 1), (0, 1), (1, 1), (-1, 0), (0, 0), (1, 0), (-1, -1), (0, -1), 9 pixel signals (1, -1) Calculate the variance of the values (corresponding to (i) in FIG. 5), select the one with the smallest variance, and correspond to the one shaded portion of the selected (a) to (i) An average value of signal values of pixels to be processed is set as a signal value after filtering of the target pixel.

  According to the edge preserving noise reduction filter, it is possible to effectively reduce random noise.

Next, the operation of the filter weight calculation unit 322 will be described. The filter weight calculation unit 322 calculates an average value of signal values of 5 × 5 pixels centered on the pixel of interest as a signal level of the pixel of interest for the current frame signal input from the outside, The filter weight coefficient GAIN = g (y) of each pixel is calculated for the signal level y, and this is output to the noise reduction unit 323. Here, as the function g (y), the following equation 2 and those shown in FIG. 6 are used.

  Equation 2 and the function g (y) shown in FIG. 6 decrease the filter weight coefficient from 1 to thβ as a linear function as the signal level y increases until the signal level y reaches thα. Therefore, when the signal level y is low, the influence of the spatial low-pass filter used in the filter unit 321 becomes large, and when the signal level y is high, the influence of the spatial low-pass filter used in the filter unit 321 becomes small. The accuracy of motion detection can be improved.

  Next, details of the motion detection unit 33 will be described. FIG. 7 is a block diagram illustrating a schematic configuration of the motion detection unit 33. The motion detection unit 33 receives a signal from the two-dimensional noise reduction processing unit 32 and the frame memory 35, and outputs a signal to the absolute value conversion unit 332; a signal is input from the subtractor 331, and the filter unit 333 And a filter unit 333 that receives a signal from the absolute value conversion unit 332 and outputs a signal to the feedback coefficient and synthesis coefficient detection unit 34.

  Here, an outline of the operation of the motion detection unit 33 will be described. First, the signal after two-dimensional noise reduction from the two-dimensional noise reduction processing unit 32 and the signal one frame before from the frame memory 35 are input to the subtractor 331. Then, the subtractor 331 calculates the difference between the two-dimensional noise-reduced signal input from the two-dimensional noise reduction processing unit 32 and the signal one frame before input from the frame memory 35, and the obtained difference signal Is output to the absolute value converting unit 332.

  Next, the absolute value converting unit 332 converts the difference signal input from the subtractor 331 into an absolute value, and outputs the obtained absolute value signal to the filter unit 333. Then, the filter unit 333 outputs the amount of motion of each pixel obtained by applying a low-pass filter to the absolute value signal input from the absolute value unit 332 to the feedback coefficient and synthesis coefficient detection unit 34. Here, by applying a low-pass filter by the filter unit 333, the separation accuracy between the noise signal and the signal indicating the motion is increased.

Next, details of the feedback coefficient and synthesis coefficient detection unit 34 will be described. The feedback coefficient and synthesis coefficient detection unit 34 calculates the feedback coefficient k1 = h (z) of each pixel with respect to the motion amount z of each pixel input from the motion detection unit 33, and outputs this to the multiplier 36. . Here, as the function h (z), for example, the following equation 3 and those shown in FIG. 8 are used.

  Equation (3) and the function h (z) shown in FIG. 8 indicate that when the motion amount z is equal to or greater than the threshold value tha, the feedback coefficient k1 decreases from 1 as a linear function as the motion amount z increases, and the motion amount z Is equal to or greater than the threshold thb, the feedback coefficient k1 is 0, thereby increasing the noise reduction effect by the noise detection unit 31 on the pixel with a small amount of motion z, that is, the effect of the three-dimensional noise reduction process. The noise reduction effect by the noise detection unit 31 for pixels with a large amount of motion z, that is, the effect of the three-dimensional noise reduction processing is reduced.

Further, the feedback coefficient and synthesis coefficient detection unit 34 calculates the synthesis coefficient k2 = j (z) of each pixel with respect to the motion amount z of each pixel input from the motion detection unit 33, and uses this to calculate the weighting addition unit 38. Output to. Here, as the function j (z), for example, the following equation 4 and those shown in FIG. 9 are used.

  In equation 4 and the function j (z) shown in FIG. 9, when the amount of motion z is equal to or greater than the threshold thA, the synthesis coefficient k2 decreases from 1 as a linear function as the amount of motion z increases, and the amount of motion z Is equal to or greater than the threshold thB, the synthesis coefficient k2 becomes 0. With this, for a pixel with a small amount of motion z, the noise reduction effect by the noise detection unit 31, that is, the effect of the three-dimensional noise reduction processing. On the other hand, the noise reduction effect by the two-dimensional noise reduction processing unit 32 is reduced, and the noise reduction effect by the noise detection unit 31, that is, the effect of the three-dimensional noise reduction processing is applied to a pixel having a large motion amount z. On the other hand, the noise reduction effect by the two-dimensional noise reduction processing unit 32 is increased. Here, the function h (z) and the function j (z) may be the same function.

  In the noise reduction device 30 having the above-described configuration, the motion detection unit 33 uses the two-dimensional noise reduction processing unit 32 to spatially convert the current frame signal input from the outside to the current frame signal input from the outside. Since the motion amount of each pixel is calculated from the two-dimensional noise-reduced signal obtained by applying the low-pass filter and the signal one frame before input from the frame memory 35, a large noise is generated in the input signal. Even if it occurs, an increase in motion amount calculation error due to the influence of the noise can be suppressed, and the effect of the three-dimensional noise reduction processing can be enhanced.

(Second Embodiment of Noise Reduction Device)
Next, a second embodiment of the noise reduction device included in the image processing unit 4 will be described. FIG. 10 is a block diagram showing a schematic configuration of a noise reduction device 40 according to the second embodiment of the present invention. In addition, in the noise reduction apparatus 40, the same code | symbol is attached | subjected to the part same as the noise reduction apparatus 30 which concerns on 1st Embodiment, and the detailed description is abbreviate | omitted. The noise reduction device 40 has a configuration different from that of the noise reduction device 30, and a noise detection unit 41 that receives signals from the outside, the frame memory 35, and the two-dimensional noise reduction processing unit 42 and outputs a signal to the multiplier 36; And a two-dimensional noise reduction processing unit 42 that outputs a signal to the motion detection unit 33, the weighting addition unit 38, and the noise detection unit 41.

  Here, an outline of the operation of the noise reduction device 40 will be described. First, the current frame signal from the outside is input to the subtractor 37, the noise detection unit 41, and the two-dimensional noise reduction processing unit 42. In addition, a signal one frame before from the frame memory 35 is input to the noise detection unit 41 and the motion detection unit 33. Then, the two-dimensional noise reduction processing unit 42 removes high-frequency components by applying a spatial low-pass filter to the current frame signal input from the outside, and moves the obtained two-dimensional noise-reduced signal. It outputs to the detection part 33 and the weighting addition part 38. FIG. The two-dimensional noise reduction processing unit 42 determines whether each pixel includes impulse noise and outputs the determination result to the noise detection unit 41.

  Next, the noise detection unit 41 inputs the signal of the current frame input from the outside in accordance with the determination result of whether or not each pixel includes impulse noise input from the two-dimensional noise reduction processing unit 42. Then, the noise detection value of each pixel is calculated from the signal of the previous frame input from the frame memory 35, and this is output to the multiplier 36.

  Next, the motion detection unit 33 calculates the motion amount of each pixel from the two-dimensional noise-reduced signal input from the two-dimensional noise reduction processing unit 42 and the signal one frame before input from the frame memory 35. This is calculated and output to the feedback coefficient and synthesis coefficient detector 34. Then, the feedback coefficient and synthesis coefficient detector 34 calculates the feedback coefficient k1 and synthesis coefficient k2 of each pixel from the amount of motion of each pixel input from the motion detector 33, and combines the feedback coefficient k1 with the multiplier 36. The coefficient k2 is output to the weighted addition unit 38.

  Then, the multiplier 36 multiplies the noise detection value of each pixel input from the noise detection unit 41 by the feedback coefficient k1 of each pixel input from the feedback coefficient and synthesis coefficient detection unit 34, and obtains The subtracted noise value of each pixel is output to the subtractor 37. Then, the subtractor 37 subtracts the subtraction noise value of each pixel input from the multiplier 36 from the signal of the current frame input from the outside, and the obtained three-dimensional noise reduced signal is sent to the weighted addition unit 38. Output.

  Next, the weighted addition unit 38 receives the three-dimensional noise-reduced signal (E) input from the subtractor 37, and the two-dimensional noise-reduced signal (A) input from the two-dimensional noise reduction processing unit 42. Are weighted and added by the synthesis coefficient k2 of each pixel input from the feedback coefficient and synthesis coefficient detection unit 34 (that is, E × k2 + A × (1−k2) is calculated for each pixel), and the obtained noise is reduced. The signal is output to the subsequent stage and the frame memory 35.

  Next, details of the two-dimensional noise reduction processing unit 42 will be described. FIG. 11 is a block diagram illustrating a schematic configuration of the two-dimensional noise reduction processing unit 42. The two-dimensional noise reduction processing unit 42 receives an external signal and outputs an impulse noise processing unit 421 that outputs a signal to the filter unit 321, the filter weight calculation unit 322, the noise reduction unit 323, and the noise detection unit 41; The same filter unit 321 as in the first embodiment, which receives a signal from the processing unit 421 and outputs a signal to the noise reduction unit 323; receives a signal from the impulse noise processing unit 421, and outputs a signal to the noise reduction unit 323 The same filter weight calculation unit 322 as that of the first embodiment is output; the signal is input from the filter unit 321, the filter weight calculation unit 322, and the impulse noise processing unit 421, and the signal is output to the motion detection unit 33. And the same noise reduction unit 323 as in the first embodiment.

  Here, an outline of the operation of the two-dimensional noise reduction processing unit 42 will be described. First, an external current frame signal is input to the impulse noise processing unit 421. Then, the impulse noise processing unit 421 determines whether or not impulse noise is included in each pixel of the signal of the current frame input from the outside, and outputs the determination result to the noise detection unit 41. In addition, the impulse noise processing unit 421 applies an impulse noise reduction filter only to pixels that are determined to include impulse noise, and the obtained post-impulse filter signal is converted into a filter unit 321, a filter weight calculation unit 322, a noise The data is output to the reduction unit 323.

  Then, the filter unit 321 removes high frequency components by applying a spatial low-pass filter to the post-impulse filter signal input from the impulse noise processing unit 421, and the obtained post-filter signal is subjected to noise. Output to the reduction unit 323. Further, the filter weight calculation unit 322 calculates the filter weight coefficient of each pixel from the signal level of the signal after the impulse filter input from the impulse noise processing unit 421, and outputs this to the noise reduction unit 323.

  Next, the noise reduction unit 323 uses the filtered signal (NR) input from the filter unit 321 and the impulse filtered signal (ORG) input from the impulse noise processing unit 421 as the filter weight calculation unit 322. The weighted addition is performed by the filter weighting coefficient (GAIN) of each pixel input (that is, NR × GAIN + ORG × (1−GAIN) is calculated for each pixel), and the obtained two-dimensional noise-reduced signal is used as a motion detection unit. To 33.

  Next, details of the impulse noise processing unit 421 will be described. The impulse noise processing unit 421 first determines whether or not impulse noise is included in each pixel of the signal of the current frame input from the outside. As this determination method, when the impulse noise determination value obtained by subtracting the average value of the signal values of the eight pixels surrounding the target pixel from the signal value of the target pixel is equal to or greater than a predetermined threshold, the impulse is applied to the target pixel. While determining that noise is included, there is known a method for determining that the target pixel does not include impulse noise when the impulse noise determination value is less than a predetermined threshold. Then, the impulse noise processing unit 421 outputs a determination result as to whether or not each pixel includes impulse noise to the noise detection unit 41.

  Further, the impulse noise determination processing unit 421 operates the impulse noise reduction filter only for pixels determined to include the impulse noise, and the obtained post-impulse filter signal is filtered by a filter unit 321, a filter weight calculation unit 322, Output to the noise reduction unit 323. Here, as the impulse noise reduction filter, for example, a median filter that can suppress the influence of an abnormal value by using an intermediate value of the signal value is used.

  As described above, the two-dimensional noise reduction processing unit 42 includes the impulse noise processing unit 421 and applies the impulse noise reduction filter to pixels determined to include the impulse noise, and thus cannot be completely removed by the smoothing filter. Therefore, it is possible to effectively remove the impulse noise that is regarded as the amount of movement.

  The two-dimensional noise reduction processing unit 42 includes the same filter unit 321 as in the first embodiment; the same filter weight calculation unit 322 as in the first embodiment; and the same noise reduction unit as in the first embodiment. Therefore, not only impulse noise but also random noise can be effectively removed.

  Next, details of the noise detection unit 41 will be described. The noise detection unit 41 detects a noise detection value for the difference signal value x of each pixel obtained by subtracting the signal of the previous frame input from the frame memory 35 from the signal of the current frame input from the outside. f (x) is calculated and output to the multiplier 36. Here, the noise detection unit 41 is input from the impulse noise processing unit 421 of the two-dimensional noise reduction processing unit 42, based on the determination result of whether each pixel includes impulse noise or not, based on Equation 1 and FIG. 3 changes the value of the threshold value th of the function f (x).

  Specifically, the noise detection unit 41 increases the th value of the function f (x) for pixels determined to include impulse noise, and for pixels determined to include no impulse noise. , The value of th of the function f (x) is decreased. In this way, by changing the th value of the function f (x) according to the presence or absence of impulse noise, the noise detection unit 41 can efficiently remove the impulse noise.

(Third Embodiment of Noise Reduction Device)
Next, a third embodiment of the noise reduction device included in the image processing unit 4 will be described. FIG. 12 is a block diagram showing a schematic configuration of a noise reduction device 50 according to the third embodiment of the present invention. In the noise reduction device 50, the same parts as those of the noise reduction device 30 according to the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted. The noise reduction device 50 has a different configuration from the noise reduction device 30 and a two-dimensional noise reduction processing unit 52 that receives signals from the outside and the motion detection unit 53 and outputs signals to the weighting addition unit 38 and the motion detection unit 53; A motion detection unit 53 that receives a signal from the frame memory 35 and the two-dimensional noise reduction processing unit 52 and outputs a signal to the feedback coefficient / combination coefficient detection unit 34 and the two-dimensional noise reduction processing unit 52.

  Here, an outline of the operation of the noise reduction device 50 will be described. First, an external current frame signal is input to the noise detector 31, the subtractor 37, and the two-dimensional noise reduction processor 52. Further, the signal one frame before from the frame memory 35 is input to the noise detection unit 31 and the motion detection unit 53. Then, the noise reduction processing unit 31 calculates the noise detection value of each pixel from the signal of the current frame input from the outside and the signal of the previous frame input from the frame memory 35, and supplies this to the multiplier 36. Output. In addition, the two-dimensional noise reduction processing unit 52 applies a spatial low-pass filter to the signal of the current frame input from the outside according to the amount of motion of the target pixel one time before input from the motion detection unit 53. The high frequency component is removed by the action, and the obtained two-dimensional noise-reduced signal is output to the weighted addition unit 38 and the motion detection unit 53.

  Next, the motion detection unit 53 calculates the motion amount of each pixel from the two-dimensional noise-reduced signal input from the two-dimensional noise reduction processing unit 52 and the signal one frame before input from the frame memory 35. This is calculated and output to the feedback coefficient and synthesis coefficient detection unit 34 and the two-dimensional noise reduction processing unit 52. Then, the feedback coefficient and synthesis coefficient detector 34 calculates the feedback coefficient k1 and synthesis coefficient k2 of each pixel from the amount of motion of each pixel input from the motion detector 33, and combines the feedback coefficient k1 with the multiplier 36. The coefficient k2 is output to the weighted addition unit 38.

  Then, the multiplier 36 multiplies the noise detection value of each pixel input from the noise detection unit 31 by the feedback coefficient k1 of each pixel input from the feedback coefficient and synthesis coefficient detection unit 34, and obtains The subtracted noise value of each pixel is output to the subtractor 37. Then, the subtractor 37 subtracts the subtraction noise value of each pixel input from the multiplier 36 from the signal of the current frame input from the outside, and the obtained three-dimensional noise reduced signal is sent to the weighted addition unit 38. Output.

  Then, the weighted addition unit 38 receives the three-dimensional noise-reduced signal (E) input from the subtractor 37, and the two-dimensional noise-reduced signal (A) input from the two-dimensional noise reduction processing unit 52. Are weighted and added by the synthesis coefficient k2 of each pixel input from the feedback coefficient and synthesis coefficient detection unit 34 (that is, E × k2 + A × (1−k2) is calculated for each pixel), and the obtained noise is reduced. The signal is output to the subsequent stage and the frame memory 35.

  Next, details of the two-dimensional noise reduction processing unit 52 will be described. FIG. 13 is a block diagram illustrating a schematic configuration of the two-dimensional noise reduction processing unit 52. The two-dimensional noise reduction processing unit 52 receives the signal from the outside and outputs the signal to the noise reduction unit 523, and the same filter unit 321 as in the first embodiment; the signal is input from the outside and the noise reduction unit 523 The same filter weight calculation unit 322 as that of the first embodiment is output, and the signal is input from the outside, the filter unit 321, the filter weight calculation unit 322, and the motion detection unit 53, and the signal is input to the motion detection unit 53 And a noise reduction unit 523 for outputting.

  Here, an outline of the operation of the two-dimensional noise reduction processing unit 52 will be described. First, an external current frame signal is input to the filter unit 321, the filter weight calculation unit 322, and the noise reduction unit 523. Then, the filter unit 321 removes high frequency components by applying a spatial low-pass filter to the current frame signal input from the outside, and outputs the obtained filtered signal to the noise reduction unit 523. The filter weight calculation unit 322 calculates a filter weight coefficient for each pixel from the signal level of the signal of the current frame input from the outside, and outputs this to the noise reduction unit 523.

  Then, the noise reduction unit 523 calculates the noise reduction adjustment amount ADJ = p (v) of the current pixel of interest with respect to the movement amount v of the pixel of interest one point before input from the motion detection unit 53, The filtered signal (NR) input from the filter unit 321 and the signal (ORG) of the current frame input from the outside, the filter weight coefficient (GAIN) of each pixel input from the filter weight calculation unit 322, and each Weighted addition is performed by the sum of the pixel noise reduction adjustment amount (ADJ) (that is, NR × (ADJ + GAIN) + ORG × (1−ADJ−GAIN) is calculated), and motion detection is performed on the obtained two-dimensional noise-reduced signal. To the unit 53. Here, ADJ + GAIN is adjusted to be 0 or more and 1 or less.

Here, as the function p (v), for example, the following equation 5 and those shown in FIG. 14 are used.

  Equation 5 and the function p (v) shown in FIG. 14 are linear functions as the amount of motion v of the pixel of interest before one point of time increases until the amount of motion v of the pixel of interest before one point of time reaches thz. In particular, when the noise reduction adjustment amount ADJ of the current pixel of interest increases from −thτ and the movement amount v of the pixel of interest one time before becomes equal to or greater than thz, the noise reduction adjustment amount ADJ becomes thτ. When the motion amount v of the pixel of interest before one point in time is small, the influence of the spatial low-pass filter used in the filter unit 321 on the current pixel of interest can be reduced, and the amount of motion of the pixel of interest before one point in time When v is large, the influence of the spatial low-pass filter used in the filter unit 321 on the current target pixel can be increased.

  In the above description, the noise reduction method of the present invention has been described by taking the imaging apparatus having the configuration shown in FIG. 1 as an example. However, the present invention is not limited to the imaging apparatus, and digital images such as liquid crystal displays and plasma televisions are used. The noise reduction method according to the present invention can also be used in a display device that performs processing. FIG. 15 shows a display device including a noise reduction device (hereinafter, included in “image processing unit”) that performs the noise reduction method of the present invention.

  The display device shown in FIG. 15 is similar to the image pickup device shown in FIG. 1, the image processing unit 4, the expansion processing unit 8, the display 9, the audio output circuit unit 10, the speaker 11, the TG 12, the CPU 13, the memory 14, the operation unit 15, And bus lines 16 and 17. A tuner unit 21 that selects a broadcast signal received externally, a demodulator unit 22 that demodulates a broadcast signal selected by the tuner unit 21, and an interface to which a compressed signal that is a digital signal input from the outside is input. 23.

  In the display device of FIG. 15, when receiving a broadcast signal, the tuner unit 21 selects a broadcast signal of a desired channel, and then the demodulator 22 demodulates the broadcast signal, thereby compressing by the MPEG compression coding method. A digital signal is obtained as a signal. When this digital signal is given to the decompression processing unit 8, the digital signal which is a compressed signal is subjected to decompression processing by the MPEG compression coding method.

  When an image display is instructed by the operation unit 15, the image signal obtained by the expansion processing by the expansion processing unit 8 is given to the image processing unit 4, and various image processing including the above-described noise reduction processing is performed. After being applied, the obtained display signal is given to the display 9 to display an image. Also, the audio signal obtained by the expansion process of the expansion processing unit 8 is given to the speaker 11 through the audio output circuit unit 10, so that the sound is output.

  In addition, this invention is not limited to said embodiment, A various change is possible in the range which does not deviate from the main point of this invention. For example, in the above embodiment, the median filter is used as the impulse noise reduction filter. However, other filters can be used as long as the filter can reduce the impulse noise. Further, the two-dimensional noise reduction processing unit 52 of the noise reduction device 50 according to the third embodiment has an impulse noise processing unit 421 similarly to the noise reduction processing unit 42 of the noise reduction device 40 according to the second embodiment, The form of the function f (x) used in the noise detection unit 31 may be changed according to the determination result of whether or not each pixel includes impulse noise.

  INDUSTRIAL APPLICABILITY The present invention is effective as a noise reduction device and a noise reduction method that reduce noise in a moving image composed of a plurality of frames. The present invention is also effective as an electronic device equipped with this noise reduction device.

The block diagram which shows the internal structure of an imaging device. FIG. 2 is a block diagram showing a schematic configuration of a noise reduction device 30. The graph which shows function f (x). FIG. 3 is a block diagram showing a schematic configuration of a two-dimensional noise reduction processing unit 32. An explanatory view showing an edge preservation type noise reduction filter. The graph which shows the function g (y). FIG. 3 is a block diagram showing a schematic configuration of a motion detection unit 33. The graph which shows the function h (z). The graph which shows function j (z). FIG. 2 is a block diagram showing a schematic configuration of a noise reduction device 40. FIG. 3 is a block diagram showing a schematic configuration of a two-dimensional noise reduction processing unit 42. FIG. 2 is a block diagram showing a schematic configuration of a noise reduction device 50. FIG. 3 is a block diagram showing a schematic configuration of a two-dimensional noise reduction processing unit 52. The graph which shows the function p (v). The block diagram which shows the internal structure of a display apparatus. FIG. 2 is a block diagram showing a schematic configuration of a noise reduction device 90.

Explanation of symbols

30, 40, 50 Noise reduction device 32, 42, 52 Two-dimensional noise reduction processing unit 33, 53 Motion detection unit 34 Feedback coefficient and synthesis coefficient detection unit (weight coefficient detection unit)
38 Weighting addition unit 322 Filter weight calculation unit (signal level gain adjustment unit)
421 Impulse noise processing unit (function changing unit)
523 Noise reduction unit (motion gain adjustment unit)

Claims (9)

A noise reduction device that receives an image signal from the outside and outputs a signal after noise reduction,
A two-dimensional noise reduction processing unit that performs a two-dimensional noise reduction process on the image signal input from the outside and obtains a signal after two-dimensional noise reduction;
Movement for obtaining a motion amount in the time axis direction of each pixel by calculating a difference value between the two-dimensional noise-reduced signal input from the two-dimensional noise reduction processing unit and the noise-reduced signal one frame before A detection unit;
The image signal input from the outside is subjected to a three-dimensional noise reduction process in which a gain is adjusted in accordance with the amount of motion in the time axis direction of each pixel input from the motion detector, and after the three-dimensional noise reduction A three-dimensional noise reduction processing unit for obtaining a signal;
Have,
The two-dimensional noise reduction processing unit performs the two-dimensional noise reduction processing including processing by an edge-preserving noise reduction filter on the image signal input from the outside . Weighting and adding the two-dimensional noise-reduced signal input from the two-dimensional noise reduction processing unit and the three-dimensional noise-reduced signal input from the three-dimensional noise reduction processing unit using a predetermined weighting factor. A weighted addition unit for obtaining the noise-reduced signal by:
A weighting factor detection unit that calculates the predetermined weighting factor according to the amount of motion in the time axis direction of each pixel input from the motion detection unit;
The noise reduction device according to claim 1, comprising:   And a signal level gain adjusting unit that adjusts a gain of the two-dimensional noise reduction processing performed by the two-dimensional noise reduction processing unit according to a signal level of each pixel of the image signal input from the outside. The noise reduction device according to claim 1 or 2.   The amount of motion for adjusting the gain of the two-dimensional noise reduction processing performed by the two-dimensional noise reduction processing unit according to the amount of motion in the time axis direction of the pixel of interest one point before input from the motion detection unit The noise reduction device according to claim 1, further comprising a gain adjustment unit. The said two-dimensional noise reduction process part performs the said two-dimensional noise reduction process including the process by an impulse noise reduction filter with respect to the said image signal input from the outside. The noise reduction device according to item 1. The three-dimensional noise reduction processing unit uses a noise detection value obtained by applying a difference value between the image signal input from the outside and the noise-reduced signal one frame before to a predetermined function. Performing the three-dimensional noise reduction process,
  6. The function changing unit according to claim 1, further comprising a function changing unit that changes the predetermined function according to a degree of impulse noise contained in each pixel of the image signal input from the outside. Noise reduction device. A noise reduction method in which an image signal is input from the outside and a signal after noise reduction is output,
A two-dimensional noise reduction processing step for performing a two-dimensional noise reduction process on the image signal input from the outside to obtain a signal after the two-dimensional noise reduction;
A motion for obtaining a motion amount in the time axis direction of each pixel by calculating a difference value between the signal after the two-dimensional noise reduction obtained in the two-dimensional noise reduction processing step and the signal after the noise reduction one frame before. A detection step;
After the three-dimensional noise reduction, the image signal input from the outside is subjected to a three-dimensional noise reduction process in which a gain is adjusted in accordance with the amount of movement of each pixel in the time axis direction obtained in the motion detection step. A three-dimensional noise reduction processing step for obtaining a signal;
Have
In the two-dimensional noise reduction processing step, the two-dimensional noise reduction processing including processing by an edge preserving noise reduction filter is performed on the image signal input from the outside . Weighting and adding the two-dimensional noise-reduced signal obtained in the two-dimensional noise reduction processing step and the three-dimensional noise-reduced signal obtained in the three-dimensional noise reduction processing step by a predetermined weighting factor. A weighted addition step of obtaining a signal after noise reduction by:
A weighting factor detection step for calculating the predetermined weighting factor according to the amount of motion in the time axis direction of each pixel obtained in the motion detection step;
The noise reduction method according to claim 7, further comprising: A moving image constituted by a plurality of frames is provided by external input or imaging, and is an electronic device including a noise reduction device that reduces noise of the moving image,
The electronic apparatus according to claim 1, wherein the noise reduction apparatus is the noise reduction apparatus according to claim 1.

Images