JP3039699B2 - Image processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus suitable for cutting out a predetermined portion of a displayed image.

[0002]

2. Description of the Related Art Various methods have been proposed for cutting out a predetermined portion of a displayed image. The first method is a method of designating all the outlines of an image to be cut out using a mouse or the like. The second method is a method of designating some points of the outline of an image to be cut out and approximately obtaining the outline. The third method is a method of extracting an outline using a color difference or a density difference. According to such a method, for example, only a person image can be extracted from a still image in which a person is shown with a predetermined landscape as a background.

[0003]

However, the first method described above has a problem that the operation is complicated and time-consuming because it is necessary to specify all the contours of the image to be cut out. Although the operability of the second method is improved as compared with the first method, it is difficult to accurately cut out the image because the specification is rough. Further, the third method is to compare the color or density difference of each pixel with a predetermined threshold value and extract an area where the color difference or density difference is larger than the threshold value as an outline. Is set,
There was a problem that an accurate contour could not be obtained locally and the processing time was long.

The present invention has been made in view of such a situation, and it is an object of the present invention to easily and accurately cut out a desired image in a short time.

[0005]

An image processing apparatus according to a first aspect of the present invention includes a means for displaying a ruled line for dividing a display image into blocks so as to be superimposed on a display image; Means for designating a predetermined one, means for determining the connection direction of the specified block, means for selecting a mask operator corresponding to the determined connection direction, and display image using the selected mask operator. Means for determining an outline of an image to be cut out.

According to a second aspect of the present invention, there is provided an image processing apparatus for displaying a ruled line for superimposing on a display image and dividing the display image into blocks, and for designating a predetermined block among the blocks divided by the ruled line. Means for determining an outline of an image to be clipped of a display image from a specified block, means for determining a connection direction of a specified block, and a method of determining an image to be clipped corresponding to the connection direction of the block. Means for performing processing of expansion and contraction of the contour.

[0007]

In the image processing apparatus according to the first aspect, a ruled line is superimposed on a display image, and, for example, an outline of an image to be cut out is specified in units of blocks divided by the ruled line. The connection direction of the block is determined, and the outline of the image is determined using a mask operator corresponding to the determined connection direction. Therefore, it is possible to cut out a desired image accurately and easily in a short time.

In the image processing apparatus according to the second aspect, the expansion and contraction processing of the outline of the image is performed in accordance with the connection direction of the blocks. Therefore, it is possible to obtain a contour that has no discontinuous portion and retains the original state.

[0009]

FIG. 2 is a block diagram showing the configuration of an embodiment of the image processing apparatus of the present invention. The CPU 1 includes a system control unit 1a, an arithmetic processing unit 1b, a work memory (RAM) 1c
It is composed of The system control unit 1a includes a ROM 2
Each part is controlled in accordance with the program written in. The operation processing unit 1b executes an operation necessary for the processing.
The work memory 1c stores data obtained as a result of the calculation as needed. The RAM 3 stores ruled line data, cursor data, direction data, Sobel operator data, and the like. The disk 4 is composed of, for example, a hard disk and temporarily stores an input image. The output memory 5 stores the finally output image data. The input / output interface 6 receives inputs from an output device 7, a keyboard 8, a mouse 9, and the like, and outputs them to each unit via a bus. In addition, data input via the bus is output to the monitor 10, the printer 11, or the like.

Next, the operation will be described with reference to the flowchart of FIG. The output device 7 outputs image data received by satellite broadcasting or image data reproduced by a video disk player, video tape recorder, or the like. This image data is temporarily stored in the disk 4 via the input / output interface 6 and the bus (step 1-1). Next, the CPU 1 reads out the image data stored on the disk 4 and executes a median filter process. That is, for example, as shown in FIG. 3, nine pixels A to I centering on a predetermined pixel A are extracted. Then, an intermediate level is selected from among the levels of the luminance signal Y of the nine pixels A to I, and that level is set as the level of the center pixel A. The image data subjected to the median filter processing in this manner is stored in the work memory 1c.
(1-2).

The image data that has been subjected to the median filter processing is called from the work memory 1c,
The information is displayed on the monitor 10 via the bus and the input / output interface 6 (1-3). Next, the CPU 1 reads the ruled line data from the RAM 3 and displays the data on the monitor 10 via the input / output interface 6. Thereby, for example, FIG.
, A still image including mountains, trees, and a house is displayed, and horizontal and vertical ruled lines are displayed so as to divide the display image into a predetermined number of blocks (1--1). 4). The user operates the keyboard 8 or the mouse 9 to specify the block to which the outline of the image to be cut out while viewing this display (1- 1).
5). That is, when the keyboard 8 or the mouse 9 is operated, the cursor moves. When the cursor comes to the position of a desired block, an arbitrary block can be selected and designated by operating the keyboard 8 or the mouse 9. FIG. 4 shows the block to which the outline of the house belongs by hatching. When the user cuts out the image of the house, the user designates the hatched block. When all blocks are specified (1
-6), the connection direction of the designated block is determined (1-7). Although details of the block connection direction determination will be described later, in this step, it is determined which direction the connection direction of the block belongs to as shown in FIG.

Next, an operator is selected according to the block connection direction determination result (1-8). That is, as shown in FIG. 6A, predetermined blocks (in the figure,
When another block is connected above and below the block (hatched block), the Sobel operator a shown in FIG. 7A is selected (1-81). FIG.
As shown in (b), when another block is adjacent to the left and right of the predetermined block, the Sobel operator b shown in FIG. 7 (b) is selected (1-82). When the block connection state is a state other than the case shown in FIGS. 6A and 6B, both the Sobel operators a and b are selected (1-83).

When the Sobel operator is selected in this way, a calculation is performed using the selected Sobel operator (1-9). This calculation is performed as follows.

That is, nine pixels centering on a predetermined pixel are selected, the data of each pixel is multiplied by the coefficient of the corresponding area of the Sobel operator, and the sum of the multiplication results is obtained. The obtained sum is newly set as data of the pixel at the center of the nine pixels. In the case of selection in step 1-83, the above-mentioned sums are individually obtained for the Sobel operators a and b, and the square root of the sum of their squares is set.

The Sobel operator a shown in FIG. 7A is suitable for extracting a contour in the vertical direction.
The Sobel operator b shown in (b) is suitable for extracting a horizontal contour. Thus, by using the Sobel operator properly, more accurate contour extraction becomes possible.

Next, the arithmetic processing section 1b executes a threshold value determination operation (1-10). This threshold determination operation is performed for each designated block as shown in FIG. 4 by, for example, a discriminant analysis method. As a result, a different threshold value is set for each block. Next, the data obtained in step 1-9 is compared with the threshold value obtained in step 1-10, and the data obtained in step 1-9 is binarized (1-11). This means that the contour has been obtained. Then, the data in the contour obtained in this manner is read from the work memory 1c, and is output from the output memory 5c.
Is written to. The data in the output memory 5 is converted into cyan (C), magenta (M), yellow (Y), and black (K) signals via the input / output interface 6, output to the printer 11, and printed. Also,
The image cut out in this way is output to the monitor 10 and displayed (1-12).

Next, referring to the flowchart of FIG.
The subroutine for block connection direction determination shown in step 1-7 in FIG. 1 will be described.

First, a block to be detected is specified (8-1), and two blocks adjacent to the block are determined (8-2). In this way, three blocks are designated, and it is determined which of the eight directions shown in FIG. 5 is the connection direction of the blocks (8-
3). This determination result is stored in the work memory 1c (8-4). The above operation is performed clockwise or counterclockwise for all the designated blocks (blocks indicated by hatching in FIG. 4) (8-5).

By the way, when the outline of the image to be cut out is extracted as described above, the outline is supposed to be a closed curve, but actually, a part of the outline is often discontinuous (a hole is formed). . Therefore, step 1-1 shown in FIG.
A process for correcting such a discontinuous portion can be inserted between 1 and 1-12. FIG. 9 shows an embodiment of such a correction process.

First, in step 9-1, it is determined whether or not the processing described below has been performed for all blocks. When the processing is completed for all the blocks, the correction processing of the contour image is completed. In this case, since the processing of all the blocks has not been completed yet, the process proceeds to step 9-2, and the position of a block adjacent to the processing target block is detected. That is, steps 9-3 to 9-6
As shown in the figure, the direction of two blocks vertically and horizontally adjacent to the processing target block indicated by hatching is determined. In the case of this embodiment, four modes are set corresponding to the adjacent directions. When blocks are adjacent above and below the block to be processed, mode 1 is set. If the block to be processed is adjacent to the left and right sides, the mode is set to mode 2. The mode 3 is set when the block to be processed is adjacent to the right side and the lower side or the upper side and the left side, respectively. Further, when the block to be processed is adjacent to the left side and the lower side or the upper side and the right side, the mode is set to mode 4. In step 9-2, when it is determined that the direction of the block adjacent to the processing target block is mode 1 to mode 4, then steps 9-3 to 9-
Proceeding to 6, the expansion / contraction direction corresponding to each mode is set. In mode 1, the expansion / contraction direction is vertical (vertical direction), and in mode 2, the expansion / contraction direction is horizontal (horizontal). In the case of mode 3, the expansion and contraction directions are the upper right direction and the lower left direction. In the case of mode 4, the expansion and contraction directions are the upper left direction and the lower right direction.

After the expansion / contraction directions corresponding to each mode are set in steps 9-3 to 9-6, the process proceeds to steps 9-7 to 9-10, where the expansion / contraction processing corresponding to each direction is executed. .

That is, in step 9-7, one pixel constituting the outline of the block to be processed is selected.
The pixel is tuned vertically. In the case of the embodiment, one pixel forming the contour is extended upward by one pixel and extended downward by one pixel. Therefore, a pixel as one point is configured as a straight line having a length of three pixels. The number of pixels to be extended can be empirically determined by experiment. Such expansion processing is performed for all the pixels constituting the contour included in the processing target block.

Next, after such expansion processing is performed,
A contraction process is performed. The contraction processing refers to deleting pixels at the ends of a plurality of pixels forming a straight line, contrary to the expansion processing. This contraction processing is performed on the condition that the contour is continuous. Such a contraction process is also performed for all pixels constituting the outline of the processing target block.

On the other hand, in step 9-8, the vertical expansion / contraction processing performed in step 9-7 is performed in the horizontal direction. Then, in steps 9-9 and 9-10, the same expansion / contraction processing is performed in the direction extending in the upper right direction and the lower left direction, or in the direction extending in the upper left direction and the lower right direction, respectively.

Thus, steps 9-7 to 9-
In 10, the expansion / contraction processing is performed in the direction corresponding to the direction adjacent to the processing target block, and the discontinuous portion of the contour is corrected to be a continuous contour.

After the expansion / contraction processing in steps 9-7 to 9-10 has been performed for all pixels constituting the contour in the processing target block, step 9-1 is performed.
Proceeding to 1, the block to be processed is shifted to the next block. Then, the process returns to step 9-1, and the above-described processing is repeatedly executed.

The above processing is schematically described as follows.
As shown in FIG. That is, when, for example, a part of the pixel data extending in the vertical direction is missing from the pixel data forming the contour as shown in FIG. 10A, the contour as shown in FIG. After the constituent pixel data is expanded in the vertical direction, the contraction processing is performed in the vertical direction so that the pixel data forming the contour is continuous as shown in FIG. 10C. Is corrected.

On the other hand, in order to correct the discontinuous portion shown in FIG. 10A, expansion processing is performed as shown in FIG. If the contraction process is performed without considering the direction so that the data is continuous, as shown in FIG. 10 (e), the pixel data of the originally existing contour may be damaged as shown by the pixel indicated by the broken line in the figure. . However, according to the present embodiment, since the expansion / contraction processing is performed in the direction corresponding to the connection direction of the blocks, such loss of the original contour data is prevented.

[0029]

As described above, according to the image processing apparatus of the first aspect, a mask operator corresponding to a specified block connection direction is selected, and an outline of an image to be cut out is selected using the mask operator. Since the determination is made, it is possible to easily and accurately extract the contour in a short time.

Further, according to the image processing apparatus of the present invention, the expansion and contraction processing of the pixels constituting the contour is executed in the direction corresponding to the connection direction of the blocks, so that the original contour is not damaged. Can be corrected to be continuous.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating the operation of the embodiment in FIG. 2;

FIG. 2 is a block diagram illustrating a configuration of an embodiment of an image processing apparatus according to the present invention.

FIG. 3 is a diagram illustrating processing of a median filter.

FIG. 4 is a diagram illustrating a display image.

FIG. 5 is a diagram showing connection directions of blocks.

FIG. 6 is a diagram showing a connection state of blocks.

FIG. 7 is a diagram illustrating a Sobel operator.

FIG. 8 is a flowchart showing details of a subroutine for determining a block connection direction in FIG. 1;

FIG. 9 is a flowchart illustrating a process of correcting a discontinuous portion of a contour.

FIG. 10 is a diagram for explaining the processing in FIG. 9;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 CPU 2 ROM 3 RAM 4 Disk 5 Output memory 6 I / O interface 7 Output device 8 Keyboard 9 Mouse 10 Monitor 11 Printer

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-4-90544 (JP, A) JP-A-60-143341 (JP, A) JP-A-58-166348 (JP, A) (58) Field (Int.Cl. ⁷ , DB name) G06T 9/20 G03F 1/00

Claims (2)

(57) [Claims] 1. A means for displaying a ruled line for dividing the display image into blocks so as to be superimposed on a display image, and a means for designating a predetermined one of the blocks divided by the ruled line. Means for determining the connection direction of the block, means for selecting a mask operator corresponding to the determined connection direction, and using the selected mask operator to define an outline of an image to be cut out of the display image. An image processing apparatus comprising: a determination unit. 2. A means for displaying a ruled line for dividing the display image into blocks so as to be superimposed on a display image, and a means for designating a predetermined one of the blocks divided by the ruled line. Means for determining a contour of an image to be clipped of the display image from the block, means for determining a connection direction of the designated block, and a contour of an image to be clipped corresponding to the connection direction of the block. Means for performing expansion and contraction processing of the image.