JP2002158872A - Image processing method, image processor and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enhance the sharpness of a character on a white background, a character on dots and an image on a print sheet while suppressing moire, and to enhance granularity. SOLUTION: An image region separating section 4 distinguishes a character, a dot and others in an image signal read in from a scanner section 1. An edge amount calculating section 3 calculates the extent of edge of an input image as an edge amount and delivers a corresponding edge amount to an image type subjected to image region separation. An edge emphasizing section 2 emphasizes the edge depending on the edge amount of each image type and delivers an emphasized image to a printer section 5.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention records an image processing method, an image processing apparatus, and an image processing program for calculating a plurality of feature values having different characteristics from a digital image signal and performing adaptive processing according to the feature values. This is a technique suitable for a recording medium, such as a printer, a digital copying machine, and a facsimile.

[0002]

2. Description of the Related Art An image processing apparatus that reads an original with a scanner such as a digital copying machine and outputs the original with a printer can handle three types of originals: a character original, a silver halide photograph (continuous tone) original, and a halftone print original. Broadly classified into types. Also, a document in which these three types of documents are mixed in the same document (hereinafter, mixed document) is handled in the same manner as the above-described document.

[0003] Since mixed originals have different required image quality evaluations depending on the types of mixed originals, it is extremely difficult to satisfy the image quality of all types of originals if uniquely similar image processing is performed. difficult.

[0004] This will be described taking filter processing as an example. In a digital copying machine, edge enhancement processing is performed in order to cope with MTF deterioration due to a scanner, and smoothing processing is performed in order to remove moire of halftone dots. Is performed. Also, if the edge emphasis is emphasized with emphasis on the character image quality, moiré may occur at the halftone dot portion, and if the smoothing processing is enhanced to suppress the moire at the halftone dot portion, the blur of the character and the silver halide photograph may be reduced. There is a problem that the sharpness is insufficient.

[0005] Various methods have been proposed for solving such problems. For example, in an apparatus described in Japanese Patent Publication No. 6-5885, an edge amount is calculated based on local information of a document in which characters and pictures are mixed, and a response is made to a character edge without responding to a halftone dot. The output of the smoothing filter and the output of the edge enhancement filter are mixed by the edge detection output using the edge detection filter. As a result, a smoothing filter is applied to the halftone dots to suppress moire.

In the above technique, the filter size for edge detection is set lower than the harmonic frequency of the halftone dot image so that a high edge amount is not calculated for the halftone dot image. In this case, if the same processing is to be performed on a halftone dot having a low screen ruling, an edge detection filter having a very large size is required. (That is, the edge amount of a small-sized character is reduced and the edge is not emphasized).

As another example, the apparatus described in Japanese Patent Application Laid-Open No. 8-181865 is provided with means for detecting a black character area and a halftone dot area, respectively, and applies a smoothing filter in accordance with the detection result of each area. The presence / absence and the degree of edge enhancement are switched.

In general, in image area separation processing for identifying an image based on the characteristics of a local image, erroneous determination is inevitable, and the processing is switched in a binary manner depending on the image area as described in the above publication. In such a case, for example, when a character on a halftone dot (determined as a normal halftone dot) is erroneously determined to be a character part, the influence of the application of the smoothing process on the output image is extremely large due to the difference in filter characteristics. As a result, an output image having a sense of incongruity in which only the portion determined to be a character is raised, resulting in remarkable image quality deterioration.

The problem of the binary image area separation processing as described above can be solved by reducing the difference in filter strength between the area determined as a character and the area determined as a picture.

[0010]

However, in a digital copying machine or the like, usually, a strong smoothing process is performed on a halftone dot region including a character on a halftone dot to perform edge emphasis. Since only weak edge enhancement is performed, characters on halftone dots tend to be blurred as compared with characters on a white background.
Therefore, when strong edge enhancement is performed on a halftone dot region in order to eliminate blurring of characters on the halftone dot, a problem of occurrence of moire occurs. Thus, it is very difficult to achieve both suppression of moiré and elimination of blurring of characters on halftone dots.

In addition, when a low-density bold character or the like is determined as a character, excessive edge emphasis is applied, so that the density of only the character edge portion increases, and this is recognized as a border, which causes a decrease in image quality. I was

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems. An object of the present invention is to calculate a plurality of feature values having different characteristics from an image, and perform adaptive processing using the feature value corresponding to the image type. By doing so, the image processing method, the image processing apparatus, and the image processing program which improved the sharpness of the characters on the white background, the characters on the halftone dots and the photographic paper image, suppressed the moiré, and improved the graininess were recorded. It is to provide a recording medium.

It is another object of the present invention to provide an image processing method, an image processing apparatus, and a recording medium on which an image processing program is recorded, in which the sharpness of characters on halftone dots is improved and moire is suppressed.

Still another object of the present invention is to provide an image processing method, an image processing apparatus, and a recording medium on which an image processing program is recorded, in which image quality deterioration due to erroneous identification in image area separation processing is suppressed.

Still another object of the present invention is to provide an image processing method, an image processing apparatus, and a recording medium on which an image processing program is recorded, in which image quality deterioration due to bordering is suppressed.

[0016]

According to the present invention, the input image data is classified into three types of image areas, namely, a character image, a halftone image, and other (continuous tone image). The edge amount is calculated from the image data using a spatial filter having different characteristics. The first spatial filter is configured by a second-order differential filter having a high-frequency spatial frequency response, and the second spatial filter is configured by a first-order differential filter having a high-frequency spatial frequency response. The edge amount for the continuous tone image is calculated by the first spatial filter,
The edge amount for the halftone image is calculated by the second spatial filter. The edge amount for the character image is calculated by mixing the edge amount for the continuous tone image and the edge amount for the halftone image. An edge amount of any image type is selected according to the image area identification result, edge enhancement processing is performed according to the edge amount, and the result is output to a printer or the like.

In the first spatial filter of the present invention, since the edge amount of the high frequency component is detected and the detection of the edge amount of the low frequency component is suppressed, the edge of the continuous tone image having the high frequency component is emphasized. Thus, sharpness is improved, and noise of low frequency components is reduced, so that graininess is improved.

In the second spatial filter of the present invention, since the edge amount of a relatively low frequency component is detected, it is possible to perform edge emphasis without responding to the frequency of a halftone dot component. Image quality degradation can be suppressed.

In the present invention, since the edge amount for the continuous tone image and the edge amount for the halftone image are mixed as the edge amount for the character image, it is optimal for a character image having a wide frequency component. Edge enhancement can be performed, and the sharpness of a character image is improved.

[0020]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be specifically described below with reference to the drawings. In order to avoid repetition of the description, the same reference numerals are used for the same or corresponding parts in a plurality of drawings in the attached drawings. (Embodiment 1) FIG. 1 is a diagram showing a configuration of an image processing apparatus according to Embodiment 1 of the present invention. The image processing apparatus according to the present embodiment includes a scanner unit 1 including an image sensor such as a CCD, an edge enhancement processing unit 2 that performs edge enhancement processing as adaptive processing according to an edge amount, and a digital image read by the scanner unit. An edge amount calculation unit 3 for calculating an edge amount from data; an image area separation processing unit 4 for identifying an image type of digital image data; and a printer unit 5 as an image output unit
It is composed of

The scanner section 1 optically reads an original containing a mixture of characters, halftone images, photographic images, etc.
To 255), and then output.
The image data is input to the edge enhancement processing unit 2, the edge amount calculation unit 3, and the image area separation processing unit 4, respectively.

The image area separation processing unit 4 converts the input image into a character image, a halftone image (characters on halftone dots are identified as halftone dots), and other (continuous tone images such as white background and photographic paper). Image areas are classified into three types. In the image area separation processing unit 4 of the present embodiment, a known image area separation technique is used. Conventionally, various image area separation methods have been proposed, but here, for example, IEICE Transactions '92 / 1 Vol. J75-D-
II No. 1 pp. Image area separation method described in 39-47 "Image area separation method for mixed image of characters / pictures (dots, photographs)". The image area separation technique used in the present embodiment is not limited to the above-described technique, and may be any technique that basically identifies a character area and a halftone area on a white background from an image.

In the edge amount calculation unit 3, as described later,
The degree of the edge of the input image data is determined by the edge amount (Edg
e) is calculated. The calculated edge amount is selected based on the identification result of the image area separation processing unit 4. In the edge enhancement processing unit 2, Edge output from the edge amount calculation unit 3 is output.
The image data input from the scanner unit 1 is adaptively subjected to edge enhancement according to the signal, and output to the printer unit 5. The printer unit 5 prints the input image data on a recording medium (not shown). The above-described image area separation processing, edge enhancement processing, and edge amount calculation processing execute processing for each pixel of input image data. In each embodiment described later, the process is similarly performed for each pixel of the input image data.

FIG. 2 is a diagram showing a configuration of the edge enhancement processing unit 2 according to the present embodiment. The edge data of the image data input to the edge enhancement processing unit 2 is extracted by a Laplacian filter 11 and a signed output LAP signal 9 bits is obtained.
(255-256).

FIG. 3 shows an example of a Laplacian filter. The Laplacian filter extracts the edge component by multiplying the target pixel by four and calculating the difference from the four neighboring pixels. The Laplacian filter is not limited to the one shown in FIG. 3, and the optimum filter shape for each system may be selected by experiments or the like in consideration of the characteristics of the scanner unit 1 and the printer unit 5.

The output signal of the Laplacian filter 11 (L
AP) is multiplied by the Edge signal (here, 8 bits) output from the edge amount calculation unit 3 by the multiplier 12 and then normalized to 9 bits (that is, the Edge signal is set to a value of 1 or less). , The multiplied value that is again a signed 9-bit signal, and the target pixel (8 bits) of the input image data.
Is added by the adder 13 to emphasize the edge of the pixel of interest, and output image data (8bi
t). When the output image data has a value exceeding 255, it is replaced with 255, and when the output image data has a value of 0 or less, it is replaced with 0.

FIG. 4 shows an edge amount calculator 3 according to this embodiment.
FIG. 3 is a diagram showing the configuration of FIG. The input image data includes an edge amount calculation filter 21 (1) and an edge amount calculation filter 24.
Each is input to (2).

The edge amount calculation filter 21 (1) has a 5 ×
Pixel block size of 5 (the target pixel is the center pixel)
The edge amount is calculated using a second derivative filter, and the edge amount calculation filter 24 (2) calculates the edge amount using a first derivative filter for a 5 × 5 pixel block size.

FIG. 5 shows an edge amount calculation filter 21.
(1) shows an example of the edge amount calculation filter 24 (2).
FIG. 5A shows second-order differential filters (f1) and (f2).
(B) is a first-order differential filter (f1), (f2)
Is shown.

FIG. 6 shows an edge amount calculation filter 21 (1).
And the spatial frequency characteristics of the edge amount calculation filter 24 (2). FIG. 6 shows that the edge amount calculation filter 21 (1) has a high response in a relatively high frequency region, and the edge amount calculation filter 24 (2) has a high response in a relatively low frequency region. This diagram is a diagram in a case where the present embodiment is applied to input image data of 600 dpi.

The edge amount calculation filter 21 (1) is configured as shown in FIG.
(A) is composed of two types of secondary differential filters f1 and f2, and performs a mask operation (convolution) between the input image data and the filter f1 for detecting the edge in the horizontal (x) direction. y) Perform a mask operation (convolution) with the filter f2 for edge detection in the direction, and sign 11 bits (−1024 to 102), respectively.
3) Output a signal.

The absolute value processing section 22 performs an absolute value process of a mask operation in the x direction and an absolute value process of a mask operation in the y direction, and outputs an unsigned 10-bit signal (0 to 102).
3) is output. The maximum value selection unit 23 selects the maximum value from the two outputs from the absolute value processing unit 22, normalizes the maximum value to 1, and outputs (8-bit signal).

The edge amount calculation filter 24 (2) is configured as shown in FIG.
(B) is composed of two types of primary differential filters f1 and f2, and includes input image data and an x-direction filter f1.
And a mask operation (convolution) between the input image data and the filter f2 in the y-direction are performed, and signed 11-bit (-1)
024 to 1023) output signals.

Similarly, the absolute value processing unit 25 performs an absolute value process of the mask operation in the x direction and an absolute value process of the mask operation in the y direction, and outputs an unsigned 10-bit signal (0
To 1023) are output. The maximum value selector 26 selects the maximum value from the two outputs from the absolute value processor 25,
The maximum value is normalized to 1 and output (8-bit signal).

Note that the filter shape (coefficient array and the like) for calculating the edge amount is not limited to the above example, and its size may be selected according to the characteristics of the image processing apparatus.

The edge amount calculated based on the output value of the edge amount calculation filter 21 (1) is output (8-bit signal) as an edge amount (Edge_kai) for a continuous tone image, and the edge amount calculation filter 24 (2) ) Are calculated based on the edge amount (Edge) for the halftone dot image.
_Ami) (8-bit signal).

At the same time, Edge_kai and Edge_am
The output value of i is input to the calculator 27, and the following calculation is performed.

Edge_moji = (Edge_kai)
+ Edge_ami) / 2 Here, Edge_moji: edge amount for a character image Edge_kai: edge amount for a continuous tone image Edge_ami: edge amount for a halftone image That is, in the arithmetic unit 27, the edge amount for a continuous tone image (Edge_kai) Edge amount for dot image (Edg
By mixing e_ami), an edge amount Edge_moji (8-bit signal) for a character image is output.

The three types of edge amounts are determined by the selector 28
And is selected by the image area identification signal output from the image area separation processing section 4. When the image area identification signal represents a character, the edge amount (Edge_moji) for a character image is selected, and when the image region identification signal represents a halftone dot, the edge amount (E) for a halftone image is selected.
dge_ami), and in other cases such as silver halide photography, the edge amount (Edge_kai) for a continuous tone image is selected and output as an Edge signal (8 bits). Then, the Edge signal is input to the multiplier 12 of the edge enhancement processing unit 4 as described above.

The first embodiment of the present invention operates as described above, but as described above, the edge amount calculation filter 21
In (1), the response is relatively high in a high frequency region. In the present embodiment, the edge amount calculation filter 21 having such characteristics
Since the edge amount of a continuous tone image or the like is detected using (1), it is possible to perform high edge enhancement even in a region having a high frequency component, and the sharpness is improved.
At the same time, by suppressing the detection of low-frequency components relatively low, the effect of noise of low-frequency components of an image (particularly, noise components of about 1 [c / mm], which are easily visually noticeable to human eyes) is suppressed. Therefore, the effect that the granularity of the output image is improved can be obtained.

In this embodiment, the edge amount calculation filter 24 (2) having a characteristic of high response in a relatively low frequency region.
Is used to detect the edge amount of the halftone image, so that it is possible to perform edge enhancement without responding to the frequency of the halftone component of the input image data, and to suppress the image quality deterioration such as moire. be able to.

Further, in this embodiment, the edge amount of the character image is detected by mixing the detection result of the edge amount of the high-frequency component and the detection result of the edge amount of the low-frequency component. Optimum edge emphasis can be performed on a character image, and the sharpness of the character image can be improved. Also, optimal edge emphasis processing can be performed on characters of various sizes (especially small letters).

As described above, according to this embodiment, the moire at the halftone dot portion is suppressed, the sharpness of the continuous tone image and the character image is improved, and the graininess is improved.

The present invention is not limited to the embodiments described above, and various modifications are possible. For example, a method for calculating an edge amount,
Various forms can be adopted for the shape of the filter, its value, and the method of selecting the edge amount, and the present invention is not limited to these methods. Further, although the present embodiment has been described by taking a digital copying machine as an example, the present embodiment can be applied to a printer having no scanner unit and the like, and can be easily realized by software processing on a computer. Further, although the present embodiment does not include the smoothing process,
Since a copying machine or the like generally performs a smoothing process, the present invention is not limited by the presence or absence of the smoothing process. Further, the present embodiment has been described by taking the edge emphasis as an example. However, for example, a case where the smoothing process is controlled in accordance with the edge amount can be realized by a similar method.

(Embodiment 2) FIG. 7 is a diagram showing a configuration of an image processing apparatus according to Embodiment 2 of the present invention. Scanner unit 1
, Optically reads a document in which characters, halftone images, photographic images, and the like are mixed, photoelectrically converts the data into 8-bit (0 to 255) digital image data, and outputs the digital image data.

The image data is sent to the edge enhancement processing unit 31.
(1), input to the edge enhancement processing unit 32 (2), the edge amount calculation unit 33 (1), the edge amount calculation unit 34 (2), and the frequency component detection unit 35, respectively. Edge enhancement processing unit 31
(1) The image data subjected to the edge enhancement processing by the edge enhancement processing unit 32 (2) is mixed by the mixer 36 based on the output signal of the frequency component detection unit 35, and is output to the printer unit 5. The printer unit 5 prints the input image data on a recording medium (not shown).

Edge enhancement processing unit 31 according to the second embodiment
The configurations of (1) and the edge enhancement processing unit 32 (2) are the same as those in FIG. 2 of the first embodiment. The Laplacian filter 11 used in the second embodiment uses the Laplacian filter (hereinafter, filter 1) shown in FIG. 8 in the edge enhancement processing unit 31 (1), and uses the Laplacian filter shown in FIG.
Is used (hereinafter, filter 2).

FIG. 9 shows the spatial frequency characteristics of the filters 1 and 2 described above. The filter 1 has a strong response to high frequencies, and the filter 2 has a strong response to relatively low frequencies.

The edge amount calculator 33 (1) calculates the edge amount for the edge enhancement processor 31 (1), and the edge amount calculator 34 (2) calculates the edge amount for the edge enhancement processor 32 (2). Is calculated.

The edge amount calculation unit 33 (1) comprises an edge amount calculation filter 21 (1), an absolute value processing unit 22, and a maximum value selection unit 23, as in FIG. 4 of the first embodiment. The filter 21 (1) calculates the edge amount (Edge_high) of the high-frequency component using the edge amount calculation filter 1 (f1, f2) shown in FIG.

The edge amount calculating section 34 (2) comprises an edge amount calculating filter 24 (2), an absolute value processing section 25 and a maximum value selecting section 26, as in FIG. 4 of the first embodiment. The filter 24 (2) calculates the edge amount (Edge_low) of the low frequency component using the edge amount calculation filter 2 (f1, f2) shown in FIG. 5B. The calculation of the edge amount by the edge amount calculation units 33 (1) and 34 (2) is the same as the processing up to the maximum value selection units 23 and 26 in FIG.

FIG. 10 is a diagram showing the configuration of the frequency component detection unit 35. The frequency component detection unit of the present embodiment simply detects the frequency component of the input image data using the edge amount calculation filter. The frequency component detector 35 (FIG. 1)
0) are the same as those shown in FIG.
The edge amount calculation filters 21 (1) and 24 (2) are also configured with the same filters as those in FIG. 4 of the first embodiment.

The edge amount calculated based on the output value of the edge amount calculation filter 21 (1) is output as the high frequency component F_high of the input image data, and is calculated based on the output value of the edge amount calculation filter 24 (2). Are output as low-frequency components F_low of the input image data, and their output values are input to the calculator 41.

The arithmetic unit 41 performs the following operation, and outputs a signal (8 bits each) of the filter 1 mixing coefficient F1 and the filter 2 mixing coefficient F2.

F1 = F_high / (F_high +
F2 = F_low / (F_high + F_low) where F1: filter 1 mixing coefficient F2: filter 2 mixing coefficient F_high: high-frequency component signal F_low: low-frequency component signal F1 and F2 detected by the frequency component detection unit 35 The mixing coefficients are Fil1, Fi of the edge enhancement processing results output from the edge enhancement processing units 31 (1), 32 (2), respectively.
The signal is input to the mixer 36 together with the l2 signal.

The mixer 36 performs the following operation and outputs an edge enhancement output Fil. Fil = (Fil1 × F1 / 256 + Fil2 × F
2/256) / 2 As described above, in the present embodiment, the output of the filter 1 (FIG. 8) that mainly emphasizes high-frequency components in the edge enhancement processing unit 31 and the enhancement of mainly low-frequency components in the edge enhancement processing unit 32 are described. The output of filter 2 (FIG. 3) that performs
Since the high and low frequency components are mixed using the detected coefficients F1 and F2, it is possible to perform edge enhancement processing adapted to the frequency characteristics of the input image, and to improve the image quality of characters, photographic paper, and the like. .

Since the frequency component detecting section of this embodiment is composed of a primary differential filter and a secondary differential filter as in the case of calculating the edge amount, the frequency component can be detected relatively easily. In addition to this, if a filter similar to the edge amount calculation filter is used, the filter can be shared, and the amount of hardware can be reduced. Further, it is possible to easily improve the resolution of the frequency by increasing the types of filters used for detecting the frequency component. Note that the method of detecting the frequency component is not limited to the above-described embodiment, and may be realized by a method using, for example, FFT (Fast Fourier Transform). Also in this embodiment, Embodiment 1
In the same manner as described above, it can be easily realized by software processing on a computer, and can be realized by a similar method when controlling the smoothing processing.

In the second embodiment, the output after edge emphasis is mixed with the frequency component signal. However, instead of detecting the frequency component, the thickness of the character is detected. The second embodiment outputs a high frequency component when a thin character is detected, and outputs a high frequency component based on the output.
By calculating the mixing coefficient in the same manner as in the case of and mixing with the mixing coefficient, the second embodiment can be modified so as to perform edge emphasis according to the thickness of the character. As a method for detecting the thickness of a character, for example, a determination method described in Japanese Patent Application Laid-Open No. 7-203198 may be used.

(Embodiment 3) The method of Embodiment 2 described above
Since highly accurate edge enhancement processing can be performed on images such as characters and photographic paper, it is possible to improve the sharpness of those images and improve the image quality.

However, when the method of Embodiment 2 is applied to a halftone image, although the sharpness of the character on the halftone dot can be improved, moire occurs and the graininess is reduced, and the image quality is degraded. May be invited. Therefore, the present embodiment relates to an embodiment which improves the image quality also for a halftone image and is suitable for a halftone image.

FIG. 11 is a diagram showing the configuration of the image processing apparatus according to the third embodiment of the present invention. In the figure, a scanner unit 1,
The operation of the printer unit 5 is the same as in the first and second embodiments.

The image data input from the scanner unit 1 is input to the character / photographic paper filter processing unit 51, the halftone dot filter processing unit 52, and the image area separation processing unit 4, respectively. The character and photographic paper filter processing unit 51 has the same configuration as that of the filter processing unit (the components 31 to 36 in FIG. 7) of the second embodiment.

The image area separation processing section 4 of this embodiment identifies an image area as to whether or not the input image data is a halftone image, and outputs an ami signal. When the ami signal is on (the input image is a halftone image), the selector 53
2 and the ami signal is off (the input image is not a halftone image), the character and photographic paper filter processing unit 5
1 is selected and output to the printer unit 5.

The dot filter processing unit 52 includes an edge enhancement processing unit 31 (1), an edge enhancement processing unit 32 (2), an edge amount calculation unit 33 (1), an edge amount calculation unit 34 (2), a frequency component It comprises a detection unit 35 and a mixer 36.
Here, the configurations and operations of the edge enhancement processing unit 31 (1), the edge enhancement processing unit 32 (2), the edge amount calculation unit 33 (1), the edge amount calculation unit 34 (2), and the mixer 36 are described in the embodiment. 2
And the same filters as in the second embodiment are used.

The structure of the frequency component detection unit 35 is also
As in the second embodiment, the components (21 to 26,
41), and is the same as the case of the second embodiment except for the operation (determination processing) of the arithmetic unit 41.

Hereinafter, the determination process performed by the arithmetic unit 41 of the third embodiment will be described in C language. if (F_high> TH_moji1 &&F_low> TH_moji2)} / * Half-tone character * / F1 = F_high / (F_high_F_low_F_low / F_low / F_low / F_low / (F_low / F_low)) F1 = 0; F2 = 255;｝ else ｛/ * low frequency halftone dot * / F1 = 255; F2 = 0;｝ where F1: filter 1 mixing coefficient F2: filter 2 mixing coefficient F_high: High frequency component signal F_low: low frequency component signal TH_moji1, TH_moji2, TH_low_screen: a predetermined constant.

The computing unit 41 of this embodiment uses the low-frequency component signal F_low and the high-frequency component signal F_high to divide the halftone image area into three of the halftone character, halftone picture, and low frequency halftone dot. Processing is performed by the user.

F_high and F_low are preset character discrimination thresholds (TH_moji1, TH_moji1,
If it is larger than _moji2), it is determined to be a halftone character, and the same edge processing as in the case of a character or photographic paper image is performed.

If F_low is lower than the low frequency ruling halftone discriminating threshold value TH_low_screen (if the low frequency component is very small) other than the characters on the halftone dots, it is determined that the image is a pattern image based on high ruling halftone dots. Edge enhancement of only low frequency components is performed. That is, in the case of a halftone image, the halftone frequency becomes the highest frequency, so that the edge enhancement of the high frequency component that causes the deterioration of the image quality is not performed.

In cases other than the above (F_l for characters other than halftone
ow is greater than or equal to TH_low_screen), it is determined to be a low-frequency halftone dot, and only high-frequency components are emphasized. This is because moiré is likely to occur when low-frequency components are emphasized at a low screen frequency halftone dot, and this is to be suppressed.

As described above, according to the present embodiment, the second embodiment
In addition to the effects described above, the sharpness of characters on halftone dots can be improved with respect to halftone images, and image quality deterioration such as generation of moiré and reduction in graininess can be prevented, thereby enabling high-quality image reproduction.

In the image area separation processing, for example, even if erroneous separation (identification error) occurs in a character image,
According to the method of the present embodiment, the difference between the processing of the character area and the processing of the non-character area due to erroneous separation is reduced, and it is possible to suppress the deterioration of the image quality at the place where the separation results are mixed. Become. Note that this embodiment can also be easily realized by software processing on a computer as in each of the above-described embodiments, and can also be applied to a smoothing process.

(Embodiment 4) Embodiments 1 to 3 described above are examples in which edge enhancement processing (spatial filter processing) is applied as adaptive processing. This embodiment is applicable to black generation and under color removal as adaptive processing. An example in which the invention is applied will be described.

FIG. 12 is a diagram showing the configuration of an image processing apparatus to which this embodiment is applied. The document is optically read by the scanner unit 1 and photoelectrically converted into an 8-bit (0 to 255) digital image signal, and then a known shading correction is performed. The image signal output from the scanner unit 1 is input to the scanner γ correction processing unit 61, and rgb (red, green, and blue) read by the scanner unit 1.
e) The signal is converted into an RGB signal of a density signal using a look-up table (LUT) or the like. The RGB signals converted into the density signals are input to the filter processing unit 62 and subjected to the spatial filter processing as described in the above-described embodiment. The RGB signals after the filter processing are output to the color correction processing unit 63.
CMY (Cyan, Magenta, Yellow)
Converted to a signal. Although various methods can be considered for the color correction processing, it is assumed here that an operation such as Expression (1) is performed.

C = α11 × R + α12 × G + α13 × B + β1 M = α21 × R + α22 × G + α23 × B + β2Y = α31 × R + α32 × G + α33 × B + β3 Equation ( 1) Here, α11 to α33 and β1 to β3 are predetermined color correction coefficients, and the output CMYK is also 8 bits (0
To 255).

Next, the BG / UCR processing section 64 generates (BG) a K signal as a black component and performs undercolor removal (UCR) from the CMY signal. K signal generation and CM
The undercolor removal from the Y signal is performed as in equation (2). K = Min (C, M, Y) × β4 C ′ = C−K × β5 M ′ = M−K × β5 Y ′ = Y−K × β5 Equation (2) where Min (C, M, Y) ) Represents the smallest of the CMY signals. Β4 (<1) and β5 (<1) are 8-bit signals with predetermined coefficients.

Next, the printer γ correction processing section 65
A gamma correction process is performed on each of the inputted CMYK density signals in order to correspond to the gamma characteristics of the printer. γ
The correction is performed using the LUT.

The CMYK data after the printer γ correction processing is output to the printer section 5 after the halftone processing section 66 performs pseudo halftone processing by known dither processing and error diffusion processing. A series of image forming processes is performed as described above, and an image is printed on paper or the like.

FIG. 13 is a diagram showing a detailed configuration of the BG / UCR processing section 64 according to this embodiment. Basically BG /
The UCR process performs the operation of the above-described equation (2). Coefficient β
4 and β5 are controlled in multiple stages according to the edge amount as described later. The CMY signal output from the color correction processing unit 63 is input to the BG processing unit 71, and a gray component (K) signal is generated by the calculation of the following equation (3).

K = Min (C, M, Y) Expression (3) In parallel with this processing, the determination unit 72 generates a Judge signal based on the image signal M by Expression (4), and calculates the edge amount calculation unit. Enter 73.

[0081] Here, TH1: a predetermined constant.

FIG. 14 is a diagram showing a configuration of the edge amount calculating section 73 according to the present embodiment. The edge amount calculation unit 73 is the same as the edge amount calculation unit 33 of the second embodiment except for the selector 81.
(1) The configuration is the same as the configuration of the edge amount calculation unit 34 (2). The calculation of the edge amount is basically the same as in the second embodiment, and the edge amount Edge_high,
The edge amount Edge_low of the low frequency component is calculated.

The selector 81 has an edge amount Edge_high of the high-frequency component and an edge amount Edge__ of the low-frequency component.
low is input, and one of them is selected by the aforementioned Judge signal. That is, when the Judge signal is 1
In the case of (high density), the selector 81 selects Edge_low, and when the Judge signal is 0 (low density), the selector 81 selects Edge_low.
e_high is selected and output to the coefficient calculation unit 74 as an Edge signal.

The Edge signal is input to a coefficient calculator 74, and the coefficient calculator 74 calculates β4 and β5 as follows. if (Edge> TH2) ｛β4 = β4min + (1.0-β4min) × Edge / 256; β5 = β5min + (1.0-β5min) × Edge / 256; {else} β4 = β4min; β5 = β5min; (5) Here, β4min, β5min, TH2: predetermined constants.

When the Edge signal has a predetermined threshold value TH2
In the following cases, β4 and β5 are the preset minimum values β4min and β5min, and the low black ratio and low UCR
Image. In the case of a high edge portion equal to or greater than the threshold value TH2, E
The output image has a high black ratio and a high UCR according to the dge signal.

Β4 output from coefficient calculating section 74 is B
The K signal output from the G processing unit 71 is multiplied by the multiplier 75 and output as a black (K ′) signal. Further, the K ′ signal is multiplied by β5 in the multiplier 76 and is subtracted from the CMY signal in the subtractor 77. U which is an output signal from the subtractor 77
The C ′, M ′, and Y ′ signals after the CR are input to the printer γ correction processing unit 65 in the next stage.

As described above, in the present embodiment, an image of high black is used in an edge portion having a high edge amount, and an image of low black is used in a flat portion having a low edge amount. The black amount and UCR amount are controlled in accordance with the edge amount. However, if the edge portion is excessively high in low-concentration, low-contrast, bold characters, black is noticeable only at the edges and the so-called bordering may deteriorate. However, by using the edge amount calculated from the high-frequency component at the edge portion of the low-density portion as in the present embodiment, the edge amount itself becomes sharper than when using the low-frequency component, and the edge amount becomes higher. It is possible to make the portion to be black thin. Therefore, it is possible to reduce the appearance of the border on the actually output image.

In this embodiment, the edge amount is switched using the image signal M. However, a signal such as lightness may be calculated and the signal may be used. Instead of binaryly switching the edge amount between a high-frequency component and a low-frequency component, the second embodiment
As described above, the edge amounts may be mixed according to the image signal M. In the above-described embodiment, the example of the black generation (BG) and under color removal (UCR) processing units has been described. However, the present invention provides a gamma correction process (for example, a gamma characteristic is moderated in a low density edge portion) or an intermediate process. Application to tone processing and the like is also possible.

As described above, the present invention may be implemented by dedicated hardware, but may be implemented by software using a general-purpose computer system. When implemented by software, a program for realizing the image processing function (processing such as edge amount calculation, edge enhancement processing, and image area separation) and processing procedure of the present invention is recorded on a recording medium or the like. The image processing function of the present invention is implemented when the program is read into the computer system from, for example, and executed by the CPU. The image data is, for example, document image data read from a scanner or the like, image data prepared in a hard disk or the like in advance, or image data captured via a network. The processing result is
Output to printer or write to hard disk. Alternatively, it is output to an external device (such as a printer) via a network.

[0090]

As described above, according to the present invention, the following effects can be obtained. (1) Since adaptive processing is performed using an edge amount of a frequency component corresponding to an image type, a character on a white background, a character on a halftone dot,
While improving the sharpness of continuous tone images such as photographic paper,
Moire is suppressed, graininess is improved, and high-quality image reproduction can be performed. (2) Since the output after the edge enhancement processing is mixed according to the frequency characteristics of the image, it is possible to achieve both sharpness of halftone characters and suppression of moiré. (3) In the filter processing for characters and the filter processing for halftone dots, the outputs after the adaptive processing are mixed, so that it is possible to suppress image quality deterioration due to erroneous identification in the image area separation processing. . (4) Since the black amount and the UCR amount are controlled using the edge amount of the high-frequency component in the low-density edge portion, it is possible to suppress the image quality deterioration due to the bordering that occurs in the low-density bold characters and the like. .

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a configuration of an image processing apparatus according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a configuration of an edge enhancement processing unit according to the first embodiment.

FIG. 3 is a diagram illustrating an example of a Laplacian filter.

FIG. 4 is a diagram illustrating a configuration of an edge amount calculation unit according to the first embodiment.

FIGS. 5A and 5B are edge amount calculation filters 1;
FIG. 2 is a diagram showing an example of the second example.

FIG. 6 is a diagram illustrating spatial frequency characteristics of edge amount calculation filters 1 and 2;

FIG. 7 is a diagram illustrating a configuration of an image processing apparatus according to a second embodiment of the present invention.

FIG. 8 is a diagram illustrating an example of a Laplacian filter used in an edge enhancement processing unit according to a second embodiment.

FIG. 9 is a diagram illustrating spatial frequency characteristics of Laplacian filters 1 and 2 used in the edge enhancement processing unit according to the second embodiment.

FIG. 10 is a diagram illustrating a configuration of a frequency component detection unit according to a second embodiment.

FIG. 11 is a diagram illustrating a configuration of an image processing apparatus according to a third embodiment of the present invention.

FIG. 12 is a diagram illustrating a configuration of an image processing apparatus to which a fourth embodiment of the present invention is applied.

FIG. 13 is a diagram illustrating a configuration of a BG / UCR processing unit according to a third embodiment.

FIG. 14 is a diagram illustrating a configuration of an edge amount calculation unit according to a third embodiment.

[Explanation of symbols]

 Reference Signs List 1 scanner unit 2, 31, 32 edge enhancement processing unit 3, 33, 34, 37 edge amount calculation unit 4 image area separation processing unit 5 printer unit 11 Laplacian filter 12, 75, 76 multiplier 13 adder 21, 24 edge amount Calculation filter 22, 25 Absolute value processing unit 23, 26 Maximum value selection unit 27, 41 Computing unit 28, 53, 81 Selector 35 Frequency component detection unit 36 Mixer 51 Character filter, photographic paper filter processing unit 52 Halftone filter processing Unit 61 scanner γ correction unit 62 filter processing unit 63 color correction unit 64 BG / UCR 65 printer γ correction unit 66 halftone processing unit 71 BG processing unit 72 determination unit 74 coefficient calculation unit 77 subtractor

 ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference) PP28 PP33 PP37 PP38 PP47 PP48 PP49 PP65 PQ08 PQ18 PQ23 RR11 5L096 BA07 FA06 FA23 FA43 FA44 FA45 FA81 GA02 GA03 GA05 GA06

Claims (58)

[Claims] 1. An image processing method for performing a predetermined adaptive process on a predetermined image and outputting the same, wherein when performing the adaptive process, a plurality of feature amounts having different characteristics calculated from the image are calculated. A feature amount corresponding to the type of the image is selected from the list, and an adaptive process is performed according to the selected feature amount. 2. The image processing method according to claim 1, wherein when the type of the image is a continuous tone image, an adaptive process is performed by selecting a feature amount of a high-frequency component calculated from the image. . 3. When the type of the image is a halftone image,
2. The image processing method according to claim 1, wherein the adaptive processing is performed by selecting a characteristic amount of a low frequency component calculated from the image. 4. When the type of the image is a character image,
2. The image processing method according to claim 1, wherein a characteristic amount of a wide frequency component including a characteristic amount of a low frequency component and a characteristic amount of a high frequency component calculated from the image is selected and the adaptive processing is performed. 5. The image processing method according to claim 4, wherein the characteristic amount of the wide frequency component is a characteristic amount obtained by mixing the characteristic amount of the low frequency component and the characteristic amount of the high frequency component. 6. An image processing method for performing predetermined adaptive processing on a predetermined image and outputting the adaptive image, wherein each of the adaptive processing is performed on the image using a plurality of feature amounts having different characteristics calculated from the image. An image processing method comprising: performing a process; mixing and outputting a plurality of outputs after each adaptive process according to a frequency characteristic of the image; 7. The frequency characteristic of the image is a ratio of a high frequency component and a low frequency component in the image, respectively.
The first output after the adaptive processing is weighted by the proportion of the high-frequency component occupying the image, and the second output after the adaptive processing is weighted by the proportion of the low-frequency component occupying the image to be mixed and output. 7. The image processing method according to claim 6, wherein: 8. A result of performing a process suitable for a first image type on a predetermined image or a result of performing a process suitable for a second image type, based on the type of the image. An image processing method for selecting and outputting based on a result of the identification, wherein a process suitable for the first image type is calculated from the image.
Perform adaptive processing on each of the images using a plurality of characteristic amounts having different characteristics, and output a plurality of outputs after each adaptive processing,
The mixed image is output in accordance with the ratio of the predetermined component in the image, and the process suitable for the second image type is performed on the image using a plurality of feature amounts having different characteristics calculated from the image. An image processing method for performing adaptive processing on each of the images and mixing or selecting a plurality of outputs after each of the adaptive processing in accordance with a predetermined amount of a predetermined component in the image. 9. The first image type is a character or a photographic paper, and the processing suitable for the character and the photographic paper is such that the first output after the adaptive processing is a ratio of a high frequency component occupying the image. 9. The image processing method according to claim 8, wherein the weighted second output after the adaptive processing is mixed and output by weighting with a ratio of a low frequency component occupying the image. 10. The second image type is a character on a halftone dot, and the processing suitable for the character on the halftone dot is such that an amount of a high-frequency component in the image is equal to or more than a predetermined first value; When the amount of the low-frequency component is equal to or more than a predetermined second value, the first output after the adaptive processing is weighted by a ratio of a high-frequency component occupying the image, and the second output after the adaptive processing is 9. The image processing method according to claim 8, wherein the image is mixed and output by weighting with a ratio of a low frequency component in the image. 11. The second image type is a halftone dot pattern,
The process suitable for the halftone dot pattern is characterized in that when the amount of the low frequency component in the image is lower than a predetermined third value, the second output after the adaptive process is selected and output. 9. The image processing method according to claim 8, wherein: 12. The second image type is a low-frequency halftone dot, and the processing suitable for the low-frequency halftone dot is performed when the amount of low-frequency components in the image is equal to or more than a predetermined third value. 9. The image processing method according to claim 8, wherein the first output after the adaptive processing is selected and output. 13. The method according to claim 13, wherein the first output after the adaptive processing is an output after the high-frequency component of the image is adaptively processed in accordance with a feature amount of the high-frequency component calculated from the image. Item 13. The image processing method according to Item 7, 9, 10 or 12. 14. The second output after the adaptive processing is an output after the low-frequency component of the image has been adaptively processed in accordance with the characteristic amount of the low-frequency component calculated from the image. 12. The image processing method according to claim 7, 9, 10, or 11, wherein: 15. An image processing method for performing predetermined adaptive processing on a predetermined image and outputting the result, wherein a density of the image is determined, and the density is calculated from the image based on the determination result.
An image processing method comprising selecting one of a plurality of feature values having different characteristics and performing adaptive processing according to the selected feature value. 16. When the density is equal to or more than a predetermined value, a characteristic amount of a low-frequency component calculated from the image is selected. When the density is less than a predetermined value, a characteristic amount of a high-frequency component calculated from the image is selected. 16. The method according to claim 15, wherein a characteristic amount is selected, and the adaptive process performs different processing according to the selected characteristic amount of the low frequency component or the characteristic amount of the high frequency component.
The image processing method described in the above. 17. The adaptive processing according to claim 16, wherein when the characteristic amount of the selected low frequency component or high frequency component exceeds a predetermined value, the adaptive processing controls the predetermined amount according to the characteristic amount. Image processing method. 18. The image processing method according to claim 16, wherein when the characteristic amount of the selected low frequency component or high frequency component is equal to or less than a predetermined value, the adaptive processing controls the predetermined amount to a minimum value. . 19. The image processing method according to claim 1, wherein the feature amount is an edge amount of the image. 20. The image processing method according to claim 19, wherein the edge amount is calculated by a plurality of filtering processes having different spatial frequency characteristics. 21. The image processing method according to claim 20, wherein said filter processing is a secondary differential filter processing for calculating an edge amount of a high-frequency component. 22. The image processing method according to claim 20, wherein the filter processing is a first-order differential filter processing for calculating an edge amount of a low-frequency component. 23. The image processing method according to claim 1, wherein said adaptive processing is edge enhancement processing. 24. The apparatus according to claim 15, wherein said adaptive processing is black generation and undercolor removal processing.
Or the image processing method according to 18. 25. The image processing method according to claim 17, wherein said predetermined amount is an amount of black and an amount of undercolor removal. 26. A first feature value calculating means for calculating a first feature value from a predetermined image, a second feature value calculating means for calculating a second feature value from the image, and a type of the image. Image type identification means for identification, selection means for selecting the first or second feature value based on the identified image type, and adaptive processing for the image in accordance with the selected feature value. An image processing apparatus comprising: an adaptive processing unit for performing the processing. 27. When the image type identification unit identifies the image type as a continuous tone image, the selection unit selects a high-frequency component feature amount as the first feature amount, and the adaptive processing unit 27. The image processing apparatus according to claim 26, wherein adaptive processing is performed on the image in accordance with a feature amount of the selected high-frequency component. 28. When the image type identification unit identifies the image type as a halftone image, the selection unit selects a low frequency component feature amount as the second feature amount, and the adaptive processing unit The image processing apparatus according to claim 26, wherein an adaptive process is performed on the image in accordance with a feature amount of the selected low-frequency component. 29. The image processing apparatus according to claim 29, further comprising: third feature value calculating means for calculating a third feature value based on the first and second feature values, wherein the selecting means selects the image identified by the image type identifying means. 27. The image processing apparatus according to claim 26, wherein the third feature amount is selected based on a seed, and the adaptive processing unit performs an adaptive process according to the selected third feature amount. 30. When the image type identification unit identifies the image type as a character image, the selection unit sets a low frequency component and a high frequency component as the third feature amount calculated by the third feature amount calculation unit. 30. The image processing apparatus according to claim 29, wherein the adaptive processing unit performs an adaptive process on the image in accordance with the selected characteristic amount including the high-frequency component and the low-frequency component. Image processing device. 31. The method according to claim 29, wherein the third feature value calculating means calculates the third feature value by mixing a feature value of a low-frequency component and a feature value of a high-frequency component. 30. The image processing device according to 30. 32. A first feature value calculating means for calculating a first feature value from a predetermined image, a second feature value calculating means for calculating a second feature value from the image, and A first step of performing an adaptive process according to the calculated first feature amount;
Adaptive processing means for performing adaptive processing on the image in accordance with the calculated second feature amount; specific component detecting means for detecting a specific component of the image; An image processing apparatus comprising: a mixing unit that mixes an output of the first adaptive processing unit and an output of the second adaptive processing unit in accordance with the detected specific component. 33. The specific component detecting means detects a ratio of a high frequency component and a low frequency component in the image as the characteristic component, and the mixing means outputs an output of the first adaptive processing means to the image. 33. The image processing apparatus according to claim 32, wherein weighting is performed by a ratio of a high frequency component occupied, and the output of the second adaptive processing unit is mixed by weighting by a ratio of a low frequency component occupying the image. 34. A first feature value calculating means for calculating a first feature value from a predetermined image; a second feature value calculating means for calculating a second feature value from the image; A first step of performing an adaptive process according to the calculated first feature amount;
Adaptive processing means for performing adaptive processing on the image in accordance with the calculated second feature amount; specific component detecting means for detecting a specific component of the image; Determining means for determining the type of the specific image based on the detected specific component; and control for mixing or selecting the output of the first adaptive processing means and the output of the second adaptive processing means in accordance with the determination result And an image processing apparatus. 35. The first feature value is a feature value of a high-frequency component calculated from the image, and the first adaptive processing means performs processing on the image according to the feature value of the calculated high-frequency component. 35. The image processing apparatus according to claim 32, wherein the high-frequency component is adaptively processed. 36. The second feature value is a feature value of a low-frequency component calculated from the image, and the second adaptive processing means calculates the second feature value in accordance with the calculated feature value of the low-frequency component. 35. The image processing apparatus according to claim 32, wherein the low frequency component of the image is adaptively processed. 37. The detected specific component is an amount of a high-frequency component and an amount of a low-frequency component occupying an image, the determination unit determines that the amount of the high-frequency component is equal to or more than a first predetermined value, and When the amount of the low frequency component is equal to or greater than the second predetermined value, the specific image is determined to be a halftone character image, and the control unit determines that the output of the first adaptive processing unit is a high frequency image occupying the image. 35. The image processing apparatus according to claim 34, wherein weighting is performed by a ratio of components, and the output of the second adaptive processing unit is mixed by weighting by a ratio of a low frequency component occupying the image. 38. The detected specific component is an amount of a high-frequency component and an amount of a low-frequency component in an image, and the determination unit determines that the amount of the low-frequency component is lower than a predetermined third value. 3. The method according to claim 1, wherein the specific image is determined as a halftone picture image, and the control unit sets the output of the first adaptive processing unit to 0 and selects only the output of the second adaptive processing unit. 35. The image processing device according to claim 34. 39. The detected specific component is an amount of a high-frequency component and an amount of a low-frequency component occupying an image, and the determination unit determines that the amount of the low-frequency component is equal to or more than a third predetermined value. The specific image is determined to be a low frequency screen image, and the control unit sets the output of the second adaptive processing unit to 0 and selects only the output of the first adaptive processing unit. The image processing device according to claim 34. 40. A first processing means comprising a means for receiving a predetermined image, performing predetermined processing on a character image and a photographic paper image, and receiving the predetermined image, 35. A second processing means comprising means for performing predetermined processing on an image, an image identification means for identifying whether or not the predetermined image is a halftone image, and Selecting means for selecting the output of the second processing means when it is identified as a dot image, and selecting the output of the first processing means when it is identified as not a halftone image. Image processing apparatus. 41. A density determining means for determining the density of a predetermined image, a first characteristic value calculating means for calculating a first characteristic value of the image, and a second characteristic value calculating means for calculating a second characteristic value of the image. Characteristic amount calculating means, selecting means for selecting the first characteristic amount or the second characteristic amount based on the determination result of the density determining means, and adaptation for performing an adaptive process according to the selected characteristic amount. An image processing apparatus comprising: processing means. 42. The adaptive processing means includes a coefficient calculating means for calculating a predetermined coefficient for controlling a predetermined amount, and when the density determined by the density determining means is less than a predetermined value,
The selection means selects a high-frequency component feature quantity as the first feature quantity, and when the density determined by the density determination means is equal to or higher than a predetermined value, the selection means sets the low-frequency component as the second feature quantity. 42. The image processing apparatus according to claim 41, wherein a characteristic amount of the component is selected, and the coefficient calculating unit calculates a predetermined coefficient in accordance with the selected characteristic amount of the low frequency component or the characteristic amount of the high frequency component. . 43. When the characteristic amount of the selected low-frequency component or the characteristic amount of the high-frequency component exceeds a predetermined value, the coefficient calculating means calculates a coefficient corresponding to the characteristic amount, and calculates the coefficient based on the calculated coefficient. The image processing apparatus according to claim 42, wherein the predetermined amount is controlled. 44. When the characteristic amount of the selected low-frequency component or the characteristic amount of the high-frequency component is equal to or smaller than a predetermined value, the coefficient calculating means calculates a minimum coefficient, and calculates the predetermined coefficient by the calculated minimum coefficient. 43. The image processing apparatus according to claim 42, wherein the quantification is controlled to a minimum value. 45. The image processing apparatus according to claim 26, wherein the first feature value calculating means calculates a feature value of a high frequency component using a secondary differential filter. 46. The image processing apparatus according to claim 26, wherein the second feature value calculating means calculates a feature value of a low frequency component using a first-order differential filter. 47. The apparatus according to claim 26, wherein the characteristic amount is an edge amount of an image.
48. The image processing apparatus according to any one of -47. 48. The image processing apparatus according to claim 26, wherein said adaptive processing is edge enhancement processing. 49. An image processing apparatus according to claim 41, wherein said adaptive processing is black generation and undercolor removal processing. 50. The image processing apparatus according to claim 42, wherein the predetermined amount is a black amount and an undercolor removal amount. 51. A function of inputting a predetermined image, a function of discriminating the image into one of a character image, a halftone image, and a continuous tone image, and calculating an edge amount having a first characteristic from the image. A function for calculating an edge amount of the second characteristic from the image, and selecting one of the first and second edge amounts based on the image type identified in the image area. A computer-readable program storing a program for causing a computer to implement a function, a function of performing edge enhancement processing on the image in accordance with the selected edge amount, and a function of outputting an image subjected to edge enhancement processing. recoding media. 52. A function for calculating a third edge amount from the first edge amount and the second edge amount, and selecting the third edge amount based on the image type identified in the image area. 52. The recording medium according to claim 51, further comprising a function of performing an edge emphasizing process according to the selected third edge amount. 53. A function of inputting a predetermined image, a function of calculating an edge amount of a first characteristic from the image, a function of calculating an edge amount of a second characteristic from the image, A function of performing a first edge enhancement process in accordance with the calculated first edge amount, and a function of performing a second edge enhancement process in accordance with the calculated second edge amount on the image. A function of detecting a frequency component of the image, and a process of mixing the output subjected to the first edge enhancement process and the output subjected to the second edge enhancement process in accordance with the detected frequency component A computer-readable recording medium on which a program for causing a computer to realize a function and a function of outputting a mixed processed image is recorded. 54. A function of inputting a predetermined image, a function of calculating an edge amount of a first characteristic from the image, a function of calculating an edge amount of a second characteristic from the image, A function of performing a first edge enhancement process in accordance with the calculated first edge amount, and a function of performing a second edge enhancement process in accordance with the calculated second edge amount on the image. A function of detecting a frequency component of the image, a function of determining the type of a halftone image based on the detected frequency component, and an output having been subjected to the first edge enhancement processing according to the determination result. And a computer-readable recording medium storing a program for causing a computer to realize a function of mixing or selecting an output subjected to the second edge enhancement processing and a function of outputting a mixed or selected image. 55. A function of inputting a predetermined image, a function of determining the density of the image, a function of calculating an edge amount of a first characteristic from the image, and a function of calculating an edge amount of a second characteristic from the image. A calculating function, a function of selecting the first edge amount or the second edge amount based on the density determination result, and a function of controlling the black amount and the undercolor removal amount according to the selected edge amount. Computer-readable recording medium on which a program for causing a computer to realize the above is recorded. 56. The recording medium according to claim 51, wherein the edge amount of the first characteristic is an edge amount of a high-frequency component. 57. The recording medium according to claim 51, wherein the edge amount of the second characteristic is an edge amount of a low frequency component. 58. The recording medium according to claim 52, wherein the edge amount of the third characteristic is an edge amount obtained by mixing an edge amount of a high-frequency component and an edge amount of a low-frequency component.