JP2002084458A - Image processing unit and method, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing unit that can accurately extract a telop. SOLUTION: An input image and a stored image are composed by shifting e.g. the phase of the stored image supplied from a storage memory so that the phase of pixels of a telop part on a field A being the input image is coincident with the phase of pixels of a telop part on the stored image as shown in Figure 4(A). As a result, a level distribution of the pixels of the input image in the telop part is conserved as shown in Figure 4(B). On the other hand, in the background, the phase of the stored image is shifted resulting in that the phase of the pixels of the input image for the part and the phase of the pixels of the stored image may be inverted and the level distribution for the case is made flat.

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 and method, and a recording medium, and more particularly, to an image processing apparatus and method and a recording medium capable of accurately extracting a specific area on an image. About.

[0002]

2. Description of the Related Art Hitherto, as a technique for extracting a characteristic part in an image, a method of extracting the characteristic part from an image using the characteristic has been devised.

For example, when a telop in an image is extracted, the telop is extracted by using those characteristics because the telop has a feature that the luminance value is larger than the background and the edge is steep.

[0004]

However, there is a problem that a telop cannot be accurately extracted when an image having similar luminance levels and sharp edges is included in the background.

[0005] The present invention has been made in view of such a situation, and for example, it is an object of the present invention to accurately extract a telop.

[0006]

According to the present invention, there is provided an image processing apparatus comprising: storage means for storing an input image; extraction means for extracting a characteristic image having a specific feature from the input image stored in the storage means; A feature image extracted by the means,
It is characterized by comprising a detecting means for detecting the movement of the characteristic image based on the input image, and a synthesizing means for synthesizing the characteristic image and the input image based on the movement of the characteristic image by adjusting the positions.

[0007] The storage means can store the image synthesized by the synthesis means.

The synthesizing means includes positioning means for performing positioning between the characteristic image and the input image in accordance with the movement of the characteristic image, and the input image and the characteristic image based on the positioning result by the positioning means. Can be synthesized.

The combining means can perform weighted addition of the input image and the characteristic image for each pixel.

[0010] The characteristic image can move differently from the image around the characteristic image.

[0011] The pixel density of the characteristic image can be higher than the pixel density of the input image.

The pixel density of the feature image can be the same as the pixel density of the input image.

[0013] The detecting means can detect the motion vector based on the characteristic image extracted by the extracting means.

The positioning means can perform positioning of the characteristic image and the input image by shifting the phase of the characteristic image.

The positioning means can perform positioning of the characteristic image and the input image by shifting the phase of the input image.

[0016] The positioning means can perform positioning between the characteristic image and the input image such that the pixel density of the characteristic image is higher than the pixel density of the input image.

The extracting means can output the extracted characteristic image.

The detecting means can output the detected motion as the motion of the characteristic image.

An image processing method according to the present invention includes a storage step of storing an input image, an extraction step of extracting a characteristic image having a specific feature from the input image stored in the processing of the storage step, and a processing of the extraction step. A detecting step of detecting a movement of the feature image based on the extracted feature image and the input image; and a combining step of aligning and combining the feature image and the input image based on the movement of the feature image. It is characterized by the following.

The program of the recording medium according to the present invention includes a storage step of storing an input image, an extraction step of extracting a characteristic image having a specific feature from the input image stored in the processing of the storage step, and a processing of the extraction step. Based on the feature image extracted in the above, based on the input image, a detection step of detecting the motion of the feature image,
A synthesizing step of synthesizing the characteristic image and the input image at the same position.

According to the image processing apparatus and method of the present invention, and a program for a recording medium, an input image is stored, a characteristic image having a specific characteristic is extracted from the stored input image, and the extracted characteristic image is The motion of the feature image is detected based on the input image, and based on the motion of the feature image, the feature image and the input image are aligned and combined.

[0022]

FIG. 1 shows an example of the configuration of an image processing apparatus to which the present invention is applied. This image processing apparatus can extract, from an input image that has been input, an image having a feature that moves differently from the background, such as a telop (hereinafter, referred to as a feature image).

The input image input to the image processing device is supplied to a delay circuit 11 and a motion vector detection circuit 15. The delay circuit 11 converts the supplied input image into
The signal is supplied to the synthesizing circuit 12 after being delayed by the time required for processing in the synthesizing circuit 12 to the phase shift circuit 16 to be described later. As a result, the input image and the corresponding image are synthesized in the synthesis circuit 12 described later.

The synthesizing circuit 12 synthesizes the input image supplied from the delay circuit 11 and the accumulated image supplied from the phase shift circuit 16 and shifted in phase and stored in the accumulation memory 13. Combined image
The data is supplied to the storage memory 13.

The storage memory 13 stores the image supplied from the synthesizing circuit 12, generates a stored image, and supplies the generated image to the extraction circuit 14 and the phase shift circuit 16.

The extraction circuit 14 extracts a characteristic image (more precisely, an image area determined to be a characteristic image) on the stored image supplied from the storage memory 13, and outputs image data of the extracted image area and its display. Move the position to motion vector detection circuit 15
And supply to external equipment.

The motion vector detecting circuit 15 receives the input image and the image data of the image area determined to be the characteristic image on the stored image and the display position thereof from the extracting circuit 14. The motion vector detection circuit 15 uses the image data and the display position supplied from the extraction circuit 14 to store an image area on the storage image extracted by the extraction circuit 14 (on the storage image determined to be a feature image). A motion vector between the image area) and an image area on the input image corresponding to the image area is detected, and the detection result is supplied to the phase shift circuit 16.

The phase shift circuit 16 stores the storage memory 1 based on the motion vector from the motion vector detection circuit 15.
3 is supplied to the synthesizing circuit 12 after shifting the phase of the stored image.

FIG. 2 shows a configuration example of the extraction circuit 14 when a telop is extracted as a feature image.

The stored image supplied from the storage memory 13 is supplied to each of the edge detection circuit 31, the level detection circuit 32, and the telop determination circuit 33.

The edge detection circuit 31 performs edge processing for each predetermined area (image area) of the stored image to detect the steepness of the edge, and outputs the detection result to the telop determination circuit 3.
Supply 3

The level detection circuit 32 detects, for example, a luminance level for each predetermined image area of the input image (an image area on which the edge detection processing is performed by the edge detection circuit 31), and outputs the detection result. This is supplied to the telop determination circuit 33.

The telop judging circuit 33 detects the steepness of the edge from the edge detecting circuit 31 and the level detecting circuit 32
The threshold value determination is performed on the level value from, and the image area on the accumulated image for which both of them have been determined to be higher than the predetermined threshold value is determined to be a telop. Since a telop usually has a high steepness and a high level value, a telop can be detected by determining a threshold value of each value in this way.

The telop determination circuit 33 acquires the image data of the image area determined to be a telop and the display position of the image area from the accumulated image, and
5 and output to an external device.

Although the configuration example of the extraction circuit 14 for extracting a telop has been described above, the extraction circuit 1
4 may have another configuration for extracting an object as a feature image.

Next, the operation of the image processing apparatus will be described with reference to the flowchart of FIG. Here, a case where a telop is extracted as a characteristic image will be described as an example.

In step S1, the motion vector detection circuit 15 extracts the image area determined by the extraction circuit 14 as a telop on the stored image and the input image at the position corresponding to the display position of the image area. A motion vector between image areas on the upper (hereinafter referred to as field A) is detected, and the detection result is supplied to the phase shift circuit 16.

Here, it is assumed that the input of the input image has already been started, the stored image is stored in the storage memory 13, and the image data of the new input image (field A) is input. The extraction circuit 14 obtains the image data of the image area determined to be a telop and the display position thereof from the stored image stored in the storage memory 13 by the method described with reference to FIG. It is supplied to the vector detection circuit 15.

Next, in step S2, as shown in FIG. 4A, the phase shift circuit 16 determines the phase of the pixel of the image area on the stored image determined to be a telop and the display of the image area. The phase of the stored image supplied from the storage memory 13 is shifted based on the motion vector from the motion vector detection circuit 15 so that the phase of the pixel in the image area on the field A at the position corresponding to the position matches. .

By the way, the phase of the pixel of the image area (for example, the background portion) which is not determined to be a telop on the accumulated image and the pixel of the image area on the field A corresponding to the image area are obtained. The phase is, for example, as shown in FIG.
Invert as shown in FIG.

The phase shift circuit 16 supplies the stored image whose phase has been shifted to the synthesizing circuit 12.

In step S3, the synthesizing circuit 12
The field A from the delay circuit 11 and the stored image whose phase has been shifted from the phase shift circuit 16 are combined according to the following equation and supplied to the storage memory 13. Composite value = (pixel value of input image × N + pixel value of accumulated image ×
N + M) N and M are coefficients.

That is, the input image and the accumulated image are weighted and added in pixel units, and the resultant value obtained here becomes the pixel value of the image supplied to the accumulation memory 13.

In the case of the examples of FIGS. 4 and 5, the phase of the pixel of the image area determined to be a telop on the stored image and the phase of the pixel of the image area on the field A corresponding to the image area are different. As shown in FIG. 4A, the phase of the pixel in the image area that is not determined to be a telop in the matched state and in the field A corresponding to the image area is determined. The input image and the accumulated image are combined in a state where the phases of the pixels in the image area are inverted, for example, as shown in FIG.

The storage memory 13 stores the image supplied from the synthesizing circuit 12.

Next, in step S4, the extraction circuit 1
4 reads out the stored image from the storage memory 13, extracts the image area determined to be a telop from the stored image in step S5,
And supply to external equipment. Thereafter, the process returns to step S1, and the subsequent processes are repeatedly executed.

As described above, the phase of the pixel of the image area determined to be a telop (characteristic image) on the accumulated image and the field A (input image) corresponding to the image area
As shown in FIG. 4A, the input image and the stored image are combined so that the phases of the pixels in the upper image region match each other. The distribution (unevenness) is maintained as shown in FIG. However, the phase of the pixel in the image area other than the telop on the input image and the accumulated image is, for example, as shown in FIG.
As shown in FIG. 5A, since the images are combined in an inverted state, after the combination, the level distribution of the image area other than the telop on the accumulated image becomes flat as shown in FIG.

That is, as a result, a higher steepness and a higher level value are detected from the telop as compared with the image area other than the telop, so that the extraction circuit 14 can accurately extract the telop.

In the above description, the case where the input image and the accumulated image are combined so that the phases of the pixels of the input image and the accumulated image match in the telop portion has been described as an example. As shown in (5), in the telop portion, the synthesis can be performed such that the phases of the pixels of the input image and the accumulated image are shifted by a predetermined phase. In this case, the density of the pixels in the telop portion is increased as shown in FIG. On the other hand, the pixel values of the image area other than the telop are averaged similarly, so that the telop can be accurately extracted as in the case described above.

In this example, the pixels of the characteristic image are
Since the density is increased, the density differs from the pixel density of the input image. Therefore, the motion vector detection circuit 15 detects a motion vector by the method described below.

Here, one image (the image corresponding to the input image) Pa shown in FIG. 7A and the image Pa shown in FIG. The following describes an example in which a motion vector is detected between an image (an image corresponding to a feature image) Pb and the image number (pixel number). In addition,
In FIG. 7, a solid line represents a so-called line, in which pixels (not shown) are arranged. On the other hand, in the figure, the dotted line
It divides the line interval into four equal parts, and there are no pixels there.

First, a high-density image Pb (FIG. 7)
(C)) shows an image Pb1 whose line position matches the image Pa, an image Pb2 whose line position is shifted downward by one line with respect to the image Pb1, and the image P
Image P whose line position is shifted down by one line with respect to b2
The image is decomposed into b3 and an image Pb4 whose line position is shifted downward by one line with respect to the image Pb3.

Next, a reference block Ba (5 × 5) as shown in FIG.
(A position corresponding to the display position of the characteristic image). Further, as shown in FIG. 8B, a reference block Bb having the same size and the same shape as the reference block Ba is, for example, on the image Pb1 or the image Pab among the images Pb1 to Pb4. Is set at the same position as the reference block Ba set to (hereinafter referred to as reference position).

Next, the difference (absolute value) between the pixel value of the pixel of the image Pa in the reference block Ba and the pixel value of the pixel of the image Pb1 in the reference block Bb set at the reference position is calculated. An absolute value sum (total value) is calculated. Then, as shown in FIG. 9, the reference block Bb is arranged on the image Pb1 in the left-right direction and the up-down direction around the reference position.
Each pixel is sequentially moved, and the absolute value sum of the difference between the pixel value of the pixel of the image Pb1 in the reference block Bb and the pixel value of the pixel of the image Pa in the reference block Ba at the moving position is calculated. You. Then, a table is generated in which the calculated absolute value sums are arranged at positions corresponding to the movement positions of the reference block Bb.

For example, when the reference block Bb is shifted from the reference position by two pixels in the left-right direction and by one pixel in the vertical direction, 5 × 3 sums of absolute values are calculated.
A table T1 as shown in FIG. 10A is generated. For example, the absolute value sum X0 of the table T1 is obtained by referring to the reference block Bb.
Is the sum of absolute values calculated when the reference position is set at the reference position (the same position as the reference block Ba). That is, the arrangement position of the absolute value sum X0 is a position corresponding to the center pixel of the reference block Ba.

As described above, the reference block B is added to the image Pb1.
When b is set and the table T1 is generated, the reference block Bb is set to the images Pb2 to Pb4, respectively, and the tables T2 to T4 in each case are generated.

In the table T2, each sum of absolute values calculated when the reference block Bb is set in the image Pb2 is calculated by adding 1 to the position of the sum of absolute values arranged in the table T1.
They are arranged at the bottom of the column.

In the table T3, each sum of absolute values calculated when the reference block Bb is set in the image Pb3 is calculated by adding 1 to the position of the sum of absolute values arranged in the table T2.
They are arranged at the bottom of the column. In the table T4, the sums of absolute values calculated when the reference block Bb is set in the image Pb4 are arranged one column below the positions of the sums of absolute values arranged in the table T3. ing.

Next, the table T generated as described above
1 to T4 are combined to generate a table T0 as shown in FIG. That is, 5 × 3 × 4 absolute value sums are arranged in the table T0.

Next, the minimum value of the sum of absolute values constituting the table T0 is detected, and the reference block Bb to which the sum of absolute values is given is detected. Then, from the position of the center pixel of the reference block Ba (the pixel of the image Pa) to the position of the center pixel (the center pixel of any of the images Pb1 to Pb4) of the reference block Bb to which the absolute value sum is given. The vector is detected as a motion vector.

According to the method described above, a motion vector can be detected even when the information densities of images are different. This method is disclosed in Japanese Patent Application Laid-Open No. H11-034814 filed earlier by the present applicant.

Although the phase of the stored image is shifted in the example of FIG. 1, the phase of the input image can be shifted. FIG. 11 shows an example of the configuration of an image processing apparatus in a case where the phase of an input image is shifted instead of an accumulated image.

In this image processing apparatus, the delay circuit 11 shown in FIG. 1 is removed, and the input image is supplied to the motion vector detection circuit 15 and the phase shift circuit 16.
The storage memory 13 supplies the stored image to the synthesizing circuit 12 instead of the phase shift circuit 16.

The operation of the image processing apparatus shown in FIG. 11 will be described with reference to the flowchart shown in FIG. Also here,
A case where a telop is extracted as a characteristic image will be described as an example.

In step S 11, the motion vector detection circuit 15 extracts the image area on the stored image extracted as the telop extracted by the extraction circuit 14 and the image area on the input image corresponding to the image area. A motion vector between them is detected, and the detection result is supplied to the phase shift circuit 16.

Next, in step S12, the phase shift circuit 16 shifts the phase of the input image based on the motion vector from the motion vector detection circuit 15.

As a result, the phase of the pixel of the image area on the stored image determined to be a telop and the phase of the pixel of the image area on the input image corresponding to the image area are shown in FIG.
As shown in (A), the phase of a pixel of an image area (for example, a background portion) on the accumulated image that is not determined to be a telop on the accumulated image and the input image corresponding to that image area The phase of the pixel in the image area is, for example, as shown in FIG.
Invert as shown in FIG.

In step S13, the combining circuit 12
Synthesizes the input image supplied from the phase shift circuit 16 and shifted in phase with the stored image from the storage memory 13 and supplies the synthesized image to the storage memory 13. Storage memory 13
Accumulates the image supplied from the synthesizing circuit 12.

In steps S14 and S15, the same processes as those in steps S4 and S5 in FIG. 3 are performed, so that the description will be omitted.

Also in the case of this example, the level distribution of the pixel values of the pixels is maintained or the density of the feature image is increased, and the pixel values of the pixels are flattened, for example, an image of the background portion. Since the stored image is supplied to the extraction circuit 14, the extraction circuit 14 can accurately extract the characteristic image.

Further, in the example of FIG. 1, only one extraction circuit 14 is provided at the subsequent stage of the storage memory 13.
As shown in FIG. 3, another extraction circuit 21 can be provided in place of the delay circuit 11 of FIG.

The extraction circuit 21 has the same configuration as the extraction circuit 14 and supplies, for example, an image area determined as a telop from the input image to the synthesis circuit 12.

The operation of the image processing apparatus of FIG. 13 will be described with reference to the flowchart of FIG. Also here,
A case where a telop is extracted as a characteristic image will be described as an example.

In step S21, the same processing as in step S1 of FIG. 3 is performed, and a description thereof will be omitted.

In step S22, the phase shift circuit 16 shifts the phase of the stored image based on the motion vector from the motion vector detection circuit 15.

As a result, the phase of the pixel in the image area on the stored image determined to be a telop by the extraction circuit 14 and the phase of the pixel in the image area on the input image determined to be a telop by the extraction circuit 21 Match as shown in FIG.

In step S23, the synthesizing circuit 12
Synthesizes the image area on the input image determined as the telop extracted by the extraction circuit 21 with the accumulated image whose phase has been shifted, supplied from the phase shift circuit 16, and The data is supplied to the storage memory 13. The storage memory 13 stores the image supplied from the synthesizing circuit 12.

In steps S24 and S25, the same processes as those in steps S4 and S5 in FIG. 3 are performed, and the description thereof will be omitted.

In the above, the case where a telop is extracted as a characteristic image has been described as an example. However, an object having a different motion from the background may be used.

The series of processes described above can be realized by hardware, but can also be realized by software. When a series of processing is realized by software, a program constituting the software is installed in a computer, and the program is executed by the computer, so that the above-described image processing apparatus is functionally realized.

FIG. 15 is a block diagram showing the configuration of an embodiment of the computer 101 functioning as the above-described image processing apparatus. CPU (Central Processing Uni
t) 111 is connected to an input / output interface 116 via a bus 115, and the CPU 111 allows the user to input a keyboard,
When a command is input from an input unit 118 composed of a mouse or the like, for example, a magnetic disk 131, an optical disk 132, or a magneto-optical disk 1 mounted on a ROM (Read Only Memory) 112, a hard disk 114 or a drive 120.
33 or a program stored in a recording medium such as the semiconductor memory 134 is stored in a random access memory (RAM).
y) Load into 113 and execute. Thus, the various processes described above (for example, the processes shown by the flowcharts in FIGS. 3, 12, and 14) are performed. In addition, CPU1
The output unit 11 outputs the processing result to an output unit 117 such as an LCD (Liquid Crystal Display) via the input / output interface 116 as necessary. The program is stored on the hard disk 114 or the ROM 112.
In advance, and can be provided to the user integrally with the computer 101, or can be stored in the magnetic disk 131 and the optical disk 1.
32, a magneto-optical disk 133, a semiconductor memory 134, or the like, or a hard disk 114 via a communication unit 119 from a satellite, a network, or the like.
Can be provided.

In the present specification, the steps of describing a program provided by a recording medium may be performed not only in chronological order but also in chronological order according to the described order. This also includes processing executed in parallel or individually.

[0083]

According to the image processing apparatus and method of the present invention and the program of the recording medium, the input image is stored,
A feature image having a specific feature is extracted from the stored input image, a movement of the feature image is detected based on the extracted feature image and the input image, and a feature image and an input image are detected based on the movement of the feature image. Since the positions are combined and synthesized, the characteristic image can be extracted with high accuracy.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration example of an image processing apparatus to which the present invention has been applied.

FIG. 2 is a block diagram showing an extraction circuit 14 of FIG.

FIG. 3 is a flowchart illustrating an operation of the image processing apparatus of FIG. 1;

FIG. 4 is a diagram showing a phase shift circuit 16 and a synthesis circuit 12 shown in FIG. 1;
It is a figure explaining operation of.

5 is a phase shift circuit 16 and a synthesis circuit 12 shown in FIG.
FIG. 6 is another diagram for explaining the operation of FIG.

FIG. 6 shows the phase shift circuit 16 and the synthesis circuit 12 shown in FIG.
FIG. 6 is another diagram for explaining the operation of FIG.

FIG. 7 is a diagram illustrating the operation of the motion vector detection circuit 15 of FIG.

FIG. 8 is a diagram illustrating a reference block Ba and a reference block Bb.

FIG. 9 is a diagram illustrating movement of a reference block Bb.

FIG. 10 is a diagram illustrating a table T.

FIG. 11 is a block diagram illustrating another configuration example of an image processing apparatus to which the present invention has been applied.

FIG. 12 is a flowchart illustrating an operation of the image processing apparatus in FIG. 11;

FIG. 13 is a block diagram illustrating another configuration example of an image processing apparatus to which the present invention has been applied.

FIG. 14 is a flowchart illustrating an operation of the image processing apparatus in FIG. 13;

FIG. 15 is a block diagram illustrating a configuration example of a computer 101.

[Explanation of symbols]

11 delay circuit, 12 synthesis circuit, 13 storage memory, 14 extraction circuit, 15 motion vector detection circuit, 16 phase shift circuit, 21 extraction circuit

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Takayoshi Fujiwara 6-35, Kita-Shinagawa, Shinagawa-ku, Tokyo Sony Corporation F-term (reference) 5C023 AA11 CA05 EA03 EA13 5L096 AA06 CA04 EA35 FA06 FA37 FA44 FA69 HA02 HA07 JA09 JA14

Claims (15)

[Claims] A storage unit configured to store an input image; an extraction unit configured to extract a feature image having a specific feature from the input image stored in the storage unit; and a feature image extracted by the extraction unit. Detecting means for detecting the movement of the characteristic image based on the input image; and synthesizing means for synthesizing the characteristic image and the input image based on the movement of the characteristic image by aligning the positions. An image processing apparatus characterized by the above-mentioned. 2. The image processing apparatus according to claim 1, wherein the storage unit stores an image synthesized by the synthesis unit. 3. The synthesizing unit includes a positioning unit that performs positioning between the characteristic image and the input image in accordance with the movement of the characteristic image, and based on a result of the positioning performed by the positioning unit. hand,
The image processing apparatus according to claim 1, wherein the input image and the feature image are combined. 4. The image processing apparatus according to claim 1, wherein the combining unit performs weighted addition of the input image and the feature image for each pixel. 5. The image processing apparatus according to claim 1, wherein the feature image moves differently from an image around the feature image. 6. The image processing apparatus according to claim 1, wherein a pixel density of the feature image is higher than a pixel density of the input image. 7. The image processing apparatus according to claim 1, wherein the pixel density of the feature image is the same as the pixel density of the input image. 8. The image processing apparatus according to claim 1, wherein the detecting unit detects a motion vector based on the feature image extracted by the extracting unit. 9. The image processing apparatus according to claim 3, wherein the positioning unit performs the positioning of the characteristic image and the input image by shifting a phase of the characteristic image. 10. The apparatus according to claim 3, wherein said positioning means performs positioning of said characteristic image and said input image by shifting a phase of said input image.
An image processing apparatus according to claim 1. 11. The apparatus according to claim 1, wherein the positioning unit performs positioning of the feature image and the input image such that a pixel density of the feature image is higher than a pixel density of the input image. 4. The image processing device according to 3. 12. The image processing apparatus according to claim 1, wherein the extraction unit outputs the extracted characteristic image. 13. The image processing apparatus according to claim 1, wherein the detection unit outputs the detected motion as the motion of the feature image. 14. A storage step for storing an input image;
An extracting step of extracting a feature image having a specific feature from the input image stored in the processing of the storing step, and the feature image extracted in the processing of the extracting step;
A detection step of detecting a movement of the feature image based on the input image; and a combining step of combining the feature image and the input image at the same position based on the movement of the feature image. Characteristic image processing method. 15. A storage step of storing an input image, an extraction step of extracting a characteristic image having a specific feature from the input image stored in the processing of the storage step, and an extraction step performed in the processing of the extraction step. The feature image;
A detection step of detecting a movement of the feature image based on the input image; and a combining step of combining the feature image and the input image at the same position based on the movement of the feature image. A recording medium on which a program for causing a computer to execute a characteristic process is recorded.