JPS60225280A - Correcting system of picture - Google Patents

Abstract

PURPOSE:To blur the background in an optional degree by segmenting an object from an original picture in the interactive mode. CONSTITUTION:After a noticed area is segmented from an object picture, picture data of the object picture stored in a picture memory 16 is taken into a central processing unit 8, and the blurring processing is performed, and the processed picture is stored in the picture memory 16 again. In the blurring processing, a low-pass filter is used to smooth the picture. When a blurred picture having an optimum background is obtained by the blurring processing, boundary line data used for segmenting is used to synthesize one picture by the central processing unit 8, and this picture is stored in the picture memory 16. The synthesized picture is displayed on a CRT monitor 19, and parts on boundary lines and detailed parts are blurred.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to computer processing of images such as photographs, and in particular, a method for extracting the background of an object in an image through simple interactive processing and quickly blurring the background. The present invention relates to an image modification method suitable for processing while adjusting intensity interactively.

[Background of the invention]

Conventionally, a first method for blurring a specific area, such as a background, in an image such as a photograph is performed at the stage of printing on photographic paper. This covers only the area other than the area you want to blur with photographic paper, and then prints an out-of-focus photo over this. This step-by-step method then covers the area you want to blur, and then prints the focused photo on top of it. The second method is to make corrections after printing on photographic paper. This method uses an airbrush to spray paint onto only the background of a photo to blur it.

The above two methods require skill, require a large amount of manpower and time, and have a high possibility of failure. In addition, there is a process that uses a computer that has an airbrush function, but in this method, a certain brightness level is assigned to each position on the image intermittently, so the brightness It is difficult to select the level and, like the airbrush used to correct photographic paper, the operation is troublesome, so it is inappropriate to use an airbrush to blur images.

As described above, with conventional methods, it is impossible to easily and quickly create an optimal blurred image interactively from a specific area in an image.

[Purpose of the invention]

The purpose of the present invention is to provide a method for quickly and optimally blurring images by using a computer to accurately blur a specific area in an image according to simple human instructions for photo editing, which requires skill. There is a particular thing.

[Summary of the invention]

Here, the principle of blurring a specific area in an image used in the present invention will be explained. First, an object (such as a person) 1 is cut out from the original image shown in Fig. 1 to obtain an image containing only the background 2 (see Fig. 2), and a spatial filter is applied to this background image to smooth it. An image with a background 3 (see Figure 3) is created. At this time, the strength of the blur can be freely set by changing the mask size and load parameters of the spatial filter, and by repeatedly applying the filter. A desired image (see FIG. 4) is obtained by incorporating the cut object l into the background image subjected to the above processing. However, simply incorporating the object into the blurred background image does not blur the portion 4 on the cutout boundary line or the details 5 contained inside the object. Therefore, in the former case.

When incorporating an object into the background, a low-pass filter is applied to some pixels adjacent to the boundary line to smooth it and blur the boundary line. In the latter case, by using a low-pass filter that limits the area to a small area and interactively determining and processing the blurring position while looking at the monitor, the image shown in Figure 5 can be obtained by adding blurring processing to the details. It will be done.

As described above, in this process, by cutting out the object from the original image in an interactive manner, it is possible to apply a blur of arbitrary strength to the background.

[Embodiments of the invention]

An embodiment of the present invention will be described in detail below using the drawings.

FIG. 6 is an overall configuration diagram of an apparatus for performing image processing according to the present invention. The target image 6 is AD converted by an image input device 7 such as a drum scanner, and then stored in the storage device 10 via the central processing unit M8. The stored image is console 9
The image is displayed on the image display 11 according to a command from the user.

While viewing this displayed image, a human operator operates a coordinate input device 12 such as a trackball to indicate a plurality of representative points on the outline of the region of interest in the image to be viewed.

This indication of the representative point is realized by the image discless device 11 shown in FIG. First, according to the coordinates input by the coordinate input device 12, the cursor generator 1
7 displays a cursor on the designated coordinates of the CRT monitor 19 via the DA converter W18. On the other hand, image data sent from the central processing unit 8 is stored in the image memory 16.

It is displayed on the CRT monitor 19 by the DA converter w18. At this time, the cursor is displayed superimposed on the image. The operator looks at the cursor displayed on the monitor 19 and uses the coordinate input device!
! 12 to bring it to a desired position, and when the setting is completed, a command is sent to the display device 11 from a command switch 15 attached to the coordinate input device 12. The cursor generator 17 receives the command and sends the coordinate values to the central processing unit 8. A person repeats the above operations one after another to designate a plurality of representative points of the contour line, and the corresponding pixel coordinates are provided to the central processing unit 8.

Next, consecutive representative points designated above are connected with a straight line, and this is used as a boundary line between the extraction area and the background.

Next, using this boundary line data, the central processing unit 8 determines whether or not each pixel of the target image is inside the extraction area. For internal pixels, the pixel intensity values of the original image are stored as they are in another image memory 20. This means that the region of interest has been cut out from the original image.

Next, background image blurring processing will be explained.

First, after cutting out a region of interest from the target image, the image data of the target image stored in the image memory 16 is loaded into the central processing unit 8, where it is subjected to blurring processing, and the processed image is stored in the image memory. 16 is stored again. Further, this image is displayed on the CRT monitor 19 via the DA converter 18. A person looks at the displayed image, determines the strength of blurring, and repeats the above operation until the optimal blurring strength is achieved.

Here, the blurring process is a process of smoothing an image using a low-pass filter. Let the intensity of the image be x(i*
j), where i and j) are pixel coordinates. At this time, the smoothed pixel value map (i'#
J') is determined by the mask processing. However, M is a parameter that specifies the size of the MXM mask,
W (k, fi) are weights. Therefore, by setting the mask size M large, the strength of smoothing is also increased, and furthermore, it is possible to manipulate the strength of smoothing by setting the weight W(k, Q). Therefore, the size M of the mask, which is a spatial filter, and the 1 weight W(k, fi
) is input from the console 9 as a parameter to set the blur strength.

After performing the blurring process and obtaining an optimal background blurred image, the central processing bag N8 uses the boundary line data at the time of cutting out, and if it is inside the boundary line, the central processing bag N8 If the pixel data is stored in the image memory 20, and if it is external, then the pixel data in the image memory 16, which is the storage area for the blurred image, is processed and combined into one image. 16.

Next, the synthesized image is displayed on the CRT monitor 19, and while looking at the CRT monitor 19, the focus position is selected by operating the coordinate input device 12, and the set coordinates are Smoothing mask processing similar to equation (1) is performed on the pixels at the position. This allows you to blur borders and details.

By performing the above operations, a desired background blurred image can be obtained. The flow of these operating procedures is shown in FIG.

In the embodiments described above, the boundary lines were specified manually, but the boundary lines can also be extracted semi-automatically as follows.

It is also possible to cut out the specified area by linear approximation, and then perform the above-mentioned blurring process after the operator performs the cutting while looking at the monitor. We will explain how the computer automatically detects the area boundary by simply specifying a point.

It is assumed that FIG. 9 is the original image 27 targeted. In order to extract the specific area (target object 29) from the original image 27, the target object 2
9 and the background 28 as shown in FIG.
If this is obtained, each pixel of the original image can be transformed into a contour line 3 using this
The target object 29 can be extracted by determining whether it is within 0 or not. Conversely, when the cut-out object is to be incorporated into the position of the original image, the position data of the contour line 30 is used for determination. Therefore, a method for extracting this contour 630 will be explained. Original image 27 as shown in FIG.
A human designates multiple representative points on the contour of the region of interest. This is 31-1°31-2. ...3l-
n (total n pieces).

This instruction serves as an input for automatic contour line extraction processing by a computer, which will be described below, and does not necessarily require high instruction accuracy. Next, the computer connects consecutive representative points indicated above with a straight line 32. This is called a rough contour line. Next, the following processing is performed for each pixel on this coarse contour line 32.

This will be explained by taking the representative point 33 as an example.

Consider an inspection line segment 34 that is orthogonal to the rough contour a32 at a point 33. The number of pixels Ω between the end points 35 and 36 of the line segment 34 takes a predetermined value. The intensity of each pixel on this line segment 34 is inspected as follows, and a correct contour point, that is, a point 37 on the boundary line between the region of interest and the background is detected.

FIG. 2 shows intensity changes of pixels on line segment 34.

In this example, the pixel intensity of the region of interest 121 is large (that is, bright), and the intensity of the background region 122 is small (that is, dark).
It turns out that it is 7. As shown in FIG. 13, this point is taken as the point where the absolute value of the differential value of the intensity change in the direction of line segment 34 in FIG. 12 is maximum.

By performing the above processing for each pixel point 33 on the rough contour line 32 and connecting the obtained points 37, a desired contour line can be obtained. At this time, it is not necessary to perform the above operation for all points on the rough contour line 32, and contour points may be set at certain intervals.

〔Effect of the invention〕

According to the present invention, background blurring can be easily performed, and the desired blurring strength can be appropriately specified while looking at the monitor. It has the effect of reducing the cost of photographic image processing such as.

[Brief explanation of the drawing]

Figure 1 is the image to be processed, Figure 2 is the background image after object extraction, Figure 3 is the image obtained by blurring the background image, and Figure 4 is the image to be processed.
The figure shows an image in which the extraction area is incorporated into the background blurred image, Figure 5 is the completed image, Figure 6 is the overall configuration diagram of the image processing device, and Figure 7
The figure is a configuration diagram of the image display, Figure 8 is a flowchart of background blurring processing, Figure 9 is the image to be processed, Figure 10 is the outline of Figure 9, and Figure 11 is a diagram showing the automatic extraction method of the outline. , FIG. 12 is a pixel intensity distribution diagram of line segment 34. FIG. 13 is a diagram showing the differential values of FIG. 12. 1...Target, 2...Background, 3...Blurred background, 4...Border line, 5...Unprocessed portion of blurring, 6...Input image, 7...・Image input device, 8...
Central processing unit, 9... Console, 10... Storage device, 11... Image display device, 12... Coordinate input device, 13... Image output device, 14... Output image, 15. ...Switch, 16...Image memory, 17
... Cursor generator, 18... DA converter, 1
9... CRT monitor, 20... Image memory, 27...
・Original image, 28...Background, 29...Object, 30・
...Contour line, 31...Representative point, 32...Rough contour line,
33...Point on rough contour line, 34...Inspection line segment, 37
...Contour points. Mt Mouth Figure 2! ! It Close to Figure 7 Figure 66 21 No. 8 (2) YI 9 Junior 10 Ro

Claims (1)

[Claims] 1. In an image processing system comprising an image input device, an output device, and a processing device, a background area is specified by a boundary line,
An image correction method characterized by interactively blurring the background by applying a spatial filter only to a specified area. 2. The background area is characterized in that the boundary line of the background area is detected from the rate of change in image intensity using a rough specified boundary line, and the background area is precisely and semi-automatically specified from the boundary line. Image correction method in Section 1.