JP4075256B2 - Image processing device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus that can suppress the occurrence of pseudo contours.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, there has been known an image processing apparatus that converts an input signal indicating a density level of each pixel with a predetermined number of gradations into a signal with a smaller number of gradations than the predetermined number of gradations. Further, an error diffusion method is known as a method for reproducing the density level of an input signal as a whole even when the gradation is reduced.
[0003]
FIG. 8 is a block diagram showing the configuration of an image processing apparatus that employs the error diffusion method.
Referring to the figure, an image processing apparatus includes an input unit 801 that inputs pixel values (input signals) one by one from a multi-tone image to be processed, and a subtracter that subtracts an error from the input pixel value. 803, threshold processing unit 807 for performing threshold processing on the output of the subtractor 803, output unit 809 for outputting the threshold processing result, and subtracting the output of the subtracter 803 from the threshold processing result And a subtractor 811 for calculating an error.
[0004]
In addition, the applicant of the present application has proposed a threshold value diffusion method for diffusing a difference between a threshold value and an output value to a threshold value used for processing of neighboring pixels in Japanese Patent Application No. 11-237492.
[0005]
FIG. 9 is a block diagram showing a configuration of an image processing apparatus using the threshold value diffusion method.
[0006]
Referring to the figure, an image processing apparatus includes an input unit 901 that inputs a pixel value of image data to be processed, a threshold processing unit 903 that performs threshold processing on the pixel value, and a threshold processing result. An output unit 905 for outputting, an initial threshold value generating unit 913 for outputting an initial threshold value Th (x), a subtractor 915 for subtracting a threshold correction value from the value of the initial threshold value generating unit 913, Inverting unit 907 that inverts the threshold processing result, subtracter 909 that outputs the correction value of the threshold by subtracting the output of subtractor 915 from the output of inversion unit 907, and the output of subtractor 909 And a β multiplier 911 that multiplies the coefficient β.
[0007]
In an image processing apparatus that employs the threshold value diffusion method, a threshold correction value is diffused to a threshold value used for processing of surrounding pixels.
[0008]
By using the image processing apparatus that outputs with the gradations reduced as described above, the gradation of the multi-gradation image can be binarized (images can be expressed in two states, whether or not dots are output). In addition, by using a plurality of threshold values, it is possible to make the gradation more than ternarization (an image is expressed in three or more states depending on the size of dots).
[0009]
[Problems to be solved by the invention]
In the feedback digital halftoning technique employing the above-described threshold value diffusion method or the like, there is a problem that a pseudo contour as shown in FIG. The pseudo contour is a phenomenon in which only specific dots (dot 1 shown in gray in FIG. 10) are continuously generated more than necessary as shown in the enlarged view on the right side of FIG.
[0010]
Conventionally, since the cause of the occurrence of such a pseudo contour has not been known, a pseudo contour has been made inconspicuous by applying a disturbance exclusively as a countermeasure.
[0011]
An object of the present invention is to provide an image processing apparatus that can prevent the occurrence of pseudo contours from the root.
[0012]
[Means for Solving the Problems]
The inventor of the present application has investigated the cause of the occurrence of the pseudo contour and invented an image processing apparatus capable of preventing the occurrence of the pseudo contour. First, the cause of the pseudo contour generation will be described.
[0013]
FIG. 11 is a block diagram illustrating a configuration of an image processing apparatus that employs a threshold value diffusion method that outputs an input value with reduced gradation. Here, a case where an input image is ternarized using two types of dots will be described as an example. That is, the image is reproduced in three states: a dark dot 2, a thin dot 1, and no dot (dot 0).
[0014]
Referring to the drawing, this apparatus inputs a density level (multi-value) of one pixel having a numerical value in the range of 0 to 1 as an input value, and outputs one of dots 0, 1, and 2.
[0015]
In such a case, it is determined which range the input value belongs to, and binarization is performed in each range. Here, there are two ranges, 0 to 0.5 (range a) and 0.5 to 1 (range b). That is, when the input value is in the range a of 0 to 0.5, either dot 0 or dot 1 is output, and when the input value is in the range b of 0.5 to 1, the dot 1 or dot 2 is output. Either one is output. In this way, the image is reproduced.
[0016]
More specifically, referring to FIG. 11, the image processing apparatus includes subtracters 103 and 109, a threshold processing unit 105, an inversion unit 107, and a β multiplication unit 111. The subtractor 103 obtains a correction value FB (also referred to as a correction value or a feedback value) of the threshold value of the neighboring pixels from the initial threshold values th1 and th2 (th1 belongs to the range a and th2 belongs to the range b). Subtraction is performed to obtain corrected threshold values Mod th1 and Mod th2.
[0017]
The threshold processing unit 105 compares the input value with the output (corrected threshold value) of the subtractor 103. Then, 0 is output if the input value <Mod th1, 0.5 if Mod th1 ≦ input value <Mod th2, and 1 if the input value ≧ Mod th2. If the output of the threshold processing unit 105 is 0, dot 0 is output. If the output of the threshold processing unit 105 is 0.5, dot 1 is output, and if the output of the threshold processing unit 105 is 1, dot 2 is output.
[0018]
The output result of the threshold processing unit 105 is inverted by the inversion unit 107. That is, in the case where the input value belongs to the range a, 0.5 is output from the inversion unit if the output result of the threshold processing unit 105 is 0, and 0 is output if it is 0.5. Further, when the input value belongs to the range b, 0.5 is output if the output result of the threshold processing unit 105 is 0.5, and 0 is output from the inverting unit if it is 1. The subtracter 109 subtracts Mod th1 from the output of the inversion unit 107 if the input value belongs to the range a, and Mod th2 if the input value belongs to the range b. The output of the subtracter 109 is multiplied by a coefficient β to obtain a correction value FB for the threshold value of the pixel.
[0019]
FIG. 12 is a diagram illustrating the relationship between the density of input image data and the density of output dots. In a range where the input is 0 → 0.5, the density of the dot 1 increases from 0 → 1. In the range where the input is 0.5 → 1, the density of dot 1 decreases from 1 → 0, and instead the density of dot 2 increases from 0 → 1. As a result, a proportional relationship is established between the input and the output as indicated by (3) for the entire image.
[0020]
Further, an image processing apparatus employing the threshold value diffusion method can be configured as shown in FIG.
[0021]
Referring to FIG. 13, the image processing apparatus includes a determination unit 201 that determines which range the input value belongs to, and a normalization unit that normalizes the input value so that the input value falls within a predetermined range. 203, a threshold processing unit 205 that performs threshold processing based on the corrected threshold Mth, an allocation unit 207 that allocates dots to be output based on the determination result by the determination unit 201 and the threshold processing result, An inversion processing unit 209 that inverts the threshold processing result, a subtractor 211 that subtracts the corrected threshold value Mth from the output of the inversion processing unit 209, a correction value memory 215 that stores the output result of the subtractor 211, and a correction Β multiplier 217 that multiplies the output of value memory 215 by coefficient β, initial threshold generator 220 that generates initial threshold th, and the output of β multiplier 217 is subtracted from initial threshold th. Composed of the subtractor 222 to modify the threshold Mth.
[0022]
Here, the input value input is assumed to have a range of 0 ≦ input ≦ 1, and the output is one of three states of 0 (no dot), out1 (= 0.5), and out2 (= 1). Is output.
[0023]
The determination unit 201 determines that the input value belongs to the range a if input <out1, and determines that the input value belongs to the range b if out1 ≦ input ≦ out2.
[0024]
When the normalization unit 203 receives the output from the determination unit 201 and the input value belongs to the range a, the normalization unit 203 outputs a value of Rin = input / out1. On the other hand, when the input value belongs to the range b, a value of Rin = (input-out1) / (out2-out1) is output.
[0025]
The threshold processing unit 205 compares the value of Rin and the corrected threshold value Mth, and outputs “0” if Rin ≦ Mth, and outputs “1” if Rin> Mth.
[0026]
The allocating unit 207 outputs “0” if the input value belongs to the range a and the threshold processing result is “0” based on the output of the determination unit 201 and the threshold processing result. If the input value belongs to the range a and the threshold processing result is “1”, out1 is output. If the input value belongs to the range b and the threshold processing result is “0”, out1 is output. If the input value belongs to the range b and the threshold processing result is “1”, out2 is output.
[0027]
The inversion processing unit 209 outputs “1” if the threshold processing result is “0”, and outputs “0” if the threshold processing result is “1”.
[0028]
Here, for the sake of simplification of explanation, the input value is 0 to 1 as in FIGS. 11 and 12, and the case of ternarizing the input value is described as an example. By further increasing the range to be processed, it is possible to perform image processing of four or more values.
[0029]
Next, the operation of the apparatus shown in FIG. 13 will be described with reference to FIG. As shown in FIG. 14 (1), the input value belongs to the range of 0-1. Here, when the input value is 0 or more and less than 0.5, the determination unit 201 determines that the input value belongs to the range a. When the input value is 0.5 or more and 1 or less, the input value belongs to the range b. Is determined. When the input value belongs to the range a, the input value is normalized so as to be in the range of 0 to 1 as shown in FIG. On the other hand, when the input value belongs to the range b, the input value is normalized so as to be in the range of 0 to 1 as shown in FIG. In both range a and range b, threshold processing is performed by threshold processing unit 207 with modified threshold Mth. In the range a, when the normalized input value is less than or equal to the correction threshold value Mth, no dot is output (dot 0), and when it exceeds the correction threshold value Mth, dot 1 (out1) Is output.
[0030]
On the other hand, in the range b, the dot 1 (out1) is output when the normalized input value is equal to or smaller than the correction threshold value Mth, and the dot 2 (out2) is output when the input value exceeds the correction threshold value Mth. . In this way, the same processing as that shown in FIGS. 11 and 12 can be performed in FIG.
[0031]
FIG. 15 is a diagram illustrating a state in which the threshold correction value (correction value or feedback value) FB is diffused. With reference to (1) in the figure, it is assumed that the input value can take a value of 0 to 1. When the input value is in the range of 0 to 0.5 (range a), it is determined whether the dot 0 (no dot) or dot 1 is set with the initial threshold value 0.25. On the other hand, when the input value is in the range of 0.5 to 1 (range b), it is determined whether to set dot 1 or dot 2 at the initial threshold value 0.75.
[0032]
Assuming that the input value is constant at 0.45 in any pixel, in the first pixel processing (1), the input value is compared with the initial threshold value 0.25 in the range a. Since input value> initial threshold value, dot 1 is output. A correction value FB is calculated based on the result of the threshold processing and is diffused to surrounding pixels.
[0033]
In the next pixel processing (2), a value obtained by subtracting the correction value FB from the initial threshold value (correction threshold value) is compared with the input value. In the process of FIG. 15, the correction value FB is accumulated by the processes of (1) to (4), and the input value is below the correction threshold value in the process of (4). Therefore, dot 1 is output in the processes (1) to (3), and dot 0 (no dot) is output in the process (4).
[0034]
Since the correction value FB decreases in the process (4), the input value again exceeds the correction threshold value in the process (5), and dot 1 is output. Then, the same processing as the processing after (2) is performed. In this way, the correction threshold value swings around the input value 0.45.
[0035]
Next, referring to FIG. 16, when the input value changes from 0.45 (10% for no dot, 90% for dot 1) to 0.55 (90% for dot 1 and 10% for dot 2). Consider.
[0036]
As shown in (1) of FIG. 16, it is assumed that the correction value FB has already been accumulated when the input value is 0.45. Here, if the input value has changed to 0.55 as shown in (2), the correction value FB is also taken over.
[0037]
In the processes after (2), the dot 1 is output until the correction value FB accumulated in (1) is canceled and the input value 0.55 exceeds the correction threshold (until the process (6)). Will continue.
[0038]
Thus, the period during which only dot 1 is output is relatively long (a dot delay occurs), which causes a false contour. When the input value is 0.55, it is normal that dot 1 is mixed at a rate of 90% and dot 2 at a rate of 10%. Therefore, it is necessary that the dot 2 appears in the first pixel or the second pixel. .
[0039]
Therefore, in the present invention, an image processing apparatus capable of preventing the occurrence of a pseudo contour by adopting the following means is provided.
[0040]
That is, according to an aspect of the present invention, an image processing apparatus is an image processing apparatus that converts an input signal indicating a density level of each pixel with a predetermined number of gradations into a signal with a smaller number of gradations than a predetermined number of gradations. A first output for comparing the input signal with a first threshold value and outputting a signal of the first gradation or the second gradation when the density level of the input signal is within the first range. And when the density level of the input signal is within a second range following the first range, the input signal is compared with a second threshold value to generate a second gradation or third gradation signal. Second output means for outputting, switching means for discriminating and switching which one of the first and second output means to use according to the density level of the input signal for each pixel, and the first or second output Output from the first or second output means and the threshold value used for comparison by the means Issue Inverted signal Based on the difference between A correction value obtained by multiplying the difference by a predetermined coefficient is output to surrounding pixels, Subsequent pixel processing The first or second threshold value is corrected by subtracting the correction value output to the peripheral pixels at the time from the initial first or second threshold value. Correction means and calculation method of correction value calculated by the correction means at the time of switching from the first output means to the second output means or at the time of switching from the second output means to the first output means Change means to change Include . The changing means inverts the sign of the correction value at the time of switching. .
[0042]
According to another aspect of the present invention, an image processing apparatus is an image processing apparatus that converts an input signal indicating a density level of each pixel with a predetermined number of gradations into a signal with a smaller number of gradations than a predetermined number of gradations. Determining means for determining which range the density level of the input signal belongs to; , Enter Normalizing means for normalizing the density level of the input signal so that the density level of the force signal falls within a predetermined range, and comparing means for comparing the normalized density level with a threshold value and outputting a comparison result An output means for outputting a signal having a smaller number of gradations than a predetermined number of gradations based on the determination result by the determination means and the comparison result by the comparison means; and a comparison means By subtracting a correction value obtained by multiplying the difference by a predetermined coefficient from the initial threshold value when processing subsequent pixels based on the difference between the inverted signal of the signal output from the threshold value and the threshold value. Correction means for correcting the threshold value; Including. The normalizing means reverses the density level of the input signal so that the density level of the input signal falls within a predetermined range when the density level of the input signal is determined to belong to a specific range by the determining means. Normalize .
[0043]
According to still another aspect of the present invention, an image processing apparatus converts an input signal indicating the density level of each pixel with a predetermined number of gradations into a signal with a smaller number of gradations than the predetermined number of gradations. An output means for outputting a value serving as a basis for threshold calculation, a threshold value calculating means for calculating at least two threshold values based on the value serving as a basis for threshold calculation, and a calculation Threshold processing means for performing threshold processing of an input signal based on at least two threshold values that have been processed, and processing results by the threshold processing means And a correction value obtained by multiplying the difference by a predetermined coefficient in the subsequent pixel processing based on the difference between the evaluation means for evaluating the output based on By subtracting from the initial threshold Correction means for correcting the value that is the basis for threshold calculation Including. The threshold processing means outputs a first gradation signal when the density level of the input signal is within the first range up to the first threshold, and the density level of the input signal is the first level. When the signal falls within the second range from the first threshold value following the range to the second threshold value, the second gradation signal is output, and the density level of the input signal follows the second range. When the signal falls within the third range equal to or greater than the threshold value of 2, a third gradation signal is output. The threshold value calculation means calculates a value obtained by multiplying a value serving as a basis for threshold value calculation and the output value of the second gradation signal as the first threshold value, and outputs the third gradation signal. The value obtained by subtracting the output value of the second gradation signal from the value is multiplied by a value serving as a basis for threshold calculation, and the value obtained by subtracting the multiplied value from the output value of the third gradation signal is the second value. Calculate as threshold value .
[0044]
DETAILED DESCRIPTION OF THE INVENTION
[First Embodiment]
FIG. 1 is a block diagram showing the configuration of the image processing apparatus according to the first embodiment of the present invention.
[0045]
This image processing apparatus has the same basic configuration as the image processing apparatus shown in FIG. 11 is different from the apparatus of FIG. 11 in that the image processing apparatus according to the present embodiment includes an inversion unit 101 that inverts the sign of the correction value FB from the neighboring pixels, and an inversion unit that inverts the sign of the output of the β multiplication unit 111. 113.
[0046]
Both the inverting unit 101 and the inverting unit 113 execute a process of multiplying by −1 when the input value is 0.5 or more. Further, when the input value is less than 0.5, the inversion unit 101 and the inversion unit 109 do nothing.
[0047]
That is, when the range to which the input value belongs is switched, the correction value calculation method is changed (more specifically, the sign of the correction value is inverted).
[0048]
When such processing is performed, when the input value is only 0.5 to 1, the correction value FB is multiplied by -1 when the correction value is sent from the β multiplier 111 to another pixel, and the correction value is captured. Since the reversal unit 101 again multiplies the correction value FB by −1, the signs of the correction values cancel each other out and processing by the threshold value diffusion method is normally performed. Further, when the input value is only 0 to 0.5, the inversion units 101 and 109 do nothing, so that the threshold value diffusion method can be normally performed. Also, when the input value moves from a range smaller than 0.5 to a larger range (or vice versa), the sign of the correction value is inverted. Thereby, generation | occurrence | production of the pseudo contour based on the delay of dot generation can be prevented.
[0049]
FIG. 2 is a diagram for explaining the operation of the image processing apparatus of FIG.
Referring to the figure, it is assumed that in the state (1), the input value is 0.45 and the upward (+ direction) correction value FB is accumulated. If the input value exceeds 0.5 and becomes 0.55 in the state (2), the sign of the correction value FB is reversed by the reversing unit 101 and the reversing unit 109. As a result, the correction value FB is in the downward direction (− direction). Thereby, the dot 2 is output immediately.
[0050]
This is the same when the input value is less than 0.5 from a value exceeding 0.5. By such processing, dot delay is prevented, so that occurrence of pseudo contour can be suppressed.
[0051]
FIG. 3 is a block diagram showing a modification of the image processing apparatus of FIG. The reversing units 101 and 113 of this apparatus execute a process of multiplying by -1 when the input value is 0.25 or less or when the input value is 0.5 or more and 0.75 or less. That is, in the present embodiment, the sign of the correction value FB is inverted not only when the input value changes across the range but also when the input value changes across the threshold value (initial threshold value). In this way, if the inversion of the correction value is performed finely, it is possible to reduce the delay of dots that occurs when the input value changes across the threshold value.
[0052]
Note that the apparatus shown in FIG. 3 is also applicable to binarization processing. That is, also in the binarization processing, it is possible to prevent dot delay by inverting the sign of the error when the input changes across the threshold value.
[0053]
[Second Embodiment]
FIG. 4 is a block diagram showing the configuration of the image processing apparatus according to the second embodiment of the present invention. This image processing apparatus has the same basic configuration as the image processing apparatus shown in FIG. However, in the present embodiment, the image processing apparatus is provided with an inverting unit 213 for inverting the sign of the output of the subtractor 211 and an inverting unit 219 for inverting the sign of the output of the β multiplier 217. .
[0054]
Both the inverting unit 213 and the inverting unit 219 perform a process of multiplying by -1 only when the input value belongs to the range b. On the other hand, when the input value belongs to the range a, the inverting unit 211 and the inverting unit 219 perform a process of multiplying by 1. As a result, the correction value calculation method is changed at the time when the range to which the input value belongs is switched as in the first embodiment. Thereby, generation | occurrence | production of a pseudo contour can be prevented similarly to 1st Embodiment.
[0055]
[Third Embodiment]
FIG. 5 is a block diagram showing a configuration of an image processing apparatus according to the third embodiment of the present invention. The basic configuration of the image processing apparatus in this embodiment is the same as that shown in FIG. However, in this embodiment, when the input value belongs to a specific range (specifically, when the input value belongs to the range b), the level of the input value is inverted and normalized. Also, the assignment unit 207 performs assignment in consideration of the result of inverting the level of the input value.
[0056]
That is, as processing in the normalization unit 203, when the input value belongs to the range a, the same processing as in FIG. 13 is performed. However, when the input value belongs to the range b, a process of outputting a value of Rin = (input−out2) / (out1−out2) is performed. This is a process of inverting the level of the input value and normalizing it when the input value belongs to the range b.
[0057]
The assigning unit 207 outputs out1 when the input value belongs to the range b and the threshold value processing result is “1”, and the input value belongs to the range b and the threshold value processing result is “0”. At some time, out2 is output.
[0058]
Specifically, referring to FIG. 6, when the input value is included in range a in the image processing apparatus according to the present embodiment (when the input value is 0 to 0.5), the processing performed is as shown in FIG. It is the same. However, when the input value belongs to the range b (when it is 0.5 to 1), the level of the input value is inverted and normalization is performed. Then, by performing threshold processing on the normalized value, when the normalized value is 0 to Mth, dot 2 is output, and when the normalized value is Mth to 1 Will output dot 1. Thereby, similarly to the above-described embodiment, it is possible to prevent the delay of dots and to prevent the generation of a pseudo contour.
[0059]
[Fourth Embodiment]
FIG. 7 is a block diagram showing a configuration of an image processing apparatus according to the fourth embodiment of the present invention. Referring to the figure, an image processing apparatus includes a threshold value processing unit 301 that performs threshold processing of input values, an output evaluation unit 303 that evaluates output based on threshold processing results, and an output evaluation unit 303. A subtractor 305 that subtracts the corrected threshold value Mth from the output of the output, a correction value memory 307 that temporarily stores the output of the subtractor 305, a β multiplier 309 that multiplies the output of the correction value memory 307 by a coefficient β, Based on an initial threshold value generator 311 for generating a threshold value th, a subtracter 313 for subtracting the output of the β multiplier 309 from the initial threshold value, and an output (corrected threshold value Mth) of the subtractor 313. And a threshold value calculation unit 315 for calculating a threshold value.
[0060]
In this embodiment, the correction threshold value Mth is used as a base value for threshold calculation, and the threshold value calculation unit 315 calculates at least two threshold values based on the correction threshold value Mth. Based on the calculated at least two threshold values, the threshold value processing unit 301 performs threshold processing of the input value.
[0061]
Based on the processing result by the threshold processing unit 301 and the correction threshold Mth, the correction threshold Mth used for subsequent pixel processing is corrected.
[0062]
Specifically, the threshold value calculation unit 315 calculates threshold values th1 and th2 by the following formula based on the corrected threshold value Mth and the output values out1 and out2.
[0063]
th1 = out1 × Mth
th2 = out2- (out2-out1) * Mth
The threshold processing unit 301 outputs 0 when input ≦ th1, and outputs out1 when th1 <input ≦ th2. Further, if th2 <input ≦ th3 (here, it is assumed that th3 = out2), out2 is output.
[0064]
The output evaluation unit 303 outputs “1” if the threshold processing result is 0, outputs “0” if the threshold processing result is out1, and outputs “0” if the threshold processing result is out2. “1” is output.
[0065]
In the present embodiment, processing is performed such that when the value of correction threshold value Mth increases, the value of threshold value th1 increases, and conversely, the value of threshold value th2 decreases. As a result, as in the first to third embodiments, even when the input value changes across out1 (= 0.5), it is possible to prevent the dot delay from occurring, and the pseudo contour Occurrence can be prevented.
[0066]
The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration of an image processing apparatus according to a first embodiment of the present invention.
FIG. 2 is a diagram for explaining the effect of the apparatus of FIG. 1;
FIG. 3 is a block diagram illustrating a modification of the image processing apparatus according to the first embodiment.
FIG. 4 is a block diagram illustrating a configuration of an image processing apparatus according to a second embodiment.
FIG. 5 is a block diagram illustrating a configuration of an image processing apparatus according to a third embodiment.
6 is a diagram showing the operation of the apparatus of FIG.
FIG. 7 is a block diagram illustrating a configuration of an image processing apparatus according to a fourth embodiment.
FIG. 8 is a block diagram illustrating a configuration of an image processing apparatus using an error diffusion method.
FIG. 9 is a block diagram illustrating a configuration of an image processing apparatus using a threshold value diffusion method.
FIG. 10 is a diagram for explaining a problem in the prior art.
FIG. 11 is a diagram illustrating an algorithm of a threshold value diffusion method.
FIG. 12 is a diagram for explaining the operation of the apparatus of FIG.
FIG. 13 is a block diagram illustrating another configuration example of the image processing apparatus using the threshold value diffusion method.
14 is a diagram showing the operation of the apparatus of FIG.
FIG. 15 is a diagram for explaining the operation of an image processing apparatus using a conventional threshold value diffusion method.
FIG. 16 is a diagram for explaining a mechanism in which a pseudo contour is generated.
[Explanation of symbols]
101, 113 Inversion unit, 103, 109 Subtractor, 105 Threshold processing unit, 107 Inversion unit, 111 β multiplication unit, 201 Discrimination unit, 203 Normalization unit, 205 Threshold processing unit, 207 Allocation unit, 209 Inversion Processing unit, 211 Subtractor, 213 Inversion unit, 215 Correction value memory, 217 β multiplication unit, 219 Inversion unit, 220 Initial threshold generation unit, 222 Subtractor, 301 Threshold processing unit, 303 Output evaluation unit, 305 Subtractor, 307 correction value memory, 309 β multiplier, 311 initial threshold value generator, 313 subtractor, 315 threshold value calculator.

Claims (3)

An image processing apparatus that converts an input signal indicating a density level of each pixel with a predetermined number of gradations into a signal with a smaller number of gradations than the predetermined number of gradations,
First output means for comparing the input signal with a first threshold value and outputting a signal of the first gradation or the second gradation when the density level of the input signal is within the first range. When,
When the density level of the input signal is within a second range following the first range, the input signal is compared with a second threshold value to obtain a second gradation or third gradation signal. Second output means for outputting;
Switching means for determining and switching for each pixel which one of the first and second output means is used in accordance with the density level of the input signal;
Multiply the difference by a predetermined coefficient based on the difference between the threshold value used for comparison by the first or second output means and the inverted signal of the signal output from the first or second output means. The correction value output to the peripheral pixel is output to the peripheral pixel, and the correction value output to the peripheral pixel in the subsequent pixel processing is subtracted from the initial first or second threshold value, thereby the first or second threshold value. Correction means for correcting the threshold value of
Calculation of the correction value calculated by the correction means at the time of switching from the first output means to the second output means or at the time of switching from the second output means to the first output means. Changing means for changing the method ,
The image processing apparatus , wherein the changing means inverts the sign of the correction value at the time of switching . An image processing apparatus that converts an input signal indicating a density level of each pixel with a predetermined number of gradations into a signal with a smaller number of gradations than the predetermined number of gradations,
Determining means for determining which range the density level of the input signal belongs to;
Normalization means for normalizing the density level of the input signal so that the density level of the input signal is within a predetermined range;
Comparing means for comparing the normalized density level with a threshold value and outputting a comparison result;
An output means for outputting a signal having a smaller number of gradations than the predetermined number of gradations based on a determination result by the determination means and a comparison result by the comparison means;
Based on the difference between the inverted signal of the signal output from the comparison means and the threshold value, a correction value obtained by multiplying the difference by a predetermined coefficient is subtracted from the initial threshold value during subsequent pixel processing. And a correction means for correcting the threshold value by
The normalization means reverses the density level of the input signal when the density level of the input signal is determined to belong to a specific range by the determination means, so that the density level of the input signal is within a predetermined range. An image processing apparatus that performs normalization to be within . An image processing apparatus that converts an input signal indicating a density level of each pixel with a predetermined number of gradations into a signal with a smaller number of gradations than the predetermined number of gradations,
An output means for outputting a value serving as a basis for threshold calculation;
Threshold calculation means for calculating at least two threshold values based on a value serving as a basis for the threshold calculation;
Threshold processing means for performing threshold processing of the input signal based on the calculated at least two threshold values;
Evaluation means for evaluating the output based on the processing result by the threshold processing means;
By subtracting, from the initial threshold value, a correction value obtained by multiplying the difference by a predetermined coefficient during subsequent pixel processing based on the difference between the values that are the basis for the threshold calculation from the evaluation of the evaluation means Correction means for correcting a value serving as a basis for the threshold calculation,
The threshold processing means outputs a first gradation signal when the density level of the input signal is within a first range up to a first threshold, and the density level of the input signal is A signal of a second gradation is output when it is within a second range from the first threshold value to the second threshold value following the first range, and the density level of the input signal is A third gradation signal is output when it is within a third range equal to or greater than the second threshold value following the second range;
The threshold calculation means calculates a value obtained by multiplying a value serving as a basis for the threshold calculation by an output value of the signal of the second gradation as the first threshold, and the third floor A value obtained by subtracting the output value of the second gradation signal from the output value of the tone signal is multiplied by a value serving as a basis for the threshold calculation, and the output value of the third gradation signal is An image processing apparatus that calculates a value obtained by subtracting a multiplied value as the second threshold value .

Images