JPH0131232B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention corrects the pixel density near the boundary line of the images to be combined to reduce the difference in density between the two images when combining a plurality of images consisting of pixels of light and dark shading to obtain one combined image. The present invention relates to a method for obtaining a loosely combined image.

2. Description of the Related Art Inserting one or more other images into one image or overlapping a portion of one image to create a combined image is performed in printing or various displays. However, when forming such a combined image, if there is a difference in pixel density between the two images near the boundary of the combined images, it will be observed as if line information other than image information exists at the boundary, causing a sense of discomfort. Sometimes.

Conventionally, as one method for eliminating the above-mentioned disadvantages, smoothing processing using an electronic computer has been performed. This conventional smoothing process combines the cropped image and the image to be fitted, then averages the image information in the vicinity along the boundary line of the combined image, or uses a two-point interpolation method. image information could not be fully utilized, and the blurred parts remained unnatural. Furthermore, in the conventional method, the boundaries of the combined images are processed again, resulting in an amount of information of about 50 MB for A-4 size, which also has the disadvantage of being time-consuming.

In view of the above points, the present invention corrects the image density constituting the image near the boundary line of a combined image so that it changes smoothly at an arbitrary conversion rate without losing image information, and blurs the image near the boundary line. There is no discomfort,
It is an object of the present invention to provide a method for forming a combination image that does not give any noticeable appearance.

The present invention will be explained in detail below with reference to FIGS. 1 to 18.

FIG. 1 shows an image A, and FIG. 2 shows an image C including an image B. FIG. However, the case where image B and image C are one continuous image is also included. Current image A
Let us obtain a combined image as shown in FIG. 4 in which image B is inserted.

Image B has a diameter R, and the boundary between images A and B is l.

FIGS. 6 to 8 illustrate cross-sectionally which parts of the boundary line l are to be blurred. In the case of Fig. 6, the part (B') that enters the A image by the width a from the boundary line l is blurred, that is, the part of the A image and the B image (strictly speaking, the C image part that is in contact with the B image) is blurred. However, this will be explained assuming that the B image is expanded.)
This means that the density of each pixel is smoothly changed at an arbitrary conversion rate. FIG. 7 shows a case where the blurred portion is located in both the A image and the B image, sandwiching the boundary line. Furthermore, FIG. 8 shows the case where the width part of a inside the boundary line l is blurred.

FIGS. 9 to 11 are planar and partial views of FIGS. 6 to 8. FIG. The portion marked with a small dot in the figure is the portion that will be blurred.

1 to 4 show the process of obtaining an enlarged 8-bit mask signal from the 1-bit mask signal of image B. Figure 12-1 shows a 1-bit mask signal created based on image B, which has not yet blurred the area near the boundary line, as shown in Figure 4, and shows R in Figure 4 and R in Figure 12-1. have the same value. FIG. 12-2 shows an 8-bit mask signal created using a programmable cursor that can create an 8-bit mask signal by automatically sweeping the 1-bit mask signal. The programmable cursor referred to here is a device that outputs signals in multiple stages on concentric circles, and the output shape and difference between stages can be changed as appropriate. In Figure 12-2, line l
The inside of is the 255/255 level, that is, the 1 level part. In the following explanation, the first level of the mask signal outputs the image signal, especially the density, as it is at 100%.
0 level outputs no image signal, i.e. 0
% density is output, and the intermediate portion between 0 and 1 is assumed to output an image signal with a density corresponding to the ratio.
Hereinafter, the line connecting the equal concentration levels will be referred to as the equal concentration conversion rate curve. Between the lines l and l' in FIG. 12-2, that is, within the width a, there is a 256-step isodensity conversion rate curve from 255/255 to 0/255, but it is not shown.

FIG. 12-3 shows a 1-bit mask signal newly created based on an arbitrary one of the 256 levels of equal density conversion rate curves for the 8-bit mask shown in FIG. 12-2. In the figure, the size of the 1-level range is R+2r>R, and this new 1-bit mask signal is an expanded 1-bit mask signal.

FIG. 12-4 shows a newly created 8-bit mask signal by automatically sweeping the new 1-bit mask signal using a new programmable cursor. If this newly created 8-bit mask signal is superimposed on the source data signal that has the image signals of both images B and C in FIG. The image in the range up to is output without density conversion, and outside of that, the density is corrected, where the pixel density in the width a range smoothly decreases from the image B side to the image C side at an arbitrary conversion rate. is also output at the same time.

Next, a mask signal that is the complement of the newly created 8-bit mask signal is created, and the 8-bit mask signal that is the complement is multiplied by the image signal of image A to convert the 1-level portion of the image A to image B. An image signal is obtained in which the density of the part overlapping the part of the B image that has been cut is corrected, and the other parts remain as they are. Note that the mask signal that is the complement of the above means that the 255/255 level becomes the 0 level, and the 0 level becomes the 0 level.
If there is a 150/255 level in between the 205/255 level, then that is 250 - 150/255 = 100/
255 level or 50/255 level becomes 255/255
It means a mask signal that is replaced by a level. The density-corrected A image signal and density-corrected B image signal obtained in this way are superimposed and combined as shown in Figure 4, and the area near the boundary line indicates the density by height. At this time, a combined image signal as shown in FIG. 14 is obtained. Then, by appropriately using a known output device for this combined image signal, it is possible to obtain a combined image in which the vicinity of the boundary line is blurred.

In addition, in the above method, when selecting an arbitrary density conversion rate curve for the 8-bit mask signal, if the highest density conversion rate curve, 255/255, is selected, it will have the same size as the previous 1-bit mask. It becomes unrealistic. In that case, the operations shown in FIGS. 12-3 and 12-4 are completely unnecessary.

The image signal of the density-corrected image B obtained by the 8-bit mask signal obtained by performing all the operations shown in FIG. It also consists of the C image area, which is not subjected to density correction, and further outside this, a blurred area of width a, whose density has been corrected and whose density is decreasing at an arbitrary rate of change in the range of 255/255 to 0/255. . However, the invention recited in claim 1 naturally includes a case where the process ends with the steps up to FIG. 12-2.

Also, in the above explanation, the conversion from a 1-bit mask signal to an 8-bit mask signal was explained, but this is not limited to an 8-bit mask signal.
If the resolution of the image is not so required, it can be selected from the range of 2 bits to 8 bits depending on the purpose of use. At present, the use of 8 bits is sufficient for fine art that requires high resolution. However, it is of course possible to use 8 bits or more. This also applies to the following description.

Next, a case will be explained in which a reduced 1-bit mask is used. In this case, if you want to fit cut image D into image A, or create a geometric pattern electrically and automatically and process the boundary between the two images so that it falls within the blur width. It is useful for

About this method, Figures 15-1 to 15-5
This will be explained with reference to the drawings and FIG. 16. First, create a 1-bit mask signal of the image to be fitted (first
Figure 5-1), then 1, which is the complement of this mask signal.
Create a bit mask signal (Figure 15-2),
Next, create an 8-bit mask signal from the complemented mask signal using a programmable cursor (Figure 15-3), then select one arbitrary equal density conversion rate curve of the 8-bit mask signal and create a new one. 1
A bit mask signal is created (Figure 15-4), and then a new 8-bit mask signal is created from the new 1-bit mask signal in the same manner as described above (first
Figure 5-5). The diameter of the equal concentration conversion rate curve lr in the 8-bit mask signal created in this way is smaller than R.

In the above method, when selecting any one equal density conversion rate curve of the 8-bit mask signal shown in Fig. 15-3 and creating a new 1-bit mask signal shown in Fig. 15-4, the selected density conversion rate curve is selected. If 0/255 is selected as the value, an 8-bit mask signal (Fig. 15-5) that can blur only the inside of the outline of image B can be created. If you want to increase the degree of reduction, please refer to Figures 15-1 to 1.
Just repeat the operations shown in Figure 5-4 as many times as necessary.

FIG. 17 graphically shows several examples of density conversion rates that can be selected by the programmable cursor. Note that instead of starting from a 1-bit mask signal, an 8-bit mask signal may be obtained directly from image B using a programmable cursor, and similar operations may be performed thereafter to obtain a desired mask signal.

FIG. 18 shows a block diagram of an image processing apparatus for carrying out the method of the present invention, in which image information is stored in an external storage means 15, and under the control of a host computer 16 using a keyboard 4 as needed. Necessary image information in the external storage means is written into the main memory 7 through the input control means 6 or displayed on the monitor 12. When it is desired to create a 1-bit mask signal for image B shown in FIG. 2, image B is displayed on the monitor and the electronic pen 3 is moved on the decisitizer table along the outline of image B on the monitor. As a result, the 1-bit mask signal (12th-1st
) is written into the secondary memory 14. Next, to create an 8-bit mask signal from this 1-bit mask signal, use the keyboard 4 to specify the desired blur width a and density conversion rate to the programmable cursor 5, and then move the programmable cursor to the sub memory 14. The written 1-bit mask signal is automatically swept to create an 8-bit mask signal as shown in FIG. 12-2, and is written and stored in another plane in the sub-memory 14. Or display it on a monitor and judge whether it is good or bad.

To select an arbitrary iso-density conversion rate curve for the 8-bit mask signal, press the necessary key on the keyboard 4 to select the iso-density conversion rate curve from the sub-memory and create a new 1-bit mask signal. . From now on, a new 8-bit mask signal will be created using the same method as described above.

Note that the mask signal, which is the complement of the mask signal, is achieved by the arithmetic means 8 and written into the sub-memory 14 again.

Next, the 8-bit blur mask signal is multiplied by the A image signal or B image signal written in the main memory to obtain the density-corrected A image signal and B image signal, or a desired combination image signal by combining these. be able to. This combined image signal is then stored in the external storage means 15 under the control of the host computer 16, and the image information stored by the storage means is further converted into a visually visible image by other suitable output means. .

In FIG. 18, 9 indicates an output control means, 10 a digital-to-analog converter, 11 a color converter, and 13 a microcomputer.

The present invention is configured as described above, and the combined image created by the method of the present invention has a density that gradually decreases from one image side to the other image side near the boundary between the two combined images. The image information of each image is gradually attenuated toward the other images. For this reason, the two images are combined very naturally, resulting in an excellent combined image that does not give any sense of discomfort to the viewer. Furthermore, it has the feature that the blur width and blur position can be controlled very precisely.

[Brief explanation of drawings]

FIG. 1 is a diagram showing image A, FIG. 2 is a diagram showing image C including image B, FIG. 3 is a diagram showing image D,
Figure 4 is a combination of images A and B, Figure 5 is a combination of images A and D, and Figures 6 to 8 are the boundaries between images A and B as shown in Figure 4. When the vicinity of line l is partly overlapped and blurred - explanatory diagrams showing various aspects shown in cross sections along the line, Figs. 9 to 11 are planar illustrations of the blurred portions corresponding to Figs. 6 to 8. 12 is a diagram showing the process of creating an expanded 8-bit mask signal from a 1-bit mask signal,
Fig. 13 is an explanatory diagram showing the level magnitude in height to explain the level of the 8-bit mask signal in the same figure as above, and Fig. 14 shows the magnitude of density when blurring near the boundary line of the combined image. Figure 15 is a diagram showing the process of creating an 8-bit mask signal reduced from a 1-bit mask signal, Figure 16 is a diagram showing the level of the 8-bit mask signal by its height, and Figure 17. 18 is a graph showing the density conversion rate as a curve, and FIG. 18 is a block diagram showing an embodiment of the apparatus for implementing the method of the present invention. In the figure, A to D each represent an image, and a is a The blur width, l is the boundary line, R is the diameter of the B image, 0 and 1 indicate the 0 level and 1 level of the digital signal, and 0/
255 to 255/255 indicate isodensity conversion rate curves between 0 and 1, 2 is a digitizer table, 3 is an electronic pen, 4 is a keyboard, 5 is a programmable cursor, 6 is an input control unit means, and 7 is a main memory. ,9
10 is an output control means, 10 is a D/A converter, 11 is a color converter, 12 is a monitor, 13 is a microcomputer, 14 is a sub memory, 5 is an external storage means, and 16 is a host computer.

Claims (1)

[Claims] 1. A method for forming a combined image in which a plurality of images are combined and the vicinity of the boundary line between both adjacent images is blurred, in which a first image and a second image to be combined are overlapped by a width a. The size and position are set based on each image signal, and the width a is set for the first image.
For each pixel signal in
A first image signal whose density is corrected by multiplying it by an appropriately selected density conversion rate that gradually decreases toward the image side is obtained, and then, for the second image, each pixel signal within the width a is multiplied by a density conversion factor of A density-corrected second image signal is obtained by multiplying by an appropriately selected density conversion rate that gradually decreases toward the first image side, and then the density-corrected first image signal and the second image signal are superimposed. A method for forming a combined image, comprising: obtaining an image signal of a combined image with blurred areas near the boundary line, and then outputting the image signal using an appropriate output means. 2. Claim 1, characterized in that the density conversion rate for the first image and the density conversion rate for the second image to be combined are complementary to each other.
A method for forming a combined image as described in Section. 3. The method of obtaining the first image signal and the second image signal whose density has been corrected by multiplying them by the density conversion rate creates a 1-bit mask signal to define the boundary between the first image and the second image. Next, from the 1-bit mask signal, a multi-bit mask signal having an equal density conversion rate curve corresponding to multiple levels of density conversion ratio is created, and then from the multi-bit mask signal, a desired equal density conversion ratio is created. Select a curve, create a new 1-bit mask signal larger than the 1-bit mask based on the selected equal density conversion rate curve, and then create a new multiple-bit mask based on the new 1-bit mask signal. A first image signal is obtained by multiplying the first image signal by the new multi-bit mask signal, and then a mask signal that is the complement of the new multi-bit mask signal is obtained. 2. The method of forming a combined image according to claim 1, wherein the second image signal is multiplied by a mask signal which is the complement of the second image signal to obtain a density-corrected second image signal. 4. The method of obtaining the first image signal and the second image signal whose density has been corrected by multiplying them by the density conversion rate creates a 1-bit mask signal to define the boundary between the first image and the second image. Next, create a 1-bit mask signal that is the complement of the 1-bit mask signal, and then create an equal density conversion rate curve corresponding to multiple levels of density conversion rate from the 1-bit mask signal that is the complement of the 1-bit mask signal. Create a multi-bit mask signal with A new reduced 1-bit mask signal is created, then a new multiple-bit mask signal is created from the 1-bit mask signal, and the density is corrected by multiplying the first image signal by this new multiple-bit mask signal. A first image signal is obtained, a mask signal that is the complement of the new multi-bit mask signal is created, and this mask signal that is the complement is used as a second mask signal.
2. A method for forming a combined image according to claim 1, which is a method for obtaining a second image signal whose density is corrected by multiplying the image signal by the image signal. 5. The method of obtaining the first image signal or the second image signal whose density has been corrected by multiplying by the density conversion rate is based on a plurality of methods having equal density conversion rate curves corresponding to multiple levels of density conversion rate directly from either image. A bit mask signal is created, then a desired equal density conversion rate curve is selected from the multiple bit mask signals, a 1-bit mask is created based on the selected equal density conversion rate curve, and then a 1-bit mask is created based on the selected equal density conversion rate curve. A new multi-bit mask signal is created based on the bit mask signal, and then this new multi-bit mask signal is multiplied by one of the above image signals to obtain one of the image signals whose density has been corrected. The method of claim 1 is a method of creating a mask signal that is the complement of a new multi-bit mask signal, and multiplying the mask signal that is the complement by another image signal to obtain another image signal whose density has been corrected. The method for forming a combined image according to item 1.