JPH01219962A - Method and device for processing image - Google Patents

Abstract

PURPOSE:To convert original image data of a flat effect of a photograph, etc., to aesthetic image data of a picture touch by replacing image data of a partial area with image data derived from the original image data in its area and each prescribed gradation data. CONSTITUTION:A CPU 1 inputs original image data through an image data I/O 26 and stores it in image memories 14-16. An edge detecting part 4 executes an extraction of an edge part, based on the original image data of the image memories 14-16. Based on a result of detection as to whether this edge exists or not, a brush pattern selecting part 8 selects small multivalued brush pattern data stored in advance in a ROM or large multivalued brush pattern data in a brush pattern rotating part 9, and a picture drawing part 10 brings this brush pattern data to picture drawing. An image synthesizing part 12 synthesizes further a texture image against the image data.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an image processing method and an apparatus thereof, and for example, an image processing method and apparatus for converting flat original image data such as a photograph into aesthetic image data with a painterly feel. Regarding the device.

[Prior Art] Conventionally, mosaic processing is an example of creating an aesthetic image by applying image processing technology. This can be done, for example, in the X direction and in the Y direction.
Image mosaic processing is performed using blocks of 5 pixels in each direction, totaling 25 pixels. Specifically, if the original image data at address (x, y) is a (x, y), then mosaic processing is performed. The subsequent image data a' (x, y) is (
1) Determined by the formula.

a' (5m-i, 5n j) = a (5m-3, 5n-3) where, i, j = pixel number (1, 2, 3, 4, 5) m, n block number (1, 2, 3, ...) That is, in the above equation (1), the center pixel data a (5m-3, 5n
-3) is the representative value, and all pixel data a' (5m-i, 5n-j) in the block is replaced with the representative value for mosaic processing. Note that this representative value is not limited to the center pixel data, but may be represented by any pixel data within the block. In some cases, the average value within a block may be used.

Furthermore, as an application of mosaic processing to pictorial expression, there are Japanese Patent Application Laid-open Nos. 179059/1982, 179060/1982, and 179061/1982. These generate the mosaic pattern at random positions using a random function, or change the size of the mosaic pattern according to the contrast or spatial frequency characteristics of the original image data.

[Problems to be Solved by the Invention] However, conventional mosaic patterns have been expressed using binary data of "O" or "l". For this reason, it is not possible to express, for example, unevenness of paint, swells of paint, etc., and only a flat impression can be obtained, making it impossible to perform high-level image conversion.

The present invention is intended to eliminate the above-mentioned drawbacks of the prior art, and its purpose is to provide an image processing method and apparatus for performing more advanced image conversion.

Another object of the present invention is to provide an image processing device that can express a stiff brush touch.

[Means for Solving the Problems] In order to achieve the above problems, the image processing method of the present invention has the following features:
The general idea is to replace image data in at least a part of the area of the original image with image data obtained from the original image data in the area and predetermined multi-tone data.

Furthermore, in order to achieve the above-mentioned object, the image processing method of the present invention obtains an image obtained by multiplying the image data of at least a part of the area of the original image by the product of the original image data in the area and predetermined multi-tone data. The outline is to replace it with data.

Further, in order to achieve the above-mentioned object, the image processing device of the present invention includes a first storage means for storing original image data, a second storage means for storing multi-gradation brush pattern data,
The outline of the present invention is to include an image synthesizing means for synthesizing part or all of the original image data of the first memory means with the brush pattern data of the second memory means.

[Function] In this configuration, the first storage means stores original image data such as a photographic image. On the other hand, the second storage means previously stores multi-gradation brush pattern data. The image synthesis means synthesizes the brush pattern data of the second storage means with part or all of the original image data of the first storage means. As a result, the original image data is added with information such as the rise of paint based on the brush pattern data, making it possible to create a painterly expression with a realistic brush touch.

[Description of Embodiments] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram of an image processing apparatus according to an embodiment of the present invention. In the figure, l is a control processor unit (CPU) that performs main control of the apparatus of this embodiment. Reference numeral 2 denotes a CPU memory, which stores, for example, the pictorial processing program shown in FIG. 2, which is executed by the CPtJ1, and parameters necessary for the processing, in a ROM or RAM (not shown). 3 is a CPU bus, which consists of an address bus, a data bus, and a control bus for the CPU 1.

Reference numeral 26 denotes image data I10, and by connecting an image reading device or an image output device (not shown) to this, it is possible to input and output the image data before and after the painting process. 14 to 16 are image memories of 8 bits per pixel, which store the image data l.
R (red), G (green), B (
Reference numeral 617 is a texture memory with 8 bits per pixel, which stores the original image data of the three primary colors (blue), respectively.
For example, image data of a canvas (knit pattern) used in oil painting is stored. Reference numeral 19 denotes a work memory of 8 bits per pixel, which temporarily stores the results of arithmetic processing performed on the image data in the image memories 14 to 16. Reference numeral 2o is a 16-bit memory with 16 bits per pixel, which temporarily stores the product-sum results etc. performed on the image data of the image memories 14-16.

Reference numerals 21 to 23 are look-up tables (LUTs) made of high-speed RAM, each having a storage capacity of 256×8 bits, allowing the CPU 1 to rewrite the gradation conversion table. Each of the LUTs 21 to 23 has eight address lines (0 to 23).
255 address), and the address data is given by image data (0 to 255 gradations) output from the image memories 14 to 16, respectively. On the other hand, the read data lines of each LUT 21 to 23 are composed of eight lines, and these are video/"C
24. In addition, the CPU bus 3 is directly connected to each LUT 21 to 23, and the CPU
U1 can read and write the contents of LUTs 21 to 23 at any time. Reference numeral 25 denotes a video controller, and by connecting a CRT display device or a video printer (not shown) to this controller, image data before and after the painting process can be monitored.

4 is an edge detection section, and the image memories 14 to 1
Edge portions in the original image are detected from the image data obtained by color-compositing each image data in 6, binarized, and the resulting binarized edge data is output. This is to make the properties and shapes of brush patterns generated different between edge portions (contrast portions) and flat portions in the original image data. Reference numeral 5 denotes a thick line processing section, which performs pattern thick line processing on the binary edge data output from the edge detection section 4. Reference numeral 6 denotes a garbage processing section, which removes isolated noise patterns with small areas from the thick line pattern that has been subjected to the thick line processing, and leaves only the necessary edge pattern portions. Reference numeral 11 denotes an edge direction detection section, which enables a brush pattern selection section 8 to be described later to select an appropriate brush pattern by detecting the presence or absence of directionality of the detected edge portion and its direction. Reference numeral 7 denotes a drawing position generating section, which includes a random function generating means (not shown), and generates a drawing position of brush pattern data to be described later at random. Reference numeral 8 denotes a brush pattern selection section, which stores a plurality of types of small multi-valued (or binary) brush pattern data in advance in a ROM (not shown), and selects from among the plurality of brush pattern data or as described later. The large multivalued (or binary) brush pattern data stored in the ROM (not shown) in the brush pattern rotation unit 9 and sequentially generated by pattern rotation processing is processed based on the detection results of the presence or absence of edges, etc. select. Reference numeral 9 denotes a brush pattern rotation unit, which is a ROM (not shown) in this embodiment.
By storing a single large multivalued (or binary) basic brush pattern data in the computer and sequentially applying a rotation effect to the basic brush pattern, multiple types of brush pattern data can be generated. . A drawing section 10 draws (synthesizes) the brush pattern data selected by the brush pattern selection section 8 at the randomly generated drawing positions on the original image data. That is, for example, in an example where the brush pattern data is composed of multiple values, the original image data in the vicinity of the generated drawing position is converted into uneven, glossy image data that looks as if it were drawn with a real brush of that color. . Reference numeral 12 denotes an image synthesis section, which further synthesizes a texture image such as a canvas on the image data for which drawing processing has been completed. 13 is an image memory controller, and the edge detection section 4
The image memories 14 to 16 and the LU 721 to
Performs 23 necessary controls.

FIG. 2 is a flowchart of the pictorial processing procedure of the embodiment. In the following explanation, the original image at address (x, y) (
concentration) data ai (x.

y). However, when the subscript i represents R, G, and B image data individually, it is written as R, G, and B, respectively. Furthermore, the image data al (x, y) is 8 bits (0 to 255 gradations can be expressed), the value of maximum density data (darkest data) is set to gradation O, and the value of minimum density data (brightest data) is is the gradation level 255.

<Step Sl> The CPUI takes in original image data R, G, and B from the outside via the image data l1026, and stores them in the image memories 14 to 16, respectively.

At this point, the contents of the LUTs 21 to 23 have standard conversion characteristics in which input and output are equal, as shown in FIG.

<Step S2> The edge detection unit 4 extracts edge portions based on the original image data R, G, and B in the image memories 14 to 16. In this edge extraction process, first, image data matching the human visibility curve is created according to equation (2).

a (x, y) = - (3aR (x, y) + 6 aa (x, y) + aa
(x, y)) In the above equation (2), the original image data R, G, B
were synthesized at a ratio of 0.3:0.6:0.1.

Since paintings are meant to be appreciated with the eyes, the original image data is first synthesized according to the human visual sensitivity curve, and edges are evaluated. At this time, each product-sum result is temporarily stored in the 16-bit memory 20 in order to perform the product-sum calculation in parentheses on the right side. The product-sum result in parentheses (l/IOL) is then stored in the work memory 9.

Next, for example, (3
X3) Edge extraction processing is performed using a matrix differential operator. FIG. 4 shows an example of the differential operator employed in this embodiment. This differential operator detects edges (contrast) that increase in brightness toward the right of the image.

By the way, in the past, the edge detection results were stored in a separate memory. However, in this embodiment, if the edge detection result is 8 bits per pixel, the threshold value for later binarization processing will be in the upper 4 bits, so the upper 4 bits of the detection result will be stored in the lower 4 bits of the image memory 14. Store.

This is due to the following reason.

In this embodiment, the image data that has been rendered in the process described later is temporarily degenerated to 8 gradations and output, but at that point, L
The relationship between input and output gradations in UTs 21 to 23 is, for example, the first
It is shown in Figure 0. That is, in FIG. 10, for example, when the input gradation ranges from O to 31 (256 gradations/8), the output gradation is 31. Thereafter, similarly, every time the input increases by 32 gradations, the output increases by one gradation. Therefore, the lower 5 bits representing the 0 to 31st gradations in the original image data R, G, and B are not actually used.
Even if the data in this part is destroyed, there will be no effect on the gradation conversion result. Therefore, in this embodiment, the image memory 14
The lower 5 bits of 16 to 16 can be freely used as a work area.

To put this into perspective, when converting image data with n gradations to m gradations, the bit area that stores gradations from O to (n/m)-1 in memory is used as the work area. This means that it will be assigned.

Next, the differential operator in FIG. 4 is rotated counterclockwise by π/4 to obtain the differential operator in FIG. 5. This differential operator detects edges that increase in brightness toward the upper right of the image. Then, the upper 4 bits of this edge detection result are compared with the detection result stored in the lower 4 bits of the image memory 14, and the larger one is stored in the lower 4 bits of the image memory 14. In the same manner, the differential operator is rotated counterclockwise by π/4, and as a result, the highest 4 bits of the largest edge detection result seen from a total of 8 directions are detected, and this is stored in the lower 4 bits of the image memory 14. Store.

As a result, the largest edge component of all pixels of the image data converted into a visible image is extracted into the lower four bits of the image memory 14.

Next, the lower 4 bits of the image memory 14 are set to 2 by a predetermined threshold.
Pixels determined to be edges (greater than the threshold value) are assigned PID 1", and pixels determined to be less than that are assigned PID °O".
, and write this into the lower 1 bit of the image memory 14.

In this way, the lower one bit of the image memory 14 stores edge pattern data regarding edge components of the original image.

The edge pattern generated in step S3 and step S2 is too thin to perform the processing described below. Therefore, edge pattern thickening processing is performed. This thickening process is performed as follows (3), where the edge pattern data of interest in the image memory 14 is a (x, y).
Do it according to the formula. That is, when a (x, y) ”1, a (x
+i,y+j)=l.

Here, 13≦i≦3. An integer of −3≦j≦3.Then, the result of this thick line processing is stored in image memory 1, for example.
Stored in the lower 1 bit of 5.

Step S4> The thickened edge pattern data obtained in step S3 above usually contains many small isolated noise patterns. The dust processing section 6 calculates the area of all the edge pattern data stored in the image memory 15 based on the determination of connectivity or non-existence, and erases the edge pattern data having a predetermined area or less as a noise pattern. This determination of the presence or absence of connectivity is based on the edge pattern data of interest in the image memory 15 a (x,
y), it is performed according to equation (4).

That is, when a (x, y) = 1, a (x+i,
Check yj)=1.

Here, 11≦i≦1. It is considered that there is continuity if at least one of the conditions of -1≦j≦1, ie, a (x+i, y+j)=1, is satisfied.

<Step S5> In step S5, the drawing start position generating section 7 generates drawing position information of the brush pattern. If the drawing positions of the brush patterns are generated regularly and sequentially, the naturalness of the pictorial expression will be lost.

Therefore, the drawing position in the embodiment is generated at a random position.

This drawing position information is first generated for the original image data R. The drawing start position generating section 7 is internally equipped with a random number generating means (not shown), and the random number generating means is, for example, a CPU.
1 to three random number generation parameters (an integer that gives a random number generation sequence in the row direction, an integer that gives a random number generation sequence in the column direction, and the number of random numbers to be generated), the receiver generates random numbers in the corresponding mode. The drawing start position generating unit 7 determines the drawing start position (x+a) of the brush pattern according to the generated random number.
,y,).

In this example, the random number generation parameter is set to the original image data G.
Since the processing of and B is the same, the drawing position and the number of drawings are also the same as those of the original image data R.

Step S6> In step S6, it is determined whether or not edge pattern data exists at the generated drawing position, and if there is no edge pattern at the drawing position (edge data = O), a routine is performed to draw with large brush pattern data from step S7 onwards. Proceed to
There is also an edge pattern at the drawing position (edge data =
If 1), the routine proceeds to a drawing routine using small brush pattern data after step Sll. This is because the edge portion of the image is drawn in detail, and the other portions are drawn roughly.

On the other hand, the pattern selection unit 8 first selects large brush pattern data or small brush pattern data in accordance with the determination in step S6. Any brush pattern data, for example, O-
It is composed of multivalued image data of n gradations (binary image data in other embodiments), and substantially a plurality of types are prepared. In this embodiment, an RO (not shown) in the brush pattern rotating section 9
One type of large brush pattern data (hereinafter, multi-value and binary are collectively referred to as brush pattern data) is stored in M, and three small brush pattern data are stored in a ROM (not shown) in the brush pattern selection section 8. I remember the type. As small brush pattern data, in order to appropriately draw the edge portion of the original image data, a pattern that is close to a circle with no directionality, a vertically long pattern suitable for vertical edges, and a horizontally long pattern suitable for horizontal edges are stored. There is.

FIGS. 6(A) to 6(D) show examples of multivalued brush pattern data of the embodiment. 6(A) shows an example of large brush pattern data, FIG. 6(B) shows an example of non-directional brush pattern data, FIG. 6(C) shows an example of vertically long brush pattern data, Figure (D) shows examples of horizontally long brush pattern data.

These brush pattern data are, literally, brush pattern data (like drawing on paper by applying paint to large or fine brushes).
However, in the embodiment, it is composed of an achromatic color representing the shape. In other words, when the direction in which the light beam is irradiated is taken into consideration in the output image, a feeling of bulge in brightness, a feeling of thickness, or a feeling of unevenness along the direction of the brush stroke will appear accordingly, and the pattern shape will have a tail at the end. is subtracting. In this embodiment, a representative example of such brush pattern data is read from, for example, an actual painting sample image or generated by image processing, and is stored in advance as brightness information such as the raised shape of a brush pattern.

Further, FIGS. 6(E) to (H) show examples of binary brush pattern data of other embodiments. FIG. 6(E) shows an example of large brush pattern data, FIG. 6(CF) shows an example of non-directional brush pattern data, FIG. 6(G) shows an example of vertically long brush pattern data, and FIG. (H") shows an example of horizontally long brush pattern data.It is the same as the case of multi-valued brush pattern data except that there is no brightness rise information.

Step S7> Step S7 is input to a routine that uses large brush pattern data. In step S7, it is first determined whether the rotation process of the large brush pattern data has been completed. If the rotation process has been completed, the process advances to step S9, where large brush pattern data is drawn. If the rotation process has not been completed, the process advances to step S8, and the brush pattern data is rotated.

Step S8> In step S8, rotation processing of large brush pattern data is performed. If multiple types of large brush pattern data are stored in the ROM in advance, access is quick. However, in this embodiment, for the purpose of saving the ROM capacity, by rotating one piece of brush pattern data, substantially the same effect as preparing a plurality of types of brush pattern data is obtained. This rotation process is performed by the brush pattern rotation unit 9. For example, if the basic position of the brush pattern data is the vertical direction,
de Perform rotation processing sequentially.

Note that this takes into account the direction in which the brush is carried when the rotation range is 20 degrees above. Moreover, within this range, there is no practical problem even if the shadow effect does not change depending on the direction of the irradiated light beam.

Specifically, the coordinates (K
, L) are determined according to equation (5).

Here, (1, J): Coordinates of input brush pattern data (xo, y
o): Center coordinate of rotation θ: Rotation angle Here, the rotation angle θ changes sequentially by ldeg, but since the drawing position generated in step S5 is random, the original image data R will end up being a random position. This means that brush pattern data in random directions appears. Moreover, a certain degree of directionality remains in the upper 20de as a whole, and it is possible to express the brush strokes (habits) unique to paintings.

<Step S9> The drawing unit 1o draws large brush pattern data at the drawing position generated in step S5.

FIG. 7(A) shows the generated drawing start position (X.

y. ) and the center position of the multivalued brush pattern data (xe, 'J
It is a figure showing the relationship of c). That is, the original image data and the multivalued brush pattern data are not aligned so that the relationship shown in FIG. ) is determined according to equation (6).

C+ (x'.y') Here, i: R, G, B aI (Xe, ye): Original image data P corresponding to the center position of the brush pattern data (x.y) Near address (x.y) Multivalued brush pattern data n: Number of gradations (x'.y') of the brush pattern data The position of the original image data corresponding to the near address (x.y), that is, the equation (6) is The color of the box pattern area (i = R) is determined by the original image data ar (Xe * yc) corresponding to the center position.

G, B) and brightness are represented, and the surrounding area is filled with paint bulge information P (x.y) of multi-valued brush pattern data.
It is drawn with some changes.

Note that the reason for dividing by (n-1) in equation (6) is to make the calculation result 8 bits. In this way, the calculation of equation (6) is performed sequentially from the upper left of the multi-value brush pattern data P (x, y), and when writing corresponding to one pixel of the multi-value brush pattern data is completed, the process proceeds to step SIO.

FIG. 7(B) shows a case where the brush pattern data is binary as another example. In this case, the above equation (6) can be expressed as equation (6)'.

C+ (x', y') = P (x, y) xa, (xc, yc) (6)' Here, P (x, y) binary brush pattern data of near address (x, y), that is, In equation (6)', the original image data a1 (Xe) corresponds to the center position of the binary brush Bakun data.

yc) and the color of the binary brush pattern area (i=R
, G, B) and the brightness, and the surrounding P (x,
The portion where y)=1 is drawn using the above-mentioned representative value.

In addition, in the above-mentioned embodiment, the original image data a i (X c
, yc) and draw multivalued or binary brush pattern data P (x, y) around it, but the invention is not limited to this. In addition, original image data a+ (Xc, yc
), or original image data that matches the position where the multivalued brush pattern data P (x, y) is greater than a predetermined value or the binary brush pattern data P (x, y) is "1". The average value of these original image data may be used, or the maximum value or minimum value of those original image data may be used.

Step SIO> It is determined whether drawing has been completed for all pixels of the brush pattern data, and if drawing has not been completed, the process returns to step S6. Therefore, if edge data is encountered during drawing of large brush pattern data, drawing of the large brush pattern data is terminated at that point, and the process proceeds to drawing processing of small brush pattern data from step Sll onwards. This is because drawing of edge portions takes priority over other drawings.

<Step Sll> The edge direction detection section 11 detects the direction of the edge data recorded in the lower 1 bit of the image memory 15, and in response to this, the brush pattern data selection section 8 selects the brush pattern data suitable for the detected direction. do. Edge direction detection is
For example, calculate the AND of the one-dimensional operator and edge data as shown in Figure 8 or Figure 9, compare the number of pixels for which the result is true in the vertical and horizontal directions, and if the difference is greater than a certain value. The direction with the largest number of pixels is determined to be the edge direction.

Specifically, the edge direction signal S is obtained according to equation (7).

S=F (T (x, y) nE (x, y)) - F
(Y (x, y)nE (x, y)) Here, E (x, y): Edge data T (x, y): Vertical operator Y (x, y): Horizontal operator F(): Based on this edge direction signal S, the function brush pattern data selection unit 8, which calculates the number of pixels for which the logical product is true, sets d to a predetermined number, and if -d≦S≦d, then "circle pattern J; if S<-d," ``Horizontal pattern J, if d<S, select ``vertical pattern''.

Step S12> The drawing unit 10 draws the brush pattern data selected in step Sll. As a result, the edge portion is drawn along the direction using small brush pattern data or elongated brush pattern data, so that a sharp pictorial expression can be achieved.

Step S13> It is determined whether drawing has been completed for all pixels of the small brush pattern data. If not completed, step S
Return to L2, and if completed, proceed to step S14.

Step S14> The CPU determines whether drawing processing has been performed for a set number of randomly generated drawings. If the set number of items has not been completed, the process returns to step S5, and if the process has been completed, the process proceeds to step S15.

(Step S15) The CPU determines whether or not the drawing process has been completed for the three sides of the image data R%G and B. If the processing has not been completed, the process returns to step SS5 to start processing the remaining surfaces. If the drawing process for all surfaces has been completed, the process advances to step S16.

<Step S16> The CPU 1 rewrites the contents of the LUTs 21 to 23 and performs gradation conversion of the image data R, G, and B for which drawing has been completed. 1st
FIG. 0 is a diagram showing an example of table conversion characteristics of LUTs 21 to 23. In the figure, the horizontal axis indicates the input gradations O to 255 of the image data R, G, and B that have been rendered. However, the lower 4-bit data of the image data R, G, and B for which drawing has been completed is meaningless because it is lost in the process of step S2. However, the output gradation on the vertical axis is in a relationship that ignores the lower 5 bits, so the image data R after drawing is
, G, and B can be properly converted. LUT21~23
The image data R, G, and B that have undergone gradation conversion are stored in the image memories 14 to 16 again. This processing is performed in memory brain units, so it can be executed at high speed.

<Step S17> Finally, the canvas image data in the texture memory 17 and the image data in the image memories 14 to 16 are combined.

Specifically, the image data G+ (x, y) after composition is obtained according to equation (8).

G+ (x, y) =aA+ (x, y)+bT (x, y)
Here, a, b: constants and a+b=1 i: R, G, B A+ (x, y): Image data of image memories 14 to 16 T (x, y): Image data of texture memory 17 Furthermore, texture In the memory 17, (bxT (x,
y)) can be stored. Also (axA
+ (x, y)) can be easily performed by writing the conversion table shown in FIG. 11 into the LUTs 21 to 23, for example. The two image data are added in each memory brain R and G.

Since the process can be performed in the entire B, the process in step S17 is also executed in real time.

Incidentally, in the description of the above-mentioned embodiment, all the drawing processing was performed by digital calculation, but the present invention is not limited to this.

For example, a robot can be equipped with several types of real brushes,
In addition, the brush pattern data with the selected or rotational angle may be drawn at the randomly generated drawing position.

Further, in the above embodiment, the brush pattern data is selected depending on the presence or absence of an edge and the direction of the edge, but the present invention is not limited to this. For example, the brush pattern data may be selected by analyzing the spatial frequency components of the original image data.

Further, in the above embodiment, a 3×3 pixel differential operator is used for edge extraction, but the present invention is not limited to this.

For example, the size and contents of the differential operator may be changed depending on the pixel size of the original image, the size of the brush pattern data, etc.

Further, in the above embodiment, a brush pattern is used as multi-gradation reference data, but the present invention is not limited to this. It is clear that the invention also encompasses the use of reference data used for other purposes.

[Effects of the Invention] As described above, according to the present invention, advanced image conversion, which was previously impossible, becomes possible.

Further, according to the present invention, for example, multi-tone brush pattern data is synthesized with flat original image data, so the brush touch is similar to that of a person actually drawing with a brush, unevenness, and paint. It is possible to faithfully express excitement, etc.

[Brief explanation of the drawing]

FIG. 1 is a block configuration diagram of an image processing apparatus according to an embodiment of the present invention, FIG. 2 is a flowchart of a painting processing procedure of the embodiment, and FIG. 3 is a gradation conversion table in the initial state of LUTs 21 to 23 of the embodiment. Figure 4 is a diagram showing an example of the characteristics. Figure 4 is a diagram showing an example of the differential operator employed in the example. Figure 5 is the differential operator when the differential operator in Figure 4 is rotated counterclockwise by π/4. Figures 6(A) to 6(D) are diagrams showing examples of multivalued brush pattern data of the embodiment, and Figures 6(E) to 6(H) are examples of binary brush pattern data of other embodiments. Figures 7 (A) and (B) show the drawing start position (X, y,) and the center position (xc) of the selected brush pattern data.
, y c ), FIG. 8 is a diagram showing an example of the vertical edge detection operator of the embodiment, FIG. 9 is a diagram showing an example of the horizontal edge detection operator of the embodiment, FIG. 10 is a diagram showing an example of tone conversion table characteristics of LUs 721 to 23 of the embodiment, and FIG. 11 is a diagram showing an example of LUT conversion characteristics during texture image synthesis. In the figure, 1...-CPU, 2...CPU memory, 3...
・CPU bus, 4... Edge detection section, 5... Thick line processing section, 6... Dust processing section, 7... Drawing start position generation section, 8... Brush pattern data selection section, 9 ... Brush pattern data rotation section, 10... Drawing section, 11... Edge direction detection section, 12... Image composition section, 13... Image memory controller, 14-20... Image memory, 21-23...LUT, 24...video bus,
25... Video controller, 26... Image data I10. Input membrane January Figure 3 Figure 4 Figure 5 (A) (B) (C) (D
) Great translation ti kuso 75 廂a = L,
Fear of Ichi, Oh Huang Cho (E)
(F) (G) (H) Figure 6 (A) Figure 7 Figure 8 (B) Figure 9

Claims (3)

[Claims] (1) An image processing method characterized by replacing image data in at least a part of an area of an original image with image data obtained from the original image data in the area and predetermined multi-tone data. (2) An image processing method characterized by replacing image data in at least a part of an area of an original image with image data obtained by multiplying the original image data in the area and predetermined multi-tone data. . (3) In an image processing device that converts original image data into aesthetic image data with a painterly touch, a first storage means for storing the original image data and a second storage means for storing multi-gradation brush pattern data. An image processing apparatus characterized by comprising: an image synthesizing means for synthesizing brush pattern data in the second storage means with part or all of the original image data in the first storage means.