WO2013027723A1 - ノイズ除去装置、ノイズ除去方法及びプログラム - Google Patents
ノイズ除去装置、ノイズ除去方法及びプログラム Download PDFInfo
- Publication number
- WO2013027723A1 WO2013027723A1 PCT/JP2012/071072 JP2012071072W WO2013027723A1 WO 2013027723 A1 WO2013027723 A1 WO 2013027723A1 JP 2012071072 W JP2012071072 W JP 2012071072W WO 2013027723 A1 WO2013027723 A1 WO 2013027723A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- pixel
- value
- area
- hierarchy
- average value
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012937 correction Methods 0.000 claims description 88
- 238000004364 calculation method Methods 0.000 claims description 27
- 238000012545 processing Methods 0.000 claims description 19
- 238000013459 approach Methods 0.000 claims description 13
- 230000003247 decreasing effect Effects 0.000 abstract 1
- 230000006870 function Effects 0.000 description 67
- 238000010586 diagram Methods 0.000 description 19
- 238000000605 extraction Methods 0.000 description 17
- 230000000694 effects Effects 0.000 description 3
- 238000012935 Averaging Methods 0.000 description 2
- 239000002131 composite material Substances 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/40—Picture signal circuits
- H04N1/409—Edge or detail enhancement; Noise or error suppression
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20076—Probabilistic image processing
Definitions
- the present invention relates to a noise removing device, a noise removing method, and a program.
- a technique for reducing random noise included in an image is an indispensable technique for reproducing a captured image more clearly.
- a technique for reducing random noise for example, there is a technique disclosed in Patent Document 1.
- Patent Document 1 is a plurality of arithmetic circuits that calculate a moving average pixel number n based on a predetermined formula for an arbitrary pixel of interest i in the main scanning direction of a color digital signal output from an input image processing circuit.
- a plurality of bit selection circuits that selectively output the target pixel i and the reference pixel j of the preceding and following n pixels, and a plurality of absolute values of differences between the output level of the target pixel i and each output level of the reference pixel j
- the circuit includes a plurality of arithmetic circuits that perform moving average processing of output signals output from the plurality of determination circuits.
- the reference pixel j is added to the moving averaging process only when the absolute value of the difference between the output level of the target pixel i and the output level of the reference pixel j is equal to or less than the threshold value, A portion where the absolute value of the difference changes steeply beyond a threshold value is excluded from the moving averaging process, and thereby noise components can be effectively removed.
- Patent Document 1 cannot remove low-frequency noise having periodicity equal to or greater than the smoothing filter size.
- the present invention has been invented in view of the above problems, and an object thereof is to provide a noise removing device, a noise removing method, and a program capable of effectively removing noise.
- the present invention obtains a pixel statistical value of a pixel in each multi-hierarchy area that includes a target pixel and the range is sequentially narrowed, and sequentially in each hierarchy, the area in the previous hierarchy having a wider range than the area in the hierarchy.
- This is a noise removal method that corrects the pixel statistical value of the region in the hierarchy using the corrected pixel statistical value and corrects the pixel of interest using the corrected pixel statistical value of the minimum range region.
- the present invention includes a pixel statistic value calculation unit that calculates a pixel statistic value of a pixel in a multi-hierarchy area including a target pixel and whose range is sequentially narrowed, and sequentially in each hierarchy, compared to the area of the hierarchy.
- the corrected pixel statistic of the area in the previous hierarchy is corrected using the corrected pixel statistic of the area in the previous hierarchy with a wide range, and the pixel of interest is corrected using the corrected pixel statistic in the area of the minimum range.
- a noise removing device having a correcting means.
- the present invention provides a computer that calculates a pixel statistic value of pixels in a multi-hierarchy area that includes a pixel of interest and whose range is sequentially narrowed. Using the corrected pixel statistic of the area in the previous previous layer, and correcting the pixel statistic in the area in the current hierarchy, and using the corrected pixel statistic in the area of the minimum range, This is a program for executing a correction process.
- FIG. 1 is a diagram for explaining the processing of the first embodiment.
- FIG. 2 is a diagram showing an example of the Func function.
- FIG. 3 is a block diagram of the noise removing apparatus according to the first embodiment.
- FIG. 4 is a diagram for explaining the processing of the second embodiment.
- FIG. 5 is a block diagram of a noise removing apparatus according to the second embodiment.
- FIG. 6 is a diagram for explaining the processing of the third embodiment.
- FIG. 7 is a block diagram of a noise removing apparatus according to the third embodiment.
- FIG. 8 is a diagram for explaining the processing of the fourth embodiment.
- FIG. 9 is a diagram illustrating a setting example of the parameter a in a space in a wide area.
- FIG. 10 is a diagram for explaining an example of setting the parameter limit in a space in a wide area.
- FIG. 11 is a block diagram of a noise removing apparatus according to the fourth embodiment.
- FIG. 1 is a diagram for explaining the processing of the first embodiment of the present invention.
- pixel statistics of pixels in the area are obtained for each of the multi-hierarchy areas including the target pixel and the range is sequentially narrowed.
- the corrected pixel statistic of the area in the previous hierarchy is corrected using the corrected pixel statistic of the area in the previous hierarchy with a wide range, and the pixel of interest is corrected using the corrected pixel statistic in the area of the minimum range. It is characterized by doing.
- the pixel statistical value of the pixel is, for example, a spatial average value of the area of each layer, and the spatial average value is an arithmetic average value, a geometric average value, a weighted average value, or the like of the pixels existing in the area. .
- the spatial average value is an arithmetic average value of pixels existing in a region.
- FIG. 1 shows the flow of processing when multi-resolution processing of three layers (A1 to A3) is performed, but it can be easily extended to other than three layers.
- the middle area The spatial average value (S2 (x, y)) in A2 (range: -k2 to k2) is corrected. Then, the spatial average value (S1 (x, y)) of the narrow area A1 (range: ⁇ k1 to k1) is corrected based on the corrected spatial average value (S2 ′ (x, y)) in the middle area A2. . By performing this correction in order, the target pixel value P in (x, y) is corrected to obtain an output pixel value P out (x, y).
- the spatial average value S3 (x, y) of the wide region (range: -k3 to k3) and the spatial average value S2 (x, y) of the middle region (range: -k2 to k2) at the pixel position (x, y) ) Is calculated as in equations (1) and (2).
- Func () represents a correction function for suppressing noise components, and corresponds to the correction function F in FIG.
- Func () is a function that has a characteristic of generating an output value that approaches the input value when the input value of the function approaches zero and the output value approaches zero and the absolute value of the input value increases. It ’s fine.
- An example of the Func function having such characteristics is shown in FIG.
- the parameters a, b, and limit of the correction function in FIG. 2 control how much the noise component is suppressed.
- the parameters are determined according to the noise characteristics of the image sensor. Each value may be determined based on an evaluation experiment in which the image quality of the actually captured image is subjectively evaluated.
- FIG. 3 is a block diagram of the noise removal apparatus according to the first embodiment.
- the noise removal apparatus includes an area pixel value extraction unit 1, a spatial average value calculation unit 2, a correction unit 3, and an output image control unit 4.
- the region pixel value extraction unit 1 is controlled by the output image control unit 4 and pixel values of pixels in the wide region A3 (range: ⁇ k3 to k3) centered on the pixel position (x, y) (target pixel).
- the pixel value of the pixel in the middle area A2 (range: -k2 to k2), the pixel value of the pixel in the narrow area A1 (range: -k1 to k1), and the input pixel value P in (x, y) Are extracted at each timing and output to the spatial average value calculation unit 2.
- the spatial average value calculation unit 2 receives the pixel value of each region from the region pixel value extraction unit 1, and calculates the spatial average value of the region. The calculated spatial average value is output to the correction unit 3.
- the correction unit 3 inputs the corrected spatial average value of the area of the previous hierarchy from the output image control unit 4 and the spatial average value of the area of the hierarchy from the spatial average value calculation unit 2, and the space of the hierarchy Correct the average value.
- the correction method is corrected using the correction function described above.
- the output image control unit 4 instructs the region pixel value extraction unit 1 to extract the pixel value of the pixel in the region of the next layer each time the sequentially corrected spatial average value is input. Each time a corrected spatial average value is input, the value is fed back to the correction unit 3. Then, it outputs one pixel P out (x, y) when is input, the P out (x, y) as an output pixel value.
- the first embodiment by correcting pixel statistic values in a wide range region to pixel statistic values in a narrow range region in order, not only high frequency noise but also low frequency noise is effectively obtained. Noise removal is possible.
- FIG. 4 is a diagram for explaining the processing of the second embodiment of the present invention.
- the Func function for suppressing the noise component in the first embodiment is changed for each layer. That is, as shown in FIG. 4, the correction function F of each layer is different for each layer.
- the parameter a of the Func function (correction function) in FIG. 2 is changed according to the amount of change in the pixel value indicated by the noise. More specifically, the function F2 is determined using the parameter a as a1 in accordance with the variation amount of the pixel value indicated by the low frequency noise appearing in the space in the wide area (for example, the range A3 in FIG. 4). Further, the function F1 is determined by setting the parameter a to a2 in accordance with the variation amount of the pixel value indicated by the medium frequency noise appearing in the space in the middle region (for example, the range A2 in FIG. 4).
- the relationship of a2> a1 is often maintained, but this is not the case.
- the function F0 is determined by setting the parameter a to a3 according to the amount of change in the pixel value indicated by the high frequency noise appearing in the space in the narrow area (for example, the range A1 in FIG. 4).
- the relationship of a3> a2> a1 is often maintained, but this is not the case.
- FIG. 5 is a block diagram of the noise removing apparatus according to the second embodiment.
- the noise removal apparatus includes an area pixel value extraction unit 1, a spatial average value calculation unit 2, a correction unit 3, an output image control unit 4, and a correction function determination unit 5.
- the noise removing apparatus is different from the noise removing apparatus according to the first embodiment in that the correction function determining unit 5 is provided.
- the correction function determination unit 5 receives the pixel values of the pixels in each region from the region pixel value extraction unit 1, and, as described above, according to the variation amount of the pixel value indicated by the noise that appears in the space of the region, the Func function The parameter a of (correction function) is determined.
- the correction unit 3 corrects the spatial average value of each layer by the Func function (correction function) determined by the correction function determination unit 5.
- FIG. 6 is a diagram for explaining the processing of the third embodiment of the present invention.
- the pixel statistical value of a pixel is, for example, a spatial average value of a region of each layer, and the spatial average value is an arithmetic average value, a geometric average value, a weighted average value, or the like of pixels existing in the region.
- a pixel statistical value is a spatial average value and the spatial average value is an arithmetic average value of pixels existing in a region will be described.
- a spatial average value S3 (x, y) that is a pixel statistic value of a space in a wide region and edge information in the region or an edge amount E3 (x, y) are used in the middle region.
- the spatial average value (S2 (x, y)) is corrected.
- the spatial average value (S1 (x, y) in the narrow region )) Is corrected.
- the input pixel value P in (x, y) is corrected to obtain the output pixel value P out (x, y).
- the edge information or the edge amount is defined by a difference value of a statistical amount (average value, median, etc.) of pixels between the upper, lower, left and right regions around the target pixel (input pixel).
- the process in each layer is the same in the process flow except that the parameters for determining the correction amount are different. Therefore, as an example, details of the process of correcting the spatial average value S2 (x, y) in the middle region using the spatial average value S3 (x, y) in the wide region will be described.
- the spatial average value S2 (x, y) is calculated as shown in equations (1) and (2).
- the edge amount E3 (x, y) in the wide area is calculated.
- the vertical edge amount EV3 (x, y) and the horizontal edge amount EH3 (x, y) are calculated as shown in Equation (4) and Equation (5), and these are calculated.
- the spatial average value S3 (x, y) in the wide area is calculated as shown in Equation (7). Correction is performed to calculate a corrected spatial average value S3 ′′ (x, y) of the wide area.
- the composite weight ⁇ 3 (x, y) is calculated using Equation (8) using threshold values hi and lo set in advance. Note that the thresholds hi and lo are thresholds determined for each hierarchy and are set to optimum values for each hierarchy, but the same values may be used.
- the correction function Func shown in FIG. 2 is used.
- the correction of the spatial average value S2 (x, y) of the middle region at the pixel position (x, y) is performed by setting the difference to (S2 (x, y) ⁇ S3 ′′ (x, y)) and correcting the image in FIG.
- the correction amount diffout obtained by the function is added to S2 (x, y), and parameters a, b, and limit in the correction function in Fig. 2 indicate the resolution to be processed and the color components to be corrected. To be determined.
- the third embodiment differs from the first and second embodiments in that the spatial average value of a wide area is corrected by the correction function of Expression (7) based on edge information. It is in.
- the substantial correction amount is controlled based on the edge information, and edge rounding can be suppressed by using this method.
- FIG. 7 is a block diagram of the noise removal apparatus of the third embodiment.
- the noise removal apparatus includes an area pixel value extraction unit 1, a spatial average value calculation unit 2, a correction unit 3, an output image control unit 4, and an edge information calculation unit 6.
- the region pixel value extraction unit 1 is controlled by the output image control unit 4 and pixel values of pixels in the wide region A3 (range: ⁇ k3 to k3) centered on the pixel position (x, y) (target pixel).
- the pixel value of the pixel in the middle area A2 range: -k2 to k2
- the pixel value of the pixel in the narrow area A1 range: -k1 to k1
- the pixel value of (target pixel) is extracted at each timing and output to the spatial average value calculation unit 2.
- the spatial average value calculation unit 2 receives the pixel value of each region from the region pixel value extraction unit 1, and calculates the spatial average value of the region. The calculated spatial average value is output to the correction unit 3.
- the edge information calculation unit 6 calculates the edge amount E3 (x, y) in the wide region A3 based on the pixel values of the pixels existing in the wide region from the region pixel value extraction unit 1. For the calculation of the edge amount, the vertical edge amount EV3 (x, y) and the horizontal edge amount EH3 (x, y) are calculated as shown in Equation (4) and Equation (5), and these are calculated using Equation (4) and Equation (5). By adding as in (6), the edge amount E3 (x, y) in the wide area A3 is calculated. Similarly, the edge amount E2 (x, y) of the middle region A2 and the edge amount E1 (x, y) of the narrow region A1 are calculated.
- the correction unit 3 uses the combined weight ⁇ 3 (x, y) obtained from the edge amount E3 (x, y) calculated by the edge information calculation unit 6 to obtain a spatial average value in a wide area as shown in Equation (7).
- S3 (x, y) is corrected, and the corrected wide area spatial average value S3 ′′ (x, y) is calculated.
- the synthesis weight ⁇ 3 (x, y) is a preset threshold value hi and Using lo, it is calculated as in equation (8).
- the spatial average value S2 (x, y) of the middle region is corrected as in Expression (9).
- a similar correction is performed using the spatial average value S1 (x, y).
- the input pixel value P in (x, y) is also performed.
- the output image control unit 4 instructs the region pixel value extraction unit 1 to extract the pixel value of the pixel in the region of the next layer each time the sequentially corrected spatial average value is input. Each time a corrected spatial average value is input, the value is fed back to the correction unit 3. Then, it outputs one pixel P out (x, y) when is input, the P out (x, y) as an output pixel value.
- the third embodiment can further suppress edge rounding.
- FIG. 8 is a diagram for explaining the processing of the fourth embodiment of the present invention.
- the edge amount E3 (x, y) calculated by Expression (6) in each layer is reflected in the Func function (correction function) that suppresses the noise component.
- the noise component of each layer is adaptively suppressed by changing the Func function (correction function) of each layer.
- the result of applying the Func function to the difference between the spatial average value S3 (x, y) in the wide region and the spatial average value S2 (x, y) in the middle region is expressed as the spatial average value S3 (
- the sum of x, y) is output as a correction value S2 ′ (x, y) of the spatial average value S2 (x, y) in the middle region.
- the parameter a is determined using the edge amount E3 (x, y) in the wide area A2 calculated by Expression (6).
- a coefficient ⁇ 3 (x, y) whose value changes as shown in Expression (11) according to the edge amount E3 (x, y) is defined.
- Hi and lo which are threshold values of E3 (x, y), are preset values.
- the thresholds hi and lo are thresholds determined for each hierarchy, and are set to optimum values for each hierarchy, but the same values may be used.
- the coefficient ⁇ 3 (x, y) defined by Equation (11) is a real number from 0 to 1.0.
- the parameter a in the Func function is set using the coefficient ⁇ 3 (x, y).
- Figure 9 shows an example of setting parameter a in a wide area.
- parameter a is expressed by the following equation.
- a_lo is a value used as the parameter a when the edge amount E3 (x, y) is smaller than the threshold value lo
- a_hi is a value used for the parameter a when the edge amount is larger than the threshold value hi.
- the parameter a is a value between a_lo and a_hi when the edge amount E3 (x, y) is the threshold value lo to hi.
- the parameter a in the Func function is similarly set for the middle region A2 and the narrow region A1.
- a method of reflecting the edge amount in the parameter limit in the Func function is also possible.
- the parameter limit in the Func function is set using the coefficient ⁇ 3 (x, y) of the equation (11).
- FIG. 10 shows an example of setting the parameter limit in the space in the wide area.
- the parameter limit can be expressed by the following equation.
- lim_lo is a value used for the parameter limit when the edge amount E3 (x, y) is smaller than the threshold value lo
- lim_hi is a value used for the parameter limit when the edge amount is larger than the threshold value hi.
- the parameter limit when the edge amount E3 (x, y) is the threshold value lo to hi is a value between lim_lo and lim_hi.
- lim_lo is a real number greater than or equal to 0
- the parameter limit in the Func function is also set for the middle and narrow areas.
- FIG. 11 is a block diagram of a noise removing apparatus according to the fourth embodiment.
- the noise removal apparatus includes an area pixel value extraction unit 1, a spatial average value calculation unit 2, a correction unit 3, an output image control unit 4, an edge information calculation unit 6, and a correction function determination. Part 7.
- the region pixel value extraction unit 1 receives the control of the output image control unit 4 and the pixel values of the pixels in the wide region A3 (range: ⁇ k3 to k3) centered on the pixel position (x, y) and the middle region
- the pixel value of the pixel in A2 (range: -k2 to k2)
- the pixel value of the pixel in the narrow area A1 (range: -k1 to k1)
- the pixel value of the input pixel value P in (x, y) Are extracted at each timing and output to the spatial average value calculation unit 2.
- the spatial average value calculation unit 2 receives the pixel values of each range from the region pixel value extraction unit 1, and calculates the spatial average value of the range. The calculated spatial average value is output to the correction unit 3.
- the edge information calculation unit 6 first calculates an edge amount E3 (x, y) in the wide region A3 based on the pixel values of the pixels existing in the wide region from the region pixel value extraction unit 1. For the calculation of the edge amount, the vertical edge amount EV3 (x, y) and the horizontal edge amount EH3 (x, y) are calculated as shown in Equation (4) and Equation (5), and these are calculated using Equation (4) and Equation (5). By adding as in (6), the edge amount E3 (x, y) in the wide area A3 is calculated. Similarly, the edge amount E2 (x, y) of the middle region A2 and the edge amount E1 (x, y) of the narrow region A1 are calculated.
- the correction function determination unit 7 obtains the parameter a in the Func function (correction function) based on the edge amount as described above, and determines the Func function (correction function). Note that the func function (correction function) may be determined so that the edge amount is reflected in the parameter limit in the func function.
- the correction unit 3 corrects the spatial average value of each region by the Func function (correction function) determined by the correction function determination unit 7.
- the correction method is corrected using the correction function described above.
- the output image control unit 4 instructs the region pixel value extraction unit 1 to extract the pixel value of the pixel in the region of the next layer each time the sequentially corrected spatial average value is input. Each time a corrected spatial average value is input, the value is fed back to the correction unit 3. Then, it outputs one pixel P out (x, y) when is input, the P out (x, y) as an output pixel value.
- the fourth embodiment can further suppress edge rounding.
- each unit can be configured by hardware, but can also be realized by a computer program.
- functions and operations similar to those of the above-described embodiments are realized by a processor that operates according to a program stored in the program memory.
- Additional remark 3 The noise removal method of Additional remark 2 which weights to the pixel statistic value after correction
- Additional remark 4 The noise removal method of Additional remark 3 which does not correct
- Additional remark 6 The noise removal method of Additional remark 5 which changes the parameter of the correction function which corrects a pixel statistical value according to the variation
- Additional remark 7 The noise removal method of Additional remark 5 which changes the parameter of the correction function which corrects a pixel statistic value for every hierarchy based on edge information.
- the function that corrects the pixel statistic value is such that the output value approaches zero as the input value of the function approaches zero, and the output value that approaches the input value increases as the absolute value of the input value increases.
- the noise removal method according to any one of appendix 1 to appendix 7, which is a function having a generated characteristic.
- the pixel statistical value calculation means which calculates the pixel statistical value of the pixel of the area
- a noise removing apparatus comprising: a correcting unit that corrects the target pixel using a statistical value.
- the correction unit corrects the pixel statistical value of the area of the hierarchy using the pixel statistical value of the area of the hierarchy and the corrected pixel statistics of the previous hierarchy and the edge information of the previous hierarchy.
- the noise removal apparatus as described.
- amendment means is a noise removal apparatus of Additional remark 12 which weights the pixel statistic value after the correction
- amendment means is a noise removal apparatus of Additional remark 13 which does not correct
- amendment means is a noise removal apparatus in any one of Additional remark 11 to Additional remark 14 which changes the parameter of the correction function which corrects a pixel statistical value for every hierarchy.
- amendment means is a noise removal apparatus of Additional remark 15 which changes the parameter of the correction function which correct
- amendment means is a noise removal apparatus of Additional remark 15 which changes the parameter of the correction function which corrects a pixel statistical value for every hierarchy based on edge information.
- the correction means has a function of generating an output value that approaches an input value when the input value approaches zero and the output value approaches zero and the absolute value of the input value increases.
- the noise removal device according to any one of supplementary note 11 to supplementary note 17, wherein the pixel statistical value is corrected.
- the noise removal device according to supplementary note 19, wherein the spatial average value is any one of an arithmetic average value, a geometric average value, and a weighted average value of pixels in a region of each layer.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
Description
本発明の第1の実施の形態を説明する。
本発明の第2の実施の形態を説明する。
本発明の第3の実施の形態を説明する。
第4の実施の形態を説明する。
ここで,Func関数は,式(6)で算出される広領域A2におけるエッジ量E3(x,y)を利用してパラメータaが決定される。
ここで、a_loは、エッジ量E3(x,y)が閾値loより小さい場合にパラメータaとして使用する値であり、a_hiはエッジ量が閾値hiより大きくなる場合のパラメータaに使用する値である。エッジ量E3(x,y)が閾値loからhiまでパラメータaは、a_loからa_hiまでの間の値となる。ここで、a_loは0以上の実数であり、a_hiは,a_hi>=a_loの実数である。
ここで、lim_loは、エッジ量E3(x,y)が閾値loより小さい場合にパラメータlimitに使用する値であり、lim_hiはエッジ量が閾値hiより大きくなる場合のパラメータlimitに使用する値である。エッジ量E3(x,y)が閾値loからhiまでのパラメータlimitは,lim_loからlim_hiまでの間の値となる。尚、lim_loは0以上の実数であり、lim_hiは,lim_hi>=lim_loの実数である。
順次各階層において、当該階層の領域よりも範囲の広い前階層の領域の補正後の画素統計値を用いて、当該階層の領域の画素統計値を補正し、
最小範囲の領域の補正後の画素統計値を用いて、前記注目画素を補正する
ノイズ除去方法。
当該階層の領域の画素統計値を、前階層の補正後の画素統計値と、前階層のエッジ情報とを用いて、当該階層の領域の画素統計値を補正する
付記1に記載のノイズ除去方法。
付記2に記載のノイズ除去方法。
付記3に記載のノイズ除去方法。
付記1から付記4のいずれかに記載のノイズ除去方法。
付記5に記載のノイズ除去方法。
付記5に記載のノイズ除去方法。
付記1から付記7のいずれかに記載のノイズ除去方法。
付記1から付記8のいずれかに記載のノイズ除去方法。
付記9に記載のノイズ除去方法。
順次各階層において、当該階層の領域よりも範囲の広い前階層の領域の補正後の画素統計値を用いて、当該階層の領域の画素統計値を補正し、最小範囲の領域の補正後の画素統計値を用いて、前記注目画素を補正する補正手段と
を有するノイズ除去装置。
前記補正手段は、当該階層の領域の画素統計値を、前階層の補正後の画素統計値と、前階層のエッジ情報とを用いて、当該階層の領域の画素統計値を補正する
付記11に記載のノイズ除去装置。
付記12に記載のノイズ除去装置。
付記13に記載のノイズ除去装置。
付記11から付記14のいずれかに記載のノイズ除去装置。
付記15に記載のノイズ除去装置。
付記15に記載のノイズ除去装置。
付記11から付記17のいずれかに記載のノイズ除去装置。
付記11から付記18のいずれかに記載のノイズ除去装置。
付記19に記載のノイズ除去装置。
注目画素を含み、範囲が順次狭くなる多階層の領域毎に、その領域の画素の画素統計値を算出する処理と、
順次各階層において、当該階層の領域よりも範囲の広い前階層の領域の補正後の画素統計値を用いて、当該階層の領域の画素統計値を補正する処理と、
最小範囲の領域の補正後の画素統計値を用いて、前記注目画素を補正する処理と
を実行させるプログラム。
2 空間平均値算出部
3 補正部
4 出力画像制御部
5 補正関数決定部
6 エッジ情報算出部
7 補正関数決定部
Claims (21)
- 注目画素を含み、範囲が順次狭くなる多階層の領域毎に、その領域の画素の画素統計値を求め、
順次各階層において、当該階層の領域よりも範囲の広い前階層の領域の補正後の画素統計値を用いて、当該階層の領域の画素統計値を補正し、
最小範囲の領域の補正後の画素統計値を用いて、前記注目画素を補正する
ノイズ除去方法。 - 各階層の領域のエッジ情報を求め、
当該階層の領域の画素統計値を、前階層の補正後の画素統計値と、前階層のエッジ情報とを用いて、当該階層の領域の画素統計値を補正する
請求項1に記載のノイズ除去方法。 - 前階層のエッジ情報に応じて、前階層の補正後の画素統計値に重みづけ行う
請求項2に記載のノイズ除去方法。 - 前階層の領域のエッジ情報が所定の閾値を超える場合、当該階層の画素統計値の補正を行わない
請求項3に記載のノイズ除去方法。 - 画素統計値を補正する補正関数のパラメータを、階層毎に変化させる
請求項1から請求項4のいずれかに記載のノイズ除去方法。 - 画素統計値を補正する補正関数のパラメータを、当該階層の領域のノイズによる画素値の変動量に応じて変化させる
請求項5に記載のノイズ除去方法。 - 画素統計値を補正する補正関数のパラメータを、エッジ情報に基づいて階層毎に変化させる
請求項5に記載のノイズ除去方法。 - 画素統計値を補正する関数は、関数の入力値がゼロに近くづくほど、出力値はゼロに近づき、入力値の絶対値が大きくなれば、その入力値に近くなる出力値を発生する特性を有する関数である
請求項1から請求項7のいずれかに記載のノイズ除去方法。 - 前記画素統計値は、各階層の領域における画素の空間平均値である
請求項1から請求項8のいずれかに記載のノイズ除去方法。 - 前記空間平均値は、各階層の領域における画素の相加平均値、相乗平均値、加重平均値のいずれかである
請求項9に記載のノイズ除去方法。 - 注目画素を含み、範囲が順次狭くなる多階層の領域毎に、その領域の画素の画素統計値を算出する画素統計値算出手段と、
順次各階層において、当該階層の領域よりも範囲の広い前階層の領域の補正後の画素統計値を用いて、当該階層の領域の画素統計値を補正し、最小範囲の領域の補正後の画素統計値を用いて、前記注目画素を補正する補正手段と
を有するノイズ除去装置。 - 各階層の領域のエッジ情報を算出するエッジ情報算出手段を有し、
前記補正手段は、当該階層の領域の画素統計値を、前階層の補正後の画素統計値と、前階層のエッジ情報とを用いて、当該階層の領域の画素統計値を補正する
請求項11に記載のノイズ除去装置。 - 前記補正手段は、前階層のエッジ情報に応じて、前階層の補正後の画素統計値に重みづけ行う
請求項12に記載のノイズ除去装置。 - 前記補正手段は、前階層の領域のエッジ情報が所定の閾値を超える場合、当該階層の画素統計値の補正を行わない
請求項13に記載のノイズ除去装置。 - 前記補正手段は、画素統計値を補正する補正関数のパラメータを、階層毎に変化させる
請求項11から請求項14のいずれかに記載のノイズ除去装置。 - 前記補正手段は、画素統計値を補正する補正関数のパラメータを、当該階層の領域のノイズによる画素値の変動量に応じて変化させる
請求項15に記載のノイズ除去装置。 - 前記補正手段は、画素統計値を補正する補正関数のパラメータを、エッジ情報に基づいて階層毎に変化させる
請求項15に記載のノイズ除去装置。 - 前記補正手段は、入力値がゼロに近くづくほど、出力値はゼロに近づき、入力値の絶対値が大きくなれば、その入力値に近くなる出力値を発生する特性を有する関数により、前記画素統計値を補正する
請求項11から請求項17のいずれかに記載のノイズ除去装置。 - 前記画素統計値算出手段は、各階層の領域における画素の空間平均値を前記画素統計値として算出する
請求項11から請求項18のいずれかに記載のノイズ除去装置。 - 前記空間平均値は、各階層の領域における画素の相加平均値、相乗平均値、加重平均値のいずれかである
請求項19に記載のノイズ除去装置。 - コンピュータに、
注目画素を含み、範囲が順次狭くなる多階層の領域毎に、その領域の画素の画素統計値を算出する処理と、
順次各階層において、当該階層の領域よりも範囲の広い前階層の領域の補正後の画素統計値を用いて、当該階層の領域の画素統計値を補正する処理と、
最小範囲の領域の補正後の画素統計値を用いて、前記注目画素を補正する処理と
を実行させるプログラム。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/239,577 US9299131B2 (en) | 2011-08-22 | 2012-08-21 | Noise reduction apparatus, noise reduction method, and computer-readable storage medium |
JP2013530023A JP6146574B2 (ja) | 2011-08-22 | 2012-08-21 | ノイズ除去装置、ノイズ除去方法及びプログラム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011-180637 | 2011-08-22 | ||
JP2011180637 | 2011-08-22 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013027723A1 true WO2013027723A1 (ja) | 2013-02-28 |
Family
ID=47746462
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2012/071072 WO2013027723A1 (ja) | 2011-08-22 | 2012-08-21 | ノイズ除去装置、ノイズ除去方法及びプログラム |
Country Status (3)
Country | Link |
---|---|
US (1) | US9299131B2 (ja) |
JP (1) | JP6146574B2 (ja) |
WO (1) | WO2013027723A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2014164734A (ja) * | 2013-02-28 | 2014-09-08 | Nec Corp | 画像処理方法、及び画像処理装置 |
US20170287116A1 (en) * | 2014-09-29 | 2017-10-05 | Nec Corporation | Image processing method, image processing device, and recording medium for storing image processing program |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013027723A1 (ja) * | 2011-08-22 | 2013-02-28 | 日本電気株式会社 | ノイズ除去装置、ノイズ除去方法及びプログラム |
JP6249173B2 (ja) * | 2012-04-26 | 2017-12-20 | 日本電気株式会社 | 画像処理方法、及び画像処理装置 |
EP2843617B1 (en) * | 2012-04-26 | 2021-01-06 | Nec Corporation | Image processing method and image processing device |
JP6213558B2 (ja) * | 2013-02-28 | 2017-10-18 | 日本電気株式会社 | 画像処理方法、及び画像処理装置 |
CN114697621A (zh) * | 2020-12-14 | 2022-07-01 | 安霸国际有限合伙企业 | 使用反指数函数优化的边缘保留噪声降低算法 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0258181A (ja) * | 1988-08-24 | 1990-02-27 | Sakata Corp | デジタル画像信号の処理方法及びそれを実施するための装置 |
JP2002010108A (ja) * | 2000-06-21 | 2002-01-11 | Fuji Photo Film Co Ltd | 画像信号処理装置および画像信号処理方法 |
Family Cites Families (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2772161B1 (fr) * | 1997-12-09 | 2000-02-25 | Trixell Sas | Procede de correction du niveau de bruit d'un detecteur d'image numerisee |
JP3664231B2 (ja) | 2000-08-09 | 2005-06-22 | 日本電気株式会社 | カラー画像処理装置 |
CN100524346C (zh) * | 2004-06-14 | 2009-08-05 | 普瑞科德公司 | 过滤数字图像的方法和系统、数字图像处理器 |
CA2616875A1 (en) * | 2004-07-30 | 2006-02-02 | Algolith Inc. | Apparatus and method for adaptive 3d artifact reducing for encoded image signal |
US7602438B2 (en) * | 2004-10-19 | 2009-10-13 | Eastman Kodak Company | Method and apparatus for capturing high quality long exposure images with a digital camera |
US7508430B1 (en) * | 2005-02-18 | 2009-03-24 | Magnachip Semiconductor, Ltd. | Method for locally reducing row noise |
EP2082566B1 (en) * | 2006-10-18 | 2016-08-31 | Robert Bosch GmbH | Image processing system, method and computer program for contrast enhancement of images |
JP5253835B2 (ja) * | 2008-02-19 | 2013-07-31 | 株式会社キーエンス | 画像生成装置、画像生成方法及びコンピュータプログラム |
JP5315158B2 (ja) * | 2008-09-12 | 2013-10-16 | キヤノン株式会社 | 画像処理装置及び画像処理方法 |
WO2013027723A1 (ja) * | 2011-08-22 | 2013-02-28 | 日本電気株式会社 | ノイズ除去装置、ノイズ除去方法及びプログラム |
US8953882B2 (en) * | 2012-05-31 | 2015-02-10 | Apple Inc. | Systems and methods for determining noise statistics of image data |
US8817120B2 (en) * | 2012-05-31 | 2014-08-26 | Apple Inc. | Systems and methods for collecting fixed pattern noise statistics of image data |
US20140056536A1 (en) * | 2012-08-27 | 2014-02-27 | Toshiba Medical Systems Corporation | Method and system for substantially removing dot noise |
-
2012
- 2012-08-21 WO PCT/JP2012/071072 patent/WO2013027723A1/ja active Application Filing
- 2012-08-21 US US14/239,577 patent/US9299131B2/en active Active
- 2012-08-21 JP JP2013530023A patent/JP6146574B2/ja active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0258181A (ja) * | 1988-08-24 | 1990-02-27 | Sakata Corp | デジタル画像信号の処理方法及びそれを実施するための装置 |
JP2002010108A (ja) * | 2000-06-21 | 2002-01-11 | Fuji Photo Film Co Ltd | 画像信号処理装置および画像信号処理方法 |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2014164734A (ja) * | 2013-02-28 | 2014-09-08 | Nec Corp | 画像処理方法、及び画像処理装置 |
US20170287116A1 (en) * | 2014-09-29 | 2017-10-05 | Nec Corporation | Image processing method, image processing device, and recording medium for storing image processing program |
Also Published As
Publication number | Publication date |
---|---|
US9299131B2 (en) | 2016-03-29 |
JPWO2013027723A1 (ja) | 2015-03-19 |
US20140205202A1 (en) | 2014-07-24 |
JP6146574B2 (ja) | 2017-06-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6146574B2 (ja) | ノイズ除去装置、ノイズ除去方法及びプログラム | |
US8818126B2 (en) | Method and apparatus for block-based image denoising | |
JP6213558B2 (ja) | 画像処理方法、及び画像処理装置 | |
JP6249173B2 (ja) | 画像処理方法、及び画像処理装置 | |
JP2012208553A (ja) | 画像処理装置、および画像処理方法、並びにプログラム | |
JP6256703B2 (ja) | 画像処理方法、及び画像処理装置 | |
JP6256704B2 (ja) | 画像処理方法、及び画像処理装置 | |
Gupta et al. | Image noise removal with detail preservation for random valued impulse noise | |
JP4635652B2 (ja) | 画像処理装置、画像処理方法、画像処理方法のプログラム及び画像処理方法のプログラムを記録した記録媒体 | |
JP2005149266A (ja) | 画像処理装置、画像処理方法及び画像処理プログラム | |
Okado et al. | Fast and high-quality regional histogram equalization | |
US20100086229A1 (en) | Image signal processing apparatus and method | |
JP6079962B2 (ja) | 画像処理方法、及び画像処理装置 | |
JP5559275B2 (ja) | 画像処理装置及びその制御方法 | |
JP5350497B2 (ja) | 動き検出装置、制御プログラム、および集積回路 | |
TWI386868B (zh) | 使用內容調適懲罰函數的移動偵測方法 | |
TWI389571B (zh) | 影像處理方法以及影像處理裝置 | |
JP6128312B2 (ja) | 画像処理方法、及び画像処理装置 | |
JP5007851B2 (ja) | 画像処理装置及び画像処理プログラム | |
JP4992438B2 (ja) | 画像処理装置および画像処理プログラム | |
Al_anizy | A METHOD OF IMPULSE NOISE DETECTION BY USING LAPLACIAN-GAUSSIAN FILTER ADAPTED WITH GENETIC ALGORITHM |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 12826017 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2013530023 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14239577 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 12826017 Country of ref document: EP Kind code of ref document: A1 |