JPH0785275A - Method and device for extracting image - Google Patents

Abstract

PURPOSE:To accurately extract an image to be extracted without executing complicated procedure even when fine ruggedness exists on a boundary or a small hole-like non-extraction area exists in the picture. CONSTITUTION:A rough outline 50 like (b) is displayed so as to be superposed to an original picture 42 in (a) and a corresponding area in the picture 42 is extracted and displayed on an outline 52 obtained by thinning the rough outline 50. The histograms of components in a color coordinate system for picture elements in the extracted original picture area 54 are found out like (d) to (f) and a point A for an extraction area and a point B for a non-extraction area in the area 54 are specified like (g). Thereby a component histogram showing that the components of the picture elements on the specified points A, B are close to the peak position of the histogram and are mutually separated is selected out of plural component histograms, the extraction area is distinguished from the non-extraction area by a threshold on the selected component histogram and a real picture extraction area 54a like (h) is extracted.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image extraction method and apparatus used in image simulations such as landscape simulations and design simulations, image database systems or exhibition systems such as electronic catalogs.

[0002]

2. Description of the Related Art As conventional image extracting methods and apparatuses, there are known a method and an apparatus for obtaining an outline of an extraction target area and extracting the outline thereof, and a method and an apparatus using a so-called chroma key method.

As the former method of extracting an image by obtaining the contour line, there is known a method of extracting an image by a method shown in FIG. 5 using an apparatus shown in FIG. That is,
Image input section (for example, TV camera or image scanner)
An original image (also referred to as an original image) 12 as shown in FIG. 5A captured by using the image processing unit 10 is image-processed by a microprocessor 14 as an image processing unit, and an original image in a RAM (for example, VRAM) 16 is stored. The image is stored in the image storage unit 16a and displayed on the CRT display 18 as an image display unit.

Then, by inputting position information from a tablet 20 as an example of a pointing device, C
The original image 12 being displayed on the RT display 18 is shown in FIG.
While drawing a rough contour line (for example, a contour line having a line width of several pixels) 22 as shown in (b) of FIG.
The rough contour image information 23 as shown in FIG. 5C is written in the internal memory (not shown) in the inside or the contour storage part 16b in the RAM 16. That is, the rough contour line image information 23 is written as pattern information in which the data of the pixels in the region where the rough contour line 22 exists is all logic "1" and the data of the pixels in the other regions is logic "0".

Then, by the thinning processing function of the microprocessor 14 which executes the thinning processing program stored in the ROM 25, the rough contour line 22 in the original image 12 is obtained.
To a thinned contour line 24 (for example, a contour line having a line width of 1 pixel) 24 as shown in FIG.
Internal memory of the microprocessor 14 and contour line storage unit 1
The rough contour image information 23 in 6b is thinned and then labeled as shown in (e) of FIG.

That is, the label "1" is given to the pixels in the region where the contour 24 exists, the label "2" is given to the pixels in the region surrounded by the contour 24, and the pixels in the other regions are given. In FIG. 5, the thinned contour line information 27 shown in FIG. 5E is stored in the contour storage unit 16b.

Then, the contour line 24 is extracted from the original image 12 by the image extracting function of the microprocessor 14 which executes the image extracting processing program stored in the ROM 25.
Extract the image contained in (that is, the image to be extracted),
The CRT display 18 displays the extracted image 12a as shown in FIG. That is, the area corresponding to the portion of the label "2" of the thinned outline information 27 labeled and shown in (e) of the same figure is the original image 12 shown in (a) or (d) of the same figure. Cut out from the
The extracted image 12a is displayed.

Further, in the latter method of extracting an image using the chroma key method, an image in which an object to be extracted is placed on a background of a specific color (for example, blue) is made in advance, and color information is converted from the image. It is known that a target image is cut out on the basis of this.

[0009]

However, in the above-mentioned conventional method for extracting an image by obtaining a contour line, the rough contour line 22 is drawn in the original image 12 being displayed by manual input using the tablet 20. After that, the characteristic feature is that the contour line 24 for image extraction is automatically obtained by the thinning processing function of the microprocessor 14 and the parameter adjustment is not necessary, but the time and effort of manual input are saved. For this reason, the rough contour line 22 having a wide line width is input and the line is automatically thinned, so that the following problems occur.

As shown in FIG. 6 (a), when there are fine irregularities at the boundaries of the object 30 which is the display image, such as the hair of a person, the skeleton of a steel tower, the spokes of a bicycle wheel, they are superimposed. The displayed rough contour line (for example, a contour line having a line width of 4 pixels) 32 is a thinned contour line (for example, a line width is 1
As shown in (b) of the figure, the area 36 outside the boundary of the target object 30 cannot be enclosed so that the pixel contour line 34 belongs to the non-extracted area, and unnecessary parts other than the target object 30 are unnecessary. Area 3
There was a problem that 6 would also be extracted.
In the figure, 38 represents the size of one pixel.

Further, as shown in FIG. 7, a small hole-shaped non-extracted region (that is, a non-target region, for example, a region having a size of about 1 pixel) 4 inside the object 30a which is a display image.
When there is 0, such a non-extracted area 40 is set as the rough contour line 3
There is a problem that it cannot be specified by 2a.

Further, in the conventional image extracting method using the chroma key method, if the distribution of the object and the background in the color space is completely separated, the fine portion can be extracted without any problem. There is a big problem that it is necessary to create an image with the target on a specific background in advance, which cannot be applied to a natural image.

In order to overcome such a problem and to apply a region division by a threshold value (threshold value, the same applies hereinafter) in a chroma key-like color space to a natural image, a method called "repeated threshold method" has been devised. Although it is actually applied, the procedure is complicated and requires a large amount of memory capacity.
There was a problem that it was not a very practical method.

The present invention has been made in view of the above-mentioned problems. Even if there is a fine uneven region at the boundary or a small hole-shaped non-extracted region inside, the extraction target is achieved without complicated procedures. It is an object of the present invention to provide an image extraction method and device capable of accurately extracting an image.

[0015]

According to a first aspect of the present invention, (1) a rough contour line is superimposed and displayed on an original image being displayed on an image display unit based on an input from a pointing device, 2) For the pixels of the original image within the contour line obtained by thinning the rough contour line, a plurality of types of histograms in which each of a plurality of constituent components of a preset color space is used as a variable are obtained, and (3) from the pointing device. Based on the input of, the point A in the extraction area and the point B in the non-extraction area in the original image in the thinned contour line are designated, and (4) the pixels of the designated points A and B are ) Detecting which position on the multiple types of histograms obtained in
(5) Based on this detection value, the pixels of the designated points A and B are close to different peak positions of the histogram from the plurality of types of histograms obtained in (2), and are separated as much as possible on the histogram. (6) A threshold for segmenting the region to which the pixels of the designated points A and B belong is determined on the histogram selected in (5) above, and (7) above (6) above is selected. The color space is divided into an extraction region and a non-extraction region based on the threshold value obtained in step 1, and the region of the original image corresponding to the divided extraction region is extracted.

In a second aspect of the invention, the color space preset in the first aspect of the invention is (R, G,
B) color coordinate system, (Y, I, Q) color coordinate system, (H, S,
I) A color coordinate system, a color coordinate system obtained by appropriately converting these color coordinate systems, or a color coordinate system using some of these color coordinate systems together. is there.

According to a third aspect of the present invention, the original image stored in the original image storage section is displayed on the image display section, and a rough contour line is drawn in the displayed image based on the input from the pointing device. In an image extracting device adapted to extract a predetermined region of an original image using a contour line obtained by thinning the contour line, a plurality of constituent components of a color space preset for pixels of the original image within the thinning contour line First means for obtaining a plurality of types of histograms, each of which is a variable, and a point A in the extraction area and a point B in the non-extraction area in the original image in the thinned contour line based on the input from the pointing device. Second means for designating the designated points and the designated points A and B
Means for detecting which position on the plurality of types of histograms obtained by the first means corresponds to, the third means;
On the basis of the detection value detected by the third means, from the plurality of types of histograms obtained by the first means, the second
A fourth means for selecting a histogram in which the pixels of the points A and B designated by the means are close to different peak positions of the histogram, and are located as far apart as possible on the histogram, and on the histogram selected by the fourth means. And a fifth means for obtaining a threshold value that divides the area to which the pixels of the designated points A and B belong, and the color space is divided into an extraction area and a non-extraction area based on the threshold value obtained by the fifth means. A sixth means for extracting a region of the original image corresponding to the extracted region is provided.

In a fourth aspect of the invention, the color space preset by the first means of the third aspect of the invention is (R,
G, B) color coordinate system, (Y, I, Q) color coordinate system, (H,
S, I) One of a color coordinate system, a color coordinate system obtained by appropriately converting these color coordinate systems, or a color coordinate system using some of these color coordinate systems in combination. It is a thing.

[0019]

According to the third aspect of the present invention, in the image extracting apparatus according to the first aspect, a plurality of types of histograms in which the plurality of constituent components of the color space preset for the pixels of the original image within the thinned contour line are used as variables are provided by the first means. The second means specifies the point A in the extraction area and the point B in the non-extraction area in the original image in the thinned contour line based on the input from the pointing device, and the third means It is detected which position on the plurality of types of histograms obtained by the first means the pixels of the points A and B correspond to.

Next, the fourth means selects the histogram of the points A and B designated by the second means from the plurality of types of histograms obtained by the first means based on the detection value detected by the third means. Of the peaks different from each other and located as far apart as possible on the histogram, the fifth means selects the area to which the pixels of the designated points A and B belong on the histogram selected by the fourth means. The sixth means calculates the threshold to be divided, and the sixth means divides the color space into the extraction area and the non-extraction area based on the threshold calculated by the fifth means, and extracts the area of the original image corresponding to the divided extraction area.

In the image extracting apparatus according to the invention of claim 4,
The color space preset by the first means of the invention of claim 3 is
(R, G, B) color coordinate system, (Y, I, Q) color coordinate system,
One is selected from the (H, S, I) color coordinate system, a color coordinate system in which these color coordinate systems are appropriately coordinate-converted, or a color coordinate system in which some of these color coordinate systems are used in combination. .

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an image extracting method and apparatus according to the present invention will be described below with reference to FIGS. In these figures, the same parts as those in FIGS. 4 and 5 are designated by the same reference numerals.
FIG. 1 shows a schematic configuration diagram of an apparatus for carrying out an image extracting method. In this figure, 10 is an image input section mainly composed of a television camera, an image scanner, etc., and this image input section 10 is shown in FIG. The original image 42 shown in (a) is taken in and the corresponding image data is output.

Reference numeral 14 denotes a microprocessor as an example of an image processing section. The microprocessor 14 is connected with a CRT display 18 as an example of an image display section and a tablet 20 as a pointing device, and a RAM ( VRAM) 44 is coupled. In the RAM 44, an original image storage unit 16a and a contour line storage unit 16b similar to those in the conventional example are provided, and a histogram storage unit 46 is provided.

Reference numeral 48 is a ROM coupled to the microprocessor 14, and in this ROM 48, the same microprocessor 14 as the ROM 25 shown in FIG.
A program for performing the thinning processing and the extraction processing is stored. In the ROM 48,
Further, a program for causing the microprocessor 14 to perform the processing shown in the following (a) to (f) is stored.

(A) A function of obtaining a plurality of types of histograms in which a plurality of constituent components of a preset color space are used as variables for the pixels of the original image within the contour obtained by the thinning processing function and the extraction processing function (Corresponding to the function achieved by the first means). Here, the color space is
(R, G, B) color coordinate system, (Y, I, Q) color coordinate system,
(H, S, I) color coordinate system, a color coordinate system in which these color coordinate systems are appropriately coordinate-converted, or a color coordinate system in which some of these color coordinate systems are used in combination, and the like. One of the coordinate systems (eg (R, G, B) color coordinate system)
It is selected and set by an instruction from the keyboard or the like by the operator.

The (R, G, B) color coordinate system represents a color coordinate system having R (red), G (green) and B (blue) as constituent components, and (Y, I, Q) color coordinates. System is Y (luminance), I
A chromaticity signal in the I-axis direction of the chromaticity diagram and a color coordinate system having Q (chromaticity signal in the Q-axis direction of the chromaticity diagram) as constituent components
(H, S, I) color coordinate system is H (hue), S (saturation), I
It represents a color coordinate system having (lightness) as a constituent component.

(B) A function of designating the point A for the extraction area and the point B for the non-extraction area in the original image 54 within the thinned contour line based on the input from the tablet 20 (achieved by the second means). Corresponding to the function). (C) A function of detecting what position on the histogram of the plurality of types (for example, R, G, B components) the pixels of the specified points A and B specified in (b) correspond to in (a). (Corresponding to the function achieved by the third means).

(D) Based on the detection values detected in (c) above, from among the plurality of types of histograms obtained in (a) above, the pixels at points A and B specified in (b) above are histograms. A function (corresponding to the function achieved by the fourth means) of a histogram (for example, an R component histogram) that is close to different peak positions and is located as far as possible on the histogram.

(E) A function (corresponding to the function achieved by the fifth means) of obtaining a threshold value (for example, Rt) that divides the region to which the pixels of the designated points A and B belong on the histogram selected in (d) above. (F) A function of dividing the color space into an extraction region and a non-extraction region based on the threshold value obtained in (e), and extracting a region of the original image 42 corresponding to the divided extraction region (achieved by sixth means). Corresponding to the function that will be).

Next, the operation of the embodiment shown in FIG. 1 will be described with reference to FIG. (A) First, similarly to the conventional example, the microprocessor 14 uses the function based on the corresponding program of the ROM 48 to obtain the image information of the original image 42 captured by the image input unit 10 from the original image storage unit 16a in the RAM 44. In addition to writing to the CRT, as shown in FIG.
The original image 42 is displayed on the display 18.

Then, similarly to the conventional example, the microprocessor 14 stores the image information of the rough contour line 50 based on the input from the tablet 20 in the RAM 44 by the function based on the corresponding program of the ROM 48. While writing in the portion 16b, the rough contour line 50 is displayed on the CRT display 18 as shown in FIG.

Further, similarly to the conventional example, the microprocessor 14 uses the thinning processing and extraction processing functions based on the thinning processing and extraction processing programs of the ROM 48 to detect the rough contour 50 in the contour storage unit 16b. Image information is thinned, a corresponding area (for example, an area including the steel tower 54a) 54 of the original image 42 is extracted in the thinned contour line 52, and the CRT is displayed as shown in (c) of FIG.
Display it on the display 18. The above-described thinning processing is performed by a method of leaving a point having a large edge strength or a method of obtaining a center line.

(B) Next, using a terminal device (not shown) (keyboard, for example), the (R, G, B) color coordinate system, the (Y, I, Q) color coordinate system, the (H, S, I) color system. Coordinate system, color coordinate system obtained by appropriately converting these color coordinate systems,
Alternatively, one color coordinate system is selected from the color coordinate systems in which some of these color coordinate systems are used together, and is set as the color space. For example, the (R, G, B) color coordinate system is selected and set (hereinafter, for convenience of description, it is assumed that the (R, G, B) color coordinate system is selected and set as a color space).

Then, the microprocessor 14 causes the RO
2 (d), (e) of FIG. 2 for the pixels in the original image region 54 extracted in (a) above by the histogram calculation function based on the histogram calculation program of M48.
A histogram is obtained for the R, G, and B components that are the constituent components of the (R, G, B) color coordinate system as in (f).
In this R, G, B component histogram, the R, G, B components of the pixels in the original image area 54 are represented by 8-bit digital values (0 to 255), and the digital values are set as the horizontal variable. , The number of pixels of each digital value is the frequency in the vertical axis direction.

(C) Next, the microprocessor 14
The area designation function based on the area designation program of the ROM 48 causes the area within the original image area 54 extracted in (b) corresponding to the input position from the tablet 20, as shown in (g) of FIG. Tower 54 which is the extraction area
A point A which is a part of a and a point B in the non-extraction area 54b are designated.

(D) Next, the microprocessor 14
Detects the position of the designated points A and B designated in (c) above on the R, G, and B component histograms by the position detection function based on the position detection program of the ROM 48.

For example, as shown in FIG. 2D, the R component Ra of the designated point A is located near one peak position of the R component histogram (for example, a digital value 77), and the R of the designated point B is R. The component Rb is located near the other peak position of the R component histogram (for example, digital value 210). Similarly, G which is the G component of the designated points A and B
a and Gb are located on the G component histogram as shown in (e) of FIG. 2 and are B and Bb which are B components of designated points A and B.
Is located on the B component histogram as shown in FIG.

(E) Then, the microprocessor 14
Is a peak selected from the R, G, and B component histograms calculated in (b) above by the histogram selection function based on the histogram selection program of the ROM 48. An R component histogram ((d) in FIG. 2) that is a histogram located close to the position and located as far as possible on the histogram is selected.

(F) Next, the microprocessor 14
Calculates the threshold value Rt that divides the area to which the pixels of the designated points A and B belong on the R component histogram selected in (e) above by the threshold value calculation function based on the threshold value calculation program of the ROM 48. The threshold value Rt is not limited to this, but is calculated by calculating the midpoint of Ra and Rb which are the R components of the designated points A and B.

(G) Next, the microprocessor 14
Uses the division extraction function based on the division extraction program of the ROM 48 to set the color space of the (R, G, B) color coordinate system into the extraction area and the non-extraction area based on the threshold value Rt calculated in (f) above. The original image area corresponding to the divided extraction area is extracted. That is, as shown in (h) of FIG. 2, the steel tower 54a, which is the extraction area in the original image area 54 extracted in (b) above, is extracted, and the CRT display 1 is displayed.
Display at 8.

In the above embodiment, the color space (R, G,
B) The color coordinate system is selected and set, and the midpoint Rt of the distance between the designated points A and B in the color space is obtained as a threshold for distinguishing the extraction region and the non-extraction region. The present invention is not limited to this. For example, a (Y, I, Q) color coordinate system,
Select one of the (H, S, I) color coordinate system, a color coordinate system in which these color coordinate systems are appropriately coordinate-converted, or a color coordinate system in which some of these color coordinate systems are used in combination. Set,
The minimum point between the peaks including the designated points A and B may be obtained as the threshold value.

For example, as shown in FIGS. 3A to 3H, the (Y, I, Q) color coordinate system is selected and set as the color space, and the extraction region and the non-extraction region are divided. The minimum point between the peaks including the designated points A and B may be obtained as a threshold for this.

That is, first, similarly to the conventional example, the original image 42 captured by the image input unit 10 is displayed on the CRT display 18 as shown in FIG. 3A, and based on the input from the tablet 20. The rough contour line 50 is displayed on the CRT display 18 as shown in (b) of the figure, and the corresponding area 54 of the original image 42 is extracted in the contour line 52 obtained by thinning the rough contour line 50. The CRT display 18 is displayed as shown in FIG.

Then, the (Y, I, Q) color coordinate system is selected by the terminal device (eg, keyboard) and set as the color space. Then, for the pixels in the original image region 54 extracted by the thinned contour line by the histogram calculation function, as shown in (d), (e), and (f) of FIG.
(I, Q) A histogram is obtained for the Y, I, Q components that are the constituent components of the color coordinate system.

Then, based on the input from the tablet 20, as shown in FIG. 3G, a point A in the extraction area 54a and a point B in the non-extraction area 54b in the original image area 54 are designated. , The specified point A, B by the position detection function
Which position on the Y, I, and Q component histogram corresponds to is detected.

For example, as shown in FIG. 3D, the Y component Ya of the designated point A is located at one peak position of the Y component histogram, and the Y component Yb of the designated point B is Y.
It shall be located at the other peak position of the component histogram. Similarly, the I components Ia and Ib of the designated points A and B are located on the I component histogram as shown in (e) of FIG. 3, and the Q components Qa and Qb of the designated points A and B are shown in FIG. It is assumed to be located on the Q component histogram as shown in (e) of.

Then, by the histogram selection function,
From the Y, I, and Q component histograms, a Y component histogram that is a histogram in which the pixels at the designated points A and B are close to different peak positions in the histogram and located as far apart as possible on the histogram is selected.

Next, the threshold value calculation function calculates a threshold value Yt for dividing the area to which the pixels of the designated points A and B belong on the selected Y component histogram. The threshold value Yt is not limited to this, but is calculated by calculating a minimum point between peaks including the designated points A and B.

Then, the division extraction function divides the color space of the (Y, I, Q) color coordinate system into an extraction area and a non-extraction area based on the threshold value Yt calculated, and the divided extraction area is divided. The original image area corresponding to is extracted. That is,
As shown in FIG. 3H, the extraction area 54 a in the original image area 54 is extracted and displayed on the CRT display 18.

In the above embodiment, the pointing device is a tablet, but the present invention is not limited to this, and a pointing device other than a tablet (for example, a mouse or a trackball) may be used.

In the above embodiment, the image display section is formed by a CRT display, but the present invention is not limited to this, and the image display section is formed by a liquid crystal display, a plasma display, an electroluminescent display or the like. You may do it.

In the above embodiment, the programs for various processes are stored in the ROM and the corresponding functions are achieved by executing these programs. However, the present invention is not limited to this, and other than the ROM. The programs for various processes may be stored in the storage medium (for example, hard disk, floppy disk, etc.) and the corresponding functions may be achieved by executing these programs.

[0053]

As described above, the image extracting method and apparatus according to the first and third aspects of the present invention extracts the corresponding region of the original image within the contour line obtained by thinning the rough contour line by inputting the rough contour line. Then, a plurality of types of histograms are obtained for the original image pixels extracted in the thinned contour line, and the point A for the extraction area and the point B for the non-extraction area are specified for the original image in the thinned contour line, It is detected which position on the plural kinds of histograms the pixels of the designated points A and B correspond to, and the best histogram is selected from the plural kinds of histograms based on the detected value. The threshold for dividing the region to which the pixels of the designated points A and B belong is obtained, and the region of the original image corresponding to the extraction region designated by the designated point A is extracted based on this threshold.

Therefore, a simple operation of inputting a rough contour line for determining a rough extraction region and inputting a designated point for determining a strict (accurate) extraction region, that is, without performing a complicated procedure, The extraction target image can be accurately extracted even if there is a fine uneven area at the boundary or a small hole-shaped non-extraction area inside.

The image extracting method and apparatus according to the second and fourth aspects of the invention have the following effects in addition to the effects of the first and third aspects of the invention. That is, the preset color space is a familiar color coordinate system (R, G, B).
Color coordinate system, (Y, I, Q) color coordinate system, (H, S, I) color coordinate system, color coordinate system obtained by appropriately converting these color coordinate systems, or some of these color coordinate systems Since one of the color coordinate systems is used in combination, it is possible to further simplify the operation of designating the specific region to be truly extracted in the rough image extraction region determined by the thinned contour line.

[Brief description of drawings]

FIG. 1 shows an embodiment of an image extraction method and apparatus according to the present invention, and is a schematic configuration diagram of an apparatus for carrying out the method of the present invention.

FIG. 2 is an explanatory view for explaining the operation of FIG. 1, in which (a) is CR
Explanatory drawing explaining the state which is displaying the original image on T display, (b) is explanatory drawing explaining the state which superimposed and displayed the rough outline in the display image of (a), (c) is (b). (D), (e), which illustrates a state in which the rough contour line is thinned and the corresponding region of the original image is extracted in the thinned contour line.
(F) is an explanatory diagram of a histogram of R, G, B components in the color space of the (R, G, B) color coordinate system, respectively, (g)
Is an explanatory view for explaining a point A for an extraction area and a point B for a non-extraction area designated in the rough extraction area of (c),
(H) is an explanatory view explaining a final extraction region.

3A and 3B are explanatory views illustrating another embodiment of the image extraction method according to the present invention, FIG. 3A is an explanatory view illustrating a state where an original image is displayed on a CRT display, and FIG. An explanatory view for explaining a state in which a rough contour line is superimposed and displayed in the display image of FIG. 7, (c) is a state in which the rough contour line of (b) is thinned and the corresponding region of the original image is extracted in the thinned contour line. (D), (e), (f) are explanatory views for explaining (Y,
(I, Q) an explanatory diagram of a histogram of Y, I, Q components in the color space of the color coordinate system, (g) is a point A for the extraction area designated in the rough extraction area of (c) and a non-extraction area FIG. 3B is an explanatory diagram illustrating a point B for use, and FIG. 3H is an explanatory diagram illustrating a final extraction region.

FIG. 4 is a schematic configuration diagram of an apparatus that implements a conventional image extraction method.

5A and 5B are explanatory diagrams illustrating the operation of FIG. 4, in which FIG. 5A is an explanatory diagram illustrating a state in which an original image is displayed, and FIG. Of the pattern information corresponding to the rough contour line information stored in the internal memory of the microprocessor or the contour line storage unit, and (d) is the thinning process. 4B is an explanatory view for explaining a state in which the rough contour line in (b) is thinned, and (e) is a description for explaining pattern information in which the pattern information about the rough contour line in (c) is thinned and labeled by the thinning process. FIG. 6F is an explanatory diagram illustrating a state in which the extracted image is displayed.

6A and 6B are explanatory views when an extraction target image having fine irregularities on a boundary is extracted using the apparatus of FIG. 4, FIG. 6A is an explanatory view for explaining before thinning processing, and FIG. It is an explanatory view explaining after conversion processing.

FIG. 7 is an explanatory diagram of a case where an extraction target image having a small hole-shaped non-extraction portion inside is extracted using the apparatus of FIG.

[Explanation of symbols]

10 ... Image input unit, 14 ... Microprocessor, 16
a ... Original image storage unit, 16b ... Contour line storage unit, 18 ... C
RT display (an example of an image display unit), 20 ... Tablet (an example of a pointing device), 42 ... Original image, 44 ... RAM, 46 ... Histogram storage unit, 48
ROM, 50 ... Coarse contour line, 52 ... Contour line obtained by thinning the rough contour line 50, 54 ... Corresponding region of the original image 42 extracted in the contour line 52, 54a ... Extraction region in the original image region 54, 54b ... Non-extracted area in original image area 54, A ... Designated point for extraction area, B ... Designated point R for non-extraction area,
G, B ... (R, G, B) color space constituents of the color coordinate system,
Ra, Rb ... R component of designated points A and B, Rt ... R component, threshold value for distinguishing extracted region and non-extracted region on the histogram, G
a, Gb ... G component of designated points A and B, Ba, Bb ... B component of designated points A and B, Y, I, Q ... (Y, I, Q) component of color space of color coordinate system, Ya , Yb, ... Y of designated points A and B
Component, Ia, Ib, ... I component of designated points A, B, Qa, Q
b, ... Q component of designated points A and B.

Claims (4)

[Claims] (1) A rough contour line is superimposed and displayed on an original image being displayed on an image display unit based on an input from a pointing device, and (2) within the contour line obtained by thinning the rough contour line. For the pixels of the original image, a plurality of types of histograms, each having a plurality of constituent components of a preset color space as variables, are obtained, and (3) based on the input from the pointing device, Specify a point A in the extraction area and a point B in the non-extraction area in the original image,
(4) It is detected which position on the plural types of histograms obtained in (2) the pixels of the designated points A and B correspond to, and (5) in (2) based on the detected value. From the obtained plural types of histograms, the designated point A,
Select a histogram in which the B pixel is close to different peak positions of the histogram, and located as far as possible on the histogram, and (6) select the specified points A and B on the histogram selected in (5) above. A threshold for dividing an area to which a pixel belongs is obtained, and (7) the color space is divided into an extraction area and a non-extraction area based on the threshold obtained in (6), and an original corresponding to the divided extraction area is obtained. An image extraction method characterized by extracting a region of an image. 2. The preset color space is (R, G, B)
Color coordinate system, (Y, I, Q) color coordinate system, (H, S, I) color coordinate system, color coordinate system obtained by appropriately converting these color coordinate systems, or some of these color coordinate systems The image extracting method according to claim 1, wherein the image extracting method is one of color coordinate systems in which 3. An original image stored in an original image storage unit is displayed on an image display unit, a rough contour line is drawn in the displayed image based on an input from a pointing device, and the rough contour line is thinned. In an image extraction device adapted to extract a predetermined region of an original image using lines, a plurality of types in which each of a plurality of constituent components of a color space preset for pixels of the original image within the thinned contour line is a variable And a second means for designating a point A in the extraction region and a point B in the non-extraction region in the original image in the thinned contour line based on the input from the pointing device. , Third means for detecting what position on the plurality of types of histograms the pixels at the designated points A and B correspond to, and the detection value detected by the third means, Points A and B designated by the second means are selected from a plurality of types of histograms obtained by the first means.
Means for selecting a histogram in which the pixels in the histogram are located close to different peak positions in the histogram and as far as possible from each other on the histogram, and the designated points A and B on the histogram selected by the fourth means. Fifth means for obtaining a threshold value for dividing an area to which a pixel belongs, and the color space is divided into an extraction area and a non-extraction area based on the threshold value obtained by the fifth means, and an original image corresponding to the divided extraction area. And a sixth means for extracting the area of FIG. 4. The color space preset by the first means is
(R, G, B) color coordinate system, (Y, I, Q) color coordinate system,
One of a (H, S, I) color coordinate system, a color coordinate system obtained by appropriately converting these color coordinate systems, or a color coordinate system using some of these color coordinate systems together. Three
The image extraction device described.