JP4040512B2 - Image processing method and image processing apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a technique for reducing noise in an input multilevel image.
[0002]
[Prior art]
Conventionally, image data captured by a digital camera or image data optically read by a scanner or the like includes various noises derived from an optical sensor such as a CCD, such as high-frequency noise and relatively large patchy noise. Low frequency noise is included. In order to reduce high-frequency noise among these noises, a low-pass filter is generally used in many cases.
[0003]
Moreover, as a high frequency noise reduction process, there is also an example using a median filter process (for example, refer to Patent Document 1).
[0004]
[Patent Document 1]
JP-A-4-235472
[Problems to be solved by the invention]
However, if the various filter processes as described above are performed on all the image data, not only noise but also high-frequency components of the image are reduced, so that the image quality is deteriorated. The various filter processes as described above are mainly aimed at reducing high-frequency noise, and have no effect on the reduction of low-frequency noise such as the relatively large spotted noise.
[0005]
The present invention has been made to solve the above-described problems, and an object thereof is to reduce low-frequency noise while suppressing adverse effects such as a decrease in resolution.
[0006]
[Means for Solving the Problems]
In order to achieve the above object, an image processing method according to the present invention includes a step of reducing an input image at a predetermined magnification to obtain a reduced image, and a window in a window of a predetermined size corresponding to the position of an input pixel from the reduced image. Extracting pixels; and An average value of the extracted pixels is obtained, a determination is made as to which of the two regions the input pixel is, a representative value is obtained for each of the determined regions, and the average value, the two representative values, and the input pixel Each difference value is obtained, and the average value and the two representative values are selected as replacement data, with the smallest difference value. A step of obtaining a difference value between the replacement data and the value of the input pixel, a step of comparing the difference value with a first threshold value, and a case where the difference value is less than the first threshold value. And a step of replacing the value of the input pixel with the replacement data.
The image processing method according to the present invention also provides an input image with a predetermined magnification. Reduce with do it Shrink Obtaining an image; and Shrink Extracting a pixel in a window of a predetermined size corresponding to the position of the input pixel from the image; An average value of the extracted pixels is obtained, a determination is made as to which of the two regions each pixel is, a representative value is obtained for each determined region, and a difference between the obtained two region representative values is obtained. The difference between the two region representative values is compared with a second threshold value, and when the difference between the two region representative values is less than or equal to the second threshold value, the average value is used as replacement data, and the average of the two region representative values is Determining, comparing the average of the two area representative values with the value of the input pixel, and selecting a smaller area representative value when the value of the input pixel is less than the average of the two area representative values; When the value of the input pixel is equal to or greater than the average of the two area representative values, the larger area representative value is selected, and when the difference between the two area representative values exceeds the second threshold, the selected representative Using the value as replacement data, and A step of obtaining a difference value between replacement data and the value of the input pixel, a step of comparing the difference value with a first threshold value, and a value of the input pixel when the difference value is less than the first threshold value. And a step of replacing with the replacement data.
[0007]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments according to the present invention will be described in detail with reference to the drawings. The embodiments described below provide disclosure to facilitate implementation by those skilled in the art of the present invention, and are only included in the technical scope of the present invention defined by the claims. Only some embodiments.
[0008]
Therefore, it will be apparent to those skilled in the art that even embodiments that are not directly described in the specification are included in the technical scope of the present invention as long as they share the same technical idea.
[0009]
Although a plurality of embodiments are described for convenience, those skilled in the art can easily understand that these are not only individually established as inventions, but of course, the invention can also be realized by appropriately combining a plurality of embodiments. I can do it.
[0010]
[First Embodiment]
FIG. 1 is a block diagram illustrating a configuration of an image processing apparatus according to the first embodiment. In FIG. 1, reference numeral 101 denotes an image reduction unit that reduces an input image at a predetermined magnification (for example, ¼). Reference numeral 102 denotes a pixel extraction unit, which extracts pixels from a reduced image into an n × m (m, n is an integer) rectangular shape (window) centered on the input pixel position. Reference numeral 103 denotes an in-window average value calculation unit that generates an average value of pixel values in the extracted window. Reference numeral 104 denotes an area determination unit. For example, the reduced pixel value in the window is compared with the average value in the window, and a pixel whose pixel value is equal to or less than the average value in the window is determined as the area 0, and other pixels are determined as the area 1. To do. Reference numeral 105 denotes an area 0 average value calculation unit that generates an in-window average value of pixels determined to be area 0 by the area determination unit 104. Reference numeral 106 denotes an area 1 average value calculation unit that generates an average value in a window of pixels determined to be area 1 by the area determination unit 104.
[0011]
Reference numeral 110 denotes a timing adjustment unit, which delays an input image (target pixel) by an amount corresponding to the latency in each processing unit. An area 0 difference value generation unit 107 generates an absolute difference between the output of the area 0 average value calculation unit 105 and the input image (target pixel) that has been delay compensated by the timing adjustment unit 110. An area 1 difference value generation unit 108 generates an absolute difference value between the output of the area 1 average value calculation unit 106 and the input image (target pixel) that has been delay compensated by the timing adjustment unit 110. Reference numeral 109 denotes an in-window difference value generation unit that generates an absolute difference value between the output of the in-window average value calculation unit 103 and the input image (target pixel) compensated for delay by the timing adjustment unit 110. Reference numeral 111 denotes a comparison unit that controls the pixel replacement unit 112 so as to select an average value that is the minimum difference value less than the threshold Th1 among the absolute difference values of the three difference value generation units 107, 108, and 109. In addition, when any difference absolute value does not become less than threshold value Th1, the pixel replacement part 112 is controlled so that an input image (target pixel) is selected. A pixel replacement unit 112 replaces the input pixel value with an average value satisfying the above condition only when a minimum difference absolute value less than the threshold Th1 exists.
[0012]
Next, a description will be given of a process in which an input image is blocked and a pixel of interest in a block of interest is replaced based on a predetermined condition in the image processing apparatus having the above-described configuration.
[0013]
FIG. 2 is a flowchart illustrating processing of the image processing apparatus according to the first embodiment. First, in step S201, a CPU (not shown) sets parameters necessary for each process. Here, the parameters to be set are the image reduction ratio, block size, window size, and threshold value, and are determined according to the nature of the input image and noise to be removed. The parameter may be determined by analyzing the input image, or may be set to a value determined in advance according to the attribute (property) of the input image and the output condition. If the parameter is always a constant value, parameter setting is omitted.
[0014]
Next, in step S202, the image reduction unit 101 reduces the input image at a predetermined reduction rate. By performing this reduction processing, the number of pixels (window size) for low-frequency noise removal described later is reduced. Note that if the reduction ratio is reduced (the output size is reduced), the subsequent window size can be reduced. However, since the degree of influence in the vicinity of the input pixel is reduced, the difference from the input pixel is increased and the pixel to be replaced is replaced. Not only decreases, but also jaggies appear due to the on / off of the substitution at the edge. If the reduction ratio is increased (the output size is increased), the subsequent window size can be increased, but the processing amount increases. Therefore, in the first embodiment, the reduction ratio is described as 1/4.
[0015]
When the reduction of the image is completed, the process proceeds to step S203, where it is determined whether or not the processing has been completed up to the end of the image. If it has been processed, the process ends. If not, the process proceeds to step S204 where the pixel extraction unit 102 inputs Reduced pixel data within a predetermined shape (window) is extracted around the reduced pixel corresponding to the pixel (target pixel) position. This corresponds to a tap (window) of the two-dimensional FIR filter, and is shifted by the movement of the input pixel (target pixel) position. As described above, since the reduction ratio is ¼, when the reduced pixel is generated from a 4 × 4 block, when the input pixel moves 4 pixels in the horizontal direction, the window shifts by 1 pixel in the horizontal direction. When the input pixel moves 4 lines in the vertical direction, the window shifts in the vertical direction by 1 line. That is, when the input pixel moves and the reduced pixel corresponding to the input pixel is shifted by one pixel, the window is also shifted by one pixel.
[0016]
Here, if the window size is 4 × 4 and the reduction ratio is 1/4, it corresponds to a 16 × 16 window on the original input image, and is particularly effective in reducing spotted noise up to a size of 8 × 8. is there. In the following description, the window size is assumed to be 4 × 4.
[0017]
Next, in step S205, replacement data is generated from the extracted pixels in the window. Although details of this processing will be described later, the in-window average value calculation unit 103, the region determination unit 104, the region 0 average value calculation unit 105, the region 1 average value calculation unit 106, the region 0 difference value generation unit 107, and the region 1 This is processing performed by the difference value generation unit 108, the in-window difference value generation unit 109, and the comparison unit 111.
[0018]
FIG. 3 is a flowchart showing replacement data generation processing in the first embodiment. First, in step S301, the region determination unit 104 determines whether the pixel in the window is the region 0 or the region 1. Specifically, the average value of pixels in the window is obtained by the average value calculation unit 103 in the window, and the pixels in each window are compared with the window average value. If the average value is equal to or less than the window average value, the region 0 is set. Set to 1. If the pixel is area 0, the process proceeds to step S302, and the area 0 average value calculation unit 105 obtains the average value of the pixels. If the pixel is region 1, the process proceeds to step S303, and the region 1 average value calculation unit 106 obtains the average value of the pixels. In step S304 or S305, the region 0 difference value generation unit 107 or the region 1 difference value generation unit 108 obtains the absolute difference value D0 or D1 between the average value of each region and the input pixel value.
[0019]
In step S306, the in-window difference value generation unit 109 obtains the absolute difference value Dw between the above window average value and the input pixel value. In steps S307 to S309, the comparison unit 111 compares the difference absolute values D0 and D1 between the average value of each region and the input pixel value with the difference absolute value Dw between the window average value and the input pixel value. As a result, an average value that is the minimum difference absolute value (minimum difference absolute value D) is used as replacement data (steps S310 to S312).
[0020]
That is, if the difference absolute value D0 of the area 0 is minimum, the average value A0 obtained by the area 0 average value calculation unit 105 is used as replacement data. Similarly, if the difference absolute value D1 of the region 1 is minimum, the average value A1 obtained by the region 1 average value calculation unit 106 is used as replacement data. If the difference value Dw in the window is minimum, the average value in the window The average value Aw obtained by the calculation unit 103 is used as replacement data.
[0021]
Returning to FIG. 2, in step S206, when the comparison unit 111 and the pixel replacement unit 112 satisfy a predetermined condition for the above-described minimum difference absolute value D, the input pixel value is replaced with the generated replacement data. When the predetermined condition is not satisfied, the input pixel value is output as it is. When the above process is repeated until the end of the input image (YES in step S203), this process ends.
[0022]
FIG. 4 is a flowchart showing the replacement process in the first embodiment. First, in step S401, the comparison unit 111 compares the minimum difference absolute value D with the threshold Th1, and if the minimum difference absolute value D is less than the threshold Th1, the process proceeds to step S402, and the pixel replacement unit 112 outputs replacement data. If the minimum difference absolute value D is greater than or equal to the threshold value Th1, the process proceeds to step S403, and the input pixel value is output.
[0023]
Through the series of processes described above, noise can be removed by replacing the input pixel with the average value of the absolute difference value that is the minimum difference less than the threshold Th1 among the obtained absolute difference values as replacement data.
[0024]
As described above, for example, when reducing patchy noise up to a size of 8 × 8, if a reduced image is not used, a 16 × 16 window is required, and it is necessary if the number of pixels in the horizontal direction is W. The capacity of the line memory is 15 W, but in the first embodiment, the required line memory capacity is 3 lines × (1/4) W = (3/4) W by using a 1/4 reduced image. And reduced to 1/20.
[0025]
As described above, according to the first embodiment, noise and a significant signal are separated by comparing the average value and the input pixel value, and by generating replacement data from the reduced image, noise can be reduced to a lower frequency component. By making the removal effect effective and using the average value for each area, the number of replacement pixels near the edge can be increased, the range in which the noise removal effect is effective can be expanded, and the simple operation has fewer harmful effects and is more effective. Noise removal effect can be obtained.
[0026]
In the first embodiment, the pixel average in the window is used for the region determination. However, the present invention is not limited to this. For example, the value of the pixel located at the center when the pixels in the window are arranged in descending order. Or the median pixel range in the window. Also, the replacement pixel candidates may be the next largest number after the maximum value or the next smallest number after the minimum value instead of the average value of each region.
[0027]
[Second Embodiment]
Next, a second embodiment according to the present invention will be described in detail with reference to the drawings.
[0028]
In the second embodiment, in a window where the difference between the average values of the two areas is small, the replacement data is used as the average value in the window.
[0029]
FIG. 5 is a block diagram illustrating a configuration of the image processing apparatus according to the second embodiment. 5, reference numerals 501 to 506 and 507 are the same as the processes 101 to 106 and 110 shown in FIG. 1 in the first embodiment, and the description thereof is omitted. Reference numeral 508 denotes an average value calculation unit, 509, 513, and 514 denote comparison units, 710 and 711 denote selection units, 712 denotes a subtraction unit, 715 denotes a difference value generation unit, and 716 denotes a pixel replacement unit. Here, only a different part from 1st Embodiment is demonstrated.
[0030]
The average value calculation unit 508 calculates an average value of the average value of region 0 and the average value of region 1. That is, if the average value of region 0 is A0 and the average value of region 1 is A1, the average value A01 of region 0 and region 1 is A01 = (A0 + A1) / 2. The comparison unit 509 compares the average value A01 of the region 0 and the region 1 with the input pixel value Pi. Here, if the input pixel value Pi is less than the average value A01 of the region 0 and the region 1, the selection unit 510 outputs the average value A0 of the region 0, and otherwise outputs the average value A1 of the region 1. In other words, the average value closer to the input pixel value Pi is selected from the selection unit 510. Further, the subtraction unit 512 obtains the difference D10 between the average values of the region 0 and the region 1, and the comparison unit 513 compares the difference D10 with the threshold value Th2. Aw is output, otherwise the output of the selection unit 510, that is, the average value closer to the input pixel value Pi is output. That is, the output of the selection unit 511 becomes replacement data. Here, since A1> A0, the difference D10 between the average values of the area 0 and the area 1 is D10 = A1-A0.
[0031]
The difference value generation unit 515 generates an absolute difference value Dt between the input pixel value Pi and the output (replacement data) of the selection unit 511. The difference absolute value Dt is compared with the threshold value Th1 by the comparison unit 514, and the pixel replacement unit 516 replaces the input pixel value Pi with the output (replacement data) of the selection unit 511 only when the difference absolute value Dt is less than the threshold value Th1. In other cases, the pixel replacement unit 516 outputs the input pixel value Pi as it is.
[0032]
Next, processing of the image processing apparatus in the second embodiment will be described. In the second embodiment, S201 to S204 in FIG. 2 in the first embodiment are the same, and here, the replacement data generation process (S205) and the replacement process (S206) will be described in detail.
[0033]
FIG. 6 is a flowchart showing replacement data generation processing in the second embodiment. First, in step S601, the area determination unit 504 determines whether the pixel in the window is the area 0 or the area 1. Specifically, the window average value calculation unit 503 calculates the average of the pixel values in the window, compares the pixels in each window with the window average value, and sets the area to 0 if the window average value is less than or equal to the window average value. Set to 1. If the pixel is area 0, the process proceeds to step S602, and the area 0 average value calculation unit 505 obtains the average value of the pixels. If the pixel is region 1, the process proceeds to step S603, and the region 1 average value calculation unit 506 obtains the average value of the pixels.
[0034]
Next, in step S604, the subtraction unit 512 obtains a difference value between the average values of the two areas, and the comparison unit 513 compares the difference value with the threshold Th2. If the obtained difference value is equal to or smaller than the threshold value Th2, the process proceeds to step S608, and the replacement data is the window average value Aw. That is, the selection unit 511 outputs the in-window average value Aw as replacement data. If the obtained difference value is larger than the threshold value Th2, the process proceeds to step S605, where the average value calculation unit 508 obtains the average value of the two area average values, and the comparison unit 509 compares it with the input pixel value Pi. In step S605, instead of comparing the average value of the two areas, a comparison is made between the addition value of the two area average values and the value obtained by doubling the input pixel data Pi (1 bit left shift). .
[0035]
If the input pixel data Pi is less than the average value of the two area average values, the process proceeds to step S606, and the replacement data is the area 0 average value A0. Otherwise, the process proceeds to step S607, and the replacement data is the area 1 average value. A1. That is, it is determined whether or not the input pixel data Pi is closer to any region average value, and the selection unit 510 outputs the closer region average value as replacement data.
[0036]
FIG. 7 is a flowchart showing replacement processing in the second embodiment. First, in step S701, a difference value D between the input pixel value and the replacement data that is the output of the selection unit 511 is obtained, and in step S702, the comparison unit 514 compares it with the threshold value Th1. If the difference value D is less than the threshold value Th1, the process advances to step S703, and the pixel replacement unit 516 outputs replacement data. If the difference value D is greater than or equal to the threshold value Th1, the process proceeds to step S704, and the input pixel value is output.
[0037]
As described above, according to the second embodiment, in a window where the difference between the average values of the two regions is small, the replacement data is set to the average value in the window, so that the noise removal effect in the flat portion can be increased. In addition, the processing is simplified.
[0038]
[Third Embodiment]
Next, a third embodiment according to the present invention will be described in detail with reference to the drawings.
[0039]
In the third embodiment, an average value of two regions is output as replacement data in accordance with the random number generated by the random number generation unit in the replacement data generation process described in the second embodiment.
[0040]
FIG. 8 is a block diagram illustrating a configuration of an image processing apparatus according to the third embodiment. 8, reference numerals 801 to 807 and 810 to 812 are the same as the processes 501 to 507 and 514 to 516 shown in FIG. 5 in the second embodiment, and the description thereof is omitted. Reference numeral 808 denotes a random number generator, and 809 denotes a selector. Here, only a different part from 2nd Embodiment is demonstrated.
[0041]
This random number generation unit 808 generates pseudo-random numbers by a known method, and controls the selection unit 809 to randomly select the average value of region 0 or the average value of region 1 as replacement data. Then, the difference value generation unit 811 generates an absolute difference value Dt between the input pixel value Pi and the output (replacement data) of the selection unit 809. The difference absolute value Dt is compared with the threshold value Th1 by the comparison unit 810, and the pixel replacement unit 812 replaces the input pixel value Pi with the output (replacement data) of the selection unit 809 only when the difference absolute value Dt is less than the threshold value Th1. In other cases, the pixel replacement unit 812 outputs the input pixel value Pi as it is.
[0042]
Here, processing of the image processing apparatus in the third embodiment will be described. Since the processing other than the replacement data generation processing is the same as that of the second embodiment, description thereof is omitted.
[0043]
FIG. 9 is a flowchart showing replacement data generation processing in the third embodiment. First, in step S901, the region determination unit 804 determines whether the pixel in the window is the region 0 or the region 1. Specifically, the window average value calculation unit 803 calculates the average of the pixel values in the window, compares the pixels in each window with the window average value, and sets the area to 0 if it is equal to or less than the window average value. Set to 1. If the pixel is area 0, the process proceeds to step S902, and the area 0 average value calculation unit 805 obtains the average value of the pixels. If the pixel is region 1, the process proceeds to step S903, and the region 1 average value calculation unit 806 obtains the average value of the pixels.
[0044]
In step S904, the random number generation unit 808 generates a 1-bit random number. If the random number is 0, the process proceeds to step S905, and the selection unit 809 outputs the area 0 average value A0 as replacement data. If the random number is not 0, the process proceeds to step S906, and the area 1 average value A1 is output as replacement data.
[0045]
As described above, according to the third embodiment, since the low frequency noise is modulated to the high frequency side by the random number by randomly selecting the average value of the region 0 or the region 1 and using it as the replacement data, the display system And visual MTF noise removal effect. Further, the processing is further simplified. Further, by limiting the number of consecutive 0s and the number of 1s in the random number output described above, the frequency characteristics of the random numbers are shifted to the high frequency side, so that a better noise removal effect can be obtained.
[0046]
[Fourth Embodiment]
Next, a fourth embodiment according to the present invention will be described in detail with reference to the drawings.
[0047]
In the fourth embodiment, the pixels in the window are selected at random according to the random number generated by the random number generation unit, and output as replacement data.
[0048]
FIG. 10 is a block diagram illustrating a configuration of an image processing apparatus according to the fourth embodiment. In FIG. 10, reference numerals 1001, 1002, and 1003 are the same as the image reduction unit, pixel extraction unit, and timing adjustment unit in the first to third embodiments, and 1006 to 1008 are comparison units in the second and third embodiments. This is the same as the difference value generation unit and the pixel replacement unit, and a description thereof will be omitted. Reference numeral 1004 denotes a random number generator, and reference numeral 1005 denotes a pixel selector. Here, only parts different from the above-described embodiment will be described.
[0049]
Similar to the third embodiment, the random number generation unit 1004 generates pseudo-random numbers by a known method, controls the pixel selection unit 1005 to randomly select pixels in the window, and uses them as replacement data. Similar to the second and third embodiments, the difference value generation unit 1007 generates an absolute difference value Dt between the input pixel value Pi and the output (replacement data) of the pixel selection unit 1005. This difference absolute value Dt is compared with the threshold value Th1 by the comparison unit 1006, and the pixel replacement unit 1008 replaces the input pixel data Pi with the output (replacement data) of the pixel selection unit 1005 only when the difference absolute value Dt is less than the threshold value Th1. . In other cases, the pixel replacement unit 1008 outputs the input pixel value Pi as it is.
[0050]
If the window size in the fourth embodiment is 4 × 4, the number of pixels in the window is 16, so the number of random digits is 4 bits.
[0051]
FIG. 11 is a diagram illustrating a configuration example of a 9-bit M-sequence pseudo-random code generation circuit. In FIG. 11, reference numeral 1101 denotes a shift register, and 1102 denotes an EXOR circuit. In this M-sequence pseudo-random code generation circuit, a value other than 0 is loaded in advance into the shift register 1101, the result of EXORing the bit corresponding to the primitive polynomial of the shift register is input to bit 0, and left-shifted. Get the next random number. In the fourth embodiment, since the number of stages of the shift register 1101 is nine, a ninth-order primitive polynomial X9 + X4 + 1 is used. Therefore, the result of EXORing bits 8 and 3 of the shift register 1101 by the EXOR circuit 1302 is used as the input of bit 0 and is shifted to the left by the next clock. With this configuration, random numbers 1 to 511 are generated in 511 cycles.
[0052]
Here, since the range of the random number generation unit 1004 is 0 to 15, random numbers of 0 to 15 in 511 periods can be obtained by using arbitrary 4 bits of the shift register 1101. Further, the random number generation means is not limited to this method, and a pseudo random number may be generated by a known method.
[0053]
In addition, the horizontal position and the vertical position in the window can be generated by pseudo random numbers, respectively, and the pixel to be replaced data can be determined.
[0054]
Further, the generation probability of the pseudo random number may be a uniform probability distribution, or may be a probability distribution depending on the distance from the pixel corresponding to the input pixel in the window. For example, in order to increase the low-frequency noise removal effect by replacement, a higher probability of occurrence may be given to a pixel farther from the pixel corresponding to the input pixel.
[0055]
In addition, if the replacement is not performed in the comparison with the threshold value described above, another new replacement data is generated from the window of interest until the replacement or the predetermined number of times is reached, and is compared with the threshold value. Thus, the probability of replacement can be increased.
[0056]
As described above, according to the fourth embodiment, since the pixels in the window are selected at random and used as replacement data, low-frequency noise is modulated to the high-frequency side by the generated random number despite the simple configuration. Therefore, a good noise removal effect by the display system or visual MTF can be obtained. Further, by inverting or non-inverting the random number output as the input pixel is updated, the frequency characteristic of the random number is shifted to the high frequency side, so that a better noise removal effect can be obtained.
[0057]
[Fifth Embodiment]
Next, a fifth embodiment according to the present invention will be described in detail with reference to the drawings.
[0058]
In the fifth embodiment, an input image is classified into a plurality of categories, and replacement data corresponding to each category is generated.
[0059]
FIG. 12 is a block diagram illustrating a configuration of an image processing apparatus according to the fifth embodiment. In FIG. 12, reference numerals 1201 to 1203 and 1207 to 1209 are the same as the processes 1001 to 1003 and 1006 to 1008 shown in FIG. 10 in the fourth embodiment, and a description thereof will be omitted. 1204 is a categorizing unit, 1205 is a replacement data generating unit, and 1206 is a replacement data selecting unit. Here, only parts different from the fourth embodiment will be described.
[0060]
The categorizing unit 1204 classifies the input image into a plurality of categories. The replacement data generation unit 1205 generates replacement data for noise removal using a plurality of algorithms as described in the first to fourth embodiments or a plurality of different threshold values. Then, the replacement data selection unit 1206 selects replacement data suitable for the category according to the category. Thereafter, the same determination as that in the fourth embodiment is performed, and the noise is removed by replacing the input pixel.
[0061]
The categories here are classified on the basis of, for example, a part that is harmful when the replacement process is performed, a part where low-frequency noise is conspicuous, a part where high-frequency noise is conspicuous, and a part where both noises are conspicuous. In particular, in the first to fourth embodiments, the appearance of harmful effects is different, and compatibility with the input image exists.
[0062]
For example, in a portion with many fine textures such as cloth, it is better to perform the averaging process, so the fourth embodiment is suitable. Also, in the dark part, the second embodiment is suitable because it looks better when averaged. Furthermore, the appearance of noise varies depending on the color gamut of the input pixel. Moreover, the level difference recognized as a flat part is also different. Therefore, the optimum value of the threshold Th2 in the second embodiment is also different.
[0063]
As described above, according to the fifth embodiment, replacement data is generated by a plurality of algorithms (including threshold control), and replacement data matching the category is selected to replace the input pixel. Noise removal effect can be obtained.
[0064]
[Sixth Embodiment]
Next, a sixth embodiment according to the present invention will be described in detail with reference to the drawings.
[0065]
In the sixth embodiment, a threshold selection unit for selecting a threshold from a plurality of thresholds according to the category is added to the fifth embodiment.
[0066]
FIG. 13 is a block diagram illustrating a configuration of an image processing apparatus according to the sixth embodiment. In FIG. 13, reference numerals 1301 to 1309 are the same as the processes shown in FIG. 12 in the fifth embodiment, and a description thereof will be omitted. Reference numeral 1310 denotes a threshold selection unit. Here, only parts different from the fifth embodiment will be described.
[0067]
The categorizing unit 1304 classifies the input image into a plurality of categories. The threshold selection unit 1310 selects a threshold Th1 suitable for the category according to the classified category. Thereafter, the same determination as that of the fifth embodiment is performed, and the noise is removed by replacing the input pixel.
[0068]
The categories here are classified on the basis of how harmful effects appear when, for example, replacement processing is performed. That is, the noise removal effect is strengthened by increasing the threshold value in a portion where the harmful effect is not noticeable, and the bad effect is made inconspicuous by reducing the threshold value in a portion where the harmful effect is noticeable.
[0069]
Further, when the input gain is different due to AGC (Auto Gain Control) or the like, the amplitude of the noise is different, so it is desirable to change the threshold Th1 also by the input gain.
[0070]
In the sixth embodiment, a plurality of replacement generation units and threshold values are associated with each other. However, the present invention is not limited to this, and the number of replacement generation units and the number of threshold values may be different. For example, replacement data may be generated from one replacement generation unit. In this case, the replacement data selection unit 1306 is omitted.
[0071]
As described above, according to the sixth embodiment, the threshold value corresponding to the category is selected and the input pixel is replaced, so that a good noise removal effect with less harmful effects can be obtained.
[0072]
[Other embodiments]
A recording medium in which a program code of software that realizes the functions of the above-described embodiments is recorded is supplied to the system or apparatus, and the computer (CPU or MPU) of the system or apparatus reads the program code stored in the recording medium. It goes without saying that the object of the present invention can also be achieved by executing.
[0073]
In this case, the program code itself read from the recording medium realizes the novel function of the present invention, and the recording medium on which the program code is recorded constitutes the present invention.
[0074]
As a recording medium for supplying the program code, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.
[0075]
In addition, the functions of the above-described embodiments are realized by executing the program code read by the computer, and the OS running on the computer is part of the actual processing based on the instruction of the program code. Alternatively, the functions of the above-described embodiment can be realized by performing all of the processes and performing the processing.
[0076]
Further, after the program code read from the recording medium is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion board is based on the instruction of the program code. The CPU of the function expansion unit or the like performs part or all of the actual processing, and the functions of the above-described embodiments can also be realized by the processing.
[0077]
The present invention can also be applied to a case where the program is distributed to a requester via a communication line such as personal computer communication from a storage medium in which a program code of software realizing the functions of the above-described embodiments is recorded. Needless to say.
[0078]
【The invention's effect】
As described above, according to the present invention, noise and significant signals are separated by comparing the average value and the input pixel value, and noise is removed to lower frequency components by generating replacement data from the reduced image. By making the effect effective and using the average value for each area, the number of replacement pixels near the edge can be increased, and the range in which the noise removal effect is effective can be expanded. Noise removal effect can be obtained. In particular, the capacity of the line memory required for processing is greatly reduced.
[0079]
Further, in a window where the difference between the two area averages is small, the replacement data is set to the average value in the window, so that the noise removal effect in the flat portion can be increased, and the processing is simplified.
[0080]
In addition, by selecting an average value of region 0 or region 1 at random and using it as replacement data, low frequency noise is modulated to a high frequency side by random numbers, and a noise removal effect by a display system or visual MTF can be obtained. Further, the processing is further simplified.
[0081]
In addition, after removing high-frequency noise by generating a reduced image, pixels in the reduced image window are randomly selected and used as replacement data, so that a good noise removal effect can be obtained without losing detail.
[0082]
In addition, by generating replacement data using multiple algorithms including threshold control and selecting replacement data that matches the category and replacing the input pixels, a good noise reduction effect that matches the characteristics of the image is obtained. It is done.
[0083]
Also, by replacing the input pixel by selecting a threshold value that matches the category, a good noise removal effect that has less adverse effects and matches the characteristics of the image can be obtained.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration of an image processing apparatus according to a first embodiment.
FIG. 2 is a flowchart illustrating processing of the image processing apparatus according to the first embodiment.
FIG. 3 is a flowchart showing replacement data generation processing in the first embodiment.
FIG. 4 is a flowchart showing replacement processing in the first embodiment.
FIG. 5 is a block diagram illustrating a configuration of an image processing apparatus according to a second embodiment.
FIG. 6 is a flowchart showing replacement data generation processing in the second embodiment.
FIG. 7 is a flowchart showing replacement processing in the second embodiment.
FIG. 8 is a block diagram illustrating a configuration of an image processing apparatus according to a third embodiment.
FIG. 9 is a flowchart showing replacement data generation processing in the third embodiment.
FIG. 10 is a block diagram illustrating a configuration of an image processing apparatus according to a fourth embodiment.
FIG. 11 is a diagram illustrating a configuration example of a 9-bit M-sequence pseudo-random code generation circuit;
FIG. 12 is a block diagram illustrating a configuration of an image processing apparatus according to a fifth embodiment.
FIG. 13 is a block diagram illustrating a configuration of an image processing apparatus according to a sixth embodiment.
[Explanation of symbols]
101 Image reduction unit
102 pixel extraction unit
103 Average value calculator in window
104 Area determination unit
105 Area 0 average value calculator
106 Region 1 average value calculator
107 area 0 difference value generation unit
108 Region 1 difference value generator
109 In-window difference value generator
110 Timing adjustment unit
111 comparator
112 Pixel replacement unit

Claims (14)

Reducing the input image at a predetermined magnification to obtain a reduced image;
Extracting a pixel in a window of a predetermined size corresponding to the position of the input pixel from the reduced image;
An average value of the extracted pixels is obtained, a determination is made as to which of the two regions the input pixel is, a representative value is obtained for each of the determined regions, and the average value, the two representative values, and the input pixel Obtaining a difference value with each value, and selecting, as replacement data, the average value and the two representative values having the smallest difference value;
Obtaining a difference value between the replacement data and the value of the input pixel;
Comparing the difference value with a first threshold;
And a step of replacing the value of the input pixel with the replacement data when the difference value is less than the first threshold value. Reducing the input image at a predetermined magnification to obtain a reduced image;
Extracting a pixel in a window of a predetermined size corresponding to the position of the input pixel from the reduced image;
An average value of the extracted pixels is obtained, a determination is made as to which of the two regions each pixel is, a representative value is obtained for each determined region, and a difference between the obtained two region representative values is obtained. A difference between the two region representative values is compared with a second threshold, and when the difference between the two region representative values is equal to or less than the second threshold, the average value is used as replacement data;
The average of the two area representative values is obtained, the average of the two area representative values is compared with the value of the input pixel, and the smaller value is obtained when the value of the input pixel is less than the average of the two area representative values. A region representative value is selected, the larger region representative value is selected when the value of the input pixel is equal to or greater than the average of the two region representative values, and the difference between the two region representative values is the second threshold value. A step of using the selected representative value as replacement data when exceeding
Obtaining a difference value between the replacement data and the value of the input pixel;
Comparing the difference value with a first threshold;
And a step of replacing the value of the input pixel with the replacement data when the difference value is less than the first threshold value.   2. The image processing method according to claim 1, wherein the step of determining which of the two areas each pixel in the window is performed by comparison with an average value in the window.   The image processing method according to claim 1, wherein the step of determining which of the two regions each pixel in the window is performed in the order of the size of the pixels in the window.   2. The image processing method according to claim 1, wherein the step of determining which of the two regions each pixel in the window is performed by comparison with a median of a pixel range in the window.   The image processing method according to claim 1, wherein the two representative values are average values in each region.   2. The image processing method according to claim 1, wherein the two representative values are a second smallest pixel and a second largest pixel in the window.   2. The image processing method according to claim 1, further comprising a step of filtering the input image, wherein the step of obtaining the reduced image obtains a reduced image by reducing the filtered input image at a predetermined magnification. Means for reducing the input image at a predetermined magnification to obtain a reduced image;
Means for extracting pixels in a window of a predetermined size corresponding to the position of the input pixel from the reduced image;
An average value of the extracted pixels is obtained, a determination is made as to which of the two regions the input pixel is, a representative value is obtained for each of the determined regions, and the average value, the two representative values, and the input pixel A difference value with respect to each value, and a means for generating replacement data by selecting the average value and the two representative values having the smallest difference value;
Means for obtaining a difference value between the replacement data and the value of the input pixel;
Means for comparing the difference value with a first threshold;
An image processing apparatus comprising: means for replacing the value of the input pixel with the replacement data when the difference value is less than the first threshold value.   A program for causing a computer to execute the image processing method according to claim 1. The computer-readable recording medium which recorded the program of Claim 10 . Means for reducing the input image at a predetermined magnification to obtain a reduced image;
Means for extracting pixels in a window of a predetermined size corresponding to the position of the input pixel from the reduced image;
An average value of the extracted pixels is obtained, a determination is made as to which of the two regions each pixel is, a representative value is obtained for each determined region, and a difference between the obtained two region representative values is obtained. A difference between the two region representative values is compared with a second threshold, and when the difference between the two region representative values is equal to or less than the second threshold, the average value is used as replacement data;
The average of the two area representative values is obtained, the average of the two area representative values is compared with the value of the input pixel, and the smaller value is obtained when the value of the input pixel is less than the average of the two area representative values. A region representative value is selected, the larger region representative value is selected when the value of the input pixel is equal to or greater than the average of the two region representative values, and the difference between the two region representative values is the second threshold value. Means for using the selected representative value as replacement data when exceeding
Means for obtaining a difference value between the replacement data and the value of the input pixel;
Means for comparing the difference value with a first threshold;
An image processing apparatus comprising: means for replacing the value of the input pixel with the replacement data when the difference value is less than the first threshold value.   A program for causing a computer to execute the image processing method according to claim 2. A computer-readable recording medium on which the program according to claim 13 is recorded.

Images