JP2001057630A - Image processing unit and image processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing unit that automatically separates an image into a major part and a background and applies blurring process only to the background area without the need for an operator to execute a troublesome work to designate a border line between the major part and the background. SOLUTION: An area division means 74 divides an image into areas, calculates each feature vector of each area, a connection extension means 76 estimates a background area on the basis of the calculated feature vector and a separate means 78 divides the image into two; a background area and a major part area. Then a mask generating means 80 overlaps a contour extract image resulting from extracting a contour of the original image by a contour extract means and an image of the major part area separated from the original image by the separation means 78 to generate a mask for the major part and a blur processing means 82 applies blur processing to the image.

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 an image processing method, and more particularly, to an image processing apparatus and the like for performing image processing on an image captured by a digital still camera or the like and printing the image. The present invention relates to a processing device and an image processing method.

[0002]

2. Description of the Related Art In general, the size of a light receiving surface of a CCD sensor, which is an image pickup device of a digital still camera, is smaller than a 35 mm film for silver halide photography, for example, and the lens focal length of a digital still camera is shorter than that of a 35 mm camera. Therefore, even if the image is photographed with the same angle of view and the same aperture, the digital still camera optically has a deeper depth of field than the 35 mm camera. In general, the aperture of a digital still camera has a narrow variable range and an open F number is often darker than that of a 35 mm camera lens. Therefore, it is often impossible to open the aperture to reduce the depth of field. As a result, in particular, a scene photographed by a digital still camera tends to be like a pan-focused image in which the entire image is in focus.

[0003] However, depending on the scene, there is naturally a case where it is desired to make only the main subject stand out and blur the background before and after. Therefore, it is necessary to easily create a print that has such an effect by image processing, even for images taken with a camera that cannot reduce the depth of field by adjusting the lens focal length and aperture of the camera. It becomes.

[0004] Japanese Patent Application Laid-Open No. H10-65923 describes the following technology. First, the operator specifies a temporary boundary between the main part and the background, performs expansion and contraction processing on the image data based on the specified temporary boundary, and performs an area including the main part, an area including only the background, and a main part. It is divided into three regions consisting of a region including an accurate boundary between the part and the background. After that, for each pixel included in the area including the accurate boundary between the main part and the background, it is determined whether the pixel belongs to the area including only the main part or the area including only the background, for example, color continuity. And an accurate boundary line between the main part and the background is detected. Then, sharpness processing is performed on the digital image data of the area of the main part and the area including the accurate boundary between the main part and the background, and the edge of the accurate boundary between the main part and the background is emphasized without discomfort. You. In addition, a boundary line obtained by slightly expanding an accurate boundary line between the main part and the background is calculated. Based on the expanded boundary line, the digital image data of the background region is subjected to interpolation processing, and the boundary line is obtained. To prevent the occurrence of bleeding in a portion close to. Thereafter, the background area is subjected to a blurring process using an IIR filter or the like. As described above,
According to the technique described in Japanese Patent Application Laid-Open No. 0-65923, even if an image is captured by a camera that cannot reduce the depth of field, it is possible to easily emphasize only the main subject by image processing and blur the background before and after. it can.

[0005]

However, in the technique described in Japanese Patent Application Laid-Open No. H10-65923, it is necessary for an operator to specify a temporary boundary between the main part and the background, and to specify the temporary boundary. There is a problem that the work is complicated because the work is specified using a mouse or the like.

The present invention has been made to solve the above problem, and automatically separates the main part from the background without the operator having to perform a complicated operation to specify the boundary between the main part and the background. It is another object of the present invention to provide an image processing apparatus and an image processing method capable of performing a blurring process only on a background area.

[0007]

In order to achieve the above object, the present invention divides an image into regions so that adjacent pixels having a data difference between pixels equal to or less than a threshold value are included in the same region,
Calculating means for calculating a characteristic amount for each area; and area estimating means for estimating a background area or a main part area from among the divided areas based on the characteristic amount calculated by the calculating means, And a blur processing means for selectively performing blur processing only on the background area estimated by the area estimating means from the entire image or selectively performing blur processing only on an area excluding the main part area.

According to the first aspect of the present invention, adjacent pixels whose data (for example, color information, luminance information, color difference information, etc.) between pixels are equal to or less than a threshold value are included in the same region by the calculating means. The image is divided into regions. Then, a feature amount (for example, an average luminance, an average color difference, an average coordinate position, and the like) for each region is calculated, and the background region or the main portion region is selected from the divided regions by the region estimation means based on the calculated feature amount. Is estimated. For example, a predetermined position in an image is set as a background region, and the feature amount of the background region is compared with the feature amount of another divided region adjacent to the background region to estimate whether or not the image is the background region. The background region can be estimated by sequentially comparing the feature amounts of the divided background region and the adjacent divided region.

Further, based on the background area or the main part area estimated by the area estimating means, the blur processing means selectively performs the blur processing only on the area excluding the estimated background area or the main part area. That is, it is possible to automatically perform the blurring process only on the background area without specifying the boundary line between the main part area and the background area.

According to a second aspect of the present invention, in the first aspect of the present invention, the blur processing means includes a step of setting the entire image to be a region excluding a background region and a background region, or a region excluding a main region and a main region. Separating means, a mask generating means for generating a mask in a region other than the background region separated by the separating means or a main part region, and an area other than the mask area generated by the mask generating means. And a processing means for performing a blurring process.

According to the invention described in claim 2, according to claim 1,
In the invention described in 5, the blur processing means is a separation means,
The background area or main area estimated by the area estimating means is composed of a mask generating means and a processing means. Divided. Then, a mask is generated by a mask generation unit in an area or a main part area other than the separated background area, and based on the generated mask,
The blurring process can be performed on the area other than the mask by the processing means.

According to a third aspect of the present invention, in the first or second aspect of the invention, the area estimating means sets at least one of four corners of the image as a background area, When the feature amount and the feature amount of the adjacent region are equal to or smaller than a predetermined threshold, the adjacent region is integrated as a background region.

According to the third aspect of the present invention, the first aspect is provided.
Alternatively, in the invention described in claim 2, the background region in the normal image is at least one of the four corners of the image. Therefore, the area estimating means determines that at least one of the four corner areas of the image is a background area and is adjacent when the feature quantity of the area adjacent to the feature quantity of the background area is equal to or smaller than a predetermined threshold value. Let the area be the background area. Therefore, the background area can be estimated based on the predetermined threshold.

According to a fourth aspect of the present invention, in the first aspect of the present invention, the area estimating means is configured to determine the area of the divided area in advance. When the value is equal to or more than 1, or when the value is equal to or less than a predetermined second value, the area is set as a background area.

According to the invention described in claim 4, according to claim 1 of the present invention,
In the invention according to any one of claims 3 to 3, when the area calculated by the calculation means is equal to or larger than a predetermined first value, that is, when the size of the divided region is very large, Alternatively, when the area of the region divided by the calculation unit is equal to or larger than a predetermined second value, that is, when the size of the region divided is very small, the region is set as a background region. Thus, the background area can be estimated.

According to a fifth aspect of the present invention, in the first aspect of the present invention, the area estimating means determines that an average coordinate position of a pixel in the divided area is It is characterized in that a region located in a region outside the circle or the ellipse smaller than the image based on the center of the image is set as the background region.

According to the invention described in claim 5, according to claim 1,
In the invention described in any one of the fourth to fourth aspects, in the case of a normal image, the main subject is located substantially at the center of the image. Therefore, the background area is estimated by setting, as a background area, an area in which the average coordinate position of the pixels of the area divided by the calculation means is located in an area outside a circle or an ellipse smaller than the image based on the center of the image. be able to.

According to a sixth aspect of the present invention, in the second aspect of the present invention, the mask generating means extracts a contour in the image, and extracts a contour of the main portion area. , So that the contour of the mask is adjusted.

According to the invention of claim 6, according to claim 2,
In the invention according to any one of claims 5 to 5, when a mask is generated in a main part region for performing the blurring process by the mask generation unit, the outline of the image is separated from the extracted outline image by the separation unit. The superimposition of the main part region thus overlapped makes it possible to adjust the outline of the main part to be masked, and to reliably mask the main part when performing the blurring process on the background area.

According to a seventh aspect of the present invention, in the first aspect of the present invention, the calculating means emphasizes a contour in the image, and calculates the contour based on the emphasized contour. And performing the area division.

According to the invention of claim 7, according to claim 1,
In the invention according to any one of claims 5 to 5, the outline in the image is emphasized, and the region is divided by the calculating unit based on the emphasized outline image, so that the boundary line of each region becomes clear. . Therefore, the boundary when the background region and the region other than the background region or the main region and the region other than the main region are separated by the separating means is also clear, and the accuracy of the boundary line between the background region and the main region is improved. Can be improved.

According to an eighth aspect of the present invention, there is provided any one of the first to seventh aspects of the present invention, further comprising a blur processing strength designating means for designating the strength of the blur processing. And

According to the invention of claim 8, according to claim 1,
In the invention according to any one of claims 7 to 7, the image processing apparatus further includes a blur processing intensity designating unit, whereby it is possible to set the intensity of the blur processing performed by the blur processing unit.

According to a ninth aspect of the present invention, in any one of the first to eighth aspects of the present invention, there is further provided switching means for switching whether or not to perform the blurring processing. .

According to the ninth aspect of the present invention, the first aspect is provided.
In the invention according to any one of claims 8 to 10, further comprising a switching means makes it possible to select whether or not to perform a blurring process.

According to a tenth aspect of the present invention, an image is divided into regions so that adjacent pixels having a data difference between pixels equal to or less than a threshold value are included in the same region, and a characteristic amount for each of the divided regions is provided. Based on the calculated feature amount, a background area or a main part area is estimated from the divided areas, and only the background area estimated from the entire image is subjected to selective blur processing. Alternatively, only the region other than the main region is selectively blurred.

According to the tenth aspect of the present invention, the image is area-divided such that adjacent pixels whose pixel data (for example, color information, luminance information, color difference information, etc.) are equal to or less than the threshold value are included in the same area. To divide. Then, a feature amount (for example, average luminance, average color difference, average coordinate position, etc.) for each region is calculated, and a background region or a main portion region is estimated from the divided regions based on the calculated feature amount. . For example, a predetermined position in an image is set as a background region, and the feature amount of the background region is compared with the feature amount of another divided region adjacent to the background region to estimate whether or not the image is the background region. The background region can be estimated by sequentially comparing the feature amounts of the divided background region and the adjacent divided region.

Then, based on the estimated background region or main region, only the region other than the background region or main region is subjected to selective blurring. By performing the blurring processing as described above, it is possible to automatically perform the blurring processing only on the background area without specifying the boundary line between the main part area and the background area.

[0029]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which the present invention is applied to a multicolor thermal printer for recording an image on a multicolor thermal sheet will be described in detail with reference to the drawings.

As shown in FIG. 1, a printer 10 is a device for thermally recording a color image on a substantially rectangular multicolor thermal recording sheet 12 and fixing the recorded image with light to form an image. It is stored in the casing 14.

On one side of the casing 14, an insertion opening 18 for the multicolor thermal recording sheet 12 to which the mounting table 16 is mounted is provided. On the other side of the casing 14, a mounting table 20 is mounted. Multicolor thermal recording sheet 1
A second outlet 22 is provided. And the insertion port 18
A transport device that transports the multicolor thermal recording sheet 12 in the direction of the insertion port 18 or the discharge port 22 is disposed between the discharge port 22 and the discharge port 22.

The transport device includes a pair of transport rollers 24 provided near the insertion port 18 and a pair of transport rollers 32 provided near the discharge port 22.
A guide plate 26 is provided downstream of the transport roller 24 in the transport direction, and a guide plate 38 is provided upstream of the transport roller 32 in the transport direction.

A thermal head 28 as a thermal recording means and a platen roller 30 are provided between the guide plates 26 and 38 so as to face each other. On the side of the transport roller 24 of the thermal head 28, a sensor 27 for detecting the front and back of the multicolor thermal recording sheet 12 and a light source 34 are arranged. Light source 3
As 4, an LED can be used and is used for detecting the multicolor thermosensitive recording sheet 12 by irradiating the conveyed multicolor thermosensitive recording sheet 12 with light having a wavelength of 300 to 450 nm, preferably 420 nm. In addition, it is also used as a fixing light source for fixing each color forming layer.

A control circuit 40 shown in FIG. 2 is housed in the casing 14. The control circuit 40 includes a main controller 42 which is configured by a microcomputer and to which a signal processing unit 44 is connected. The signal processing unit converts image data consisting of RGB (red, green, blue) color signals into a luminance signal Y and a color difference signal Cb,
A signal converter 50 having a function of converting into image data composed of Cr and a RAM 52 for temporarily storing the image data output from the signal converter 50 are provided.

A digital camera 46 or a computer 48 is selectively connected to the signal processing section 44. The digital camera 46 outputs a luminance signal Y and color difference signals Cb and Cr.
Digital image data of a natural image consisting of
Computer-generated RG from computer 48
Computer graphics (CG) consisting of B color signals
Digital image data of any one of an image, a composite image obtained by combining a CG image and a natural image, a composite image obtained by combining a text image and a natural image, and a composite image obtained by combining a template image and a natural image is input. Is done. Note that the text image and the template image can be configured by CG images.

The signal processing unit 44 sends the R
When image data composed of GB color signals is input,
The signals are converted into a luminance signal Y and color difference signals Cb and Cr. When image data including image data composed of Cb, Cr, and Y signals input from a digital camera is input,
The data is temporarily stored in the RAM 52 without being converted into Cb, Cr, and Y signals.

The main controller 42 includes the printer 1
An operation unit 55 for performing the operation of the printer 10 is provided.
Is configured to control the operation. Any one of the Y, M, and C color data is output to the output side of the main controller 42 based on a signal from the main controller.
A driver 58 for driving the thermal head 28 is connected via a switch 58 for selecting one data and a line buffer 60, and a sensor 27 for receiving the reflected light of the light emitted from the light source 34 is further connected. A driver 66 for driving a rotation drive source 64 for rotating the platen roller 30 and a driver 68 for driving the light source 34 and the fixing light source 36 are connected.

The thermal head 28 is provided with a heating element 70 connected to a power supply 72 near the platen roller 30.

Further, the main controller 42 is configured to be capable of performing a blurring process on only the background region of the input image (described later). In addition,
The instruction as to whether or not to perform the blurring process is made when the operator inputs information indicating whether or not to perform the blurring process to the operation unit 55. For example, it is possible to provide a blurring processing changeover switch in the operation unit 55, and to select whether or not to perform the blurring processing by switching the switch.

Next, the above-described background blurring processing will be described.

The above-described blurring process is executed by the main controller 42 by software, which will be described later. When this blurring process is represented by functional blocks, as shown in FIG. It can be represented by an extraction unit 84, a separation unit 78, a mask generation unit 80, and a blur processing unit 82. The area dividing means 74 and the connection expanding means 76 correspond to the calculating means and the area estimating means of the present invention, respectively.

Assuming that the image data representing the image is given by the luminance signal Y and the color difference signals Cb and Cr, the area dividing means 74 performs a raster scan of the two-dimensional data of the luminance signal Y while performing the luminance difference between adjacent pixels. Is a predetermined threshold (Δ
Y) or less and the hue (Cb / Cr distance (ΔC) or Cb / Cr hue angle difference (ΔH)) is equal to or less than a predetermined threshold, and are integrated as the same area, and this condition is not satisfied Is divided into a new area. Then, such processing is sequentially performed, and area division is performed while labeling each area. Note that R, G, and B may be used instead of the luminance signal Y and the color difference signals Cb and Cr.

Further, after the area is divided, the area dividing means 74 calculates an area (number of pixels), an average density, an average Cb, an average Cr, and an average coordinate position (average X, average Y) for each area distinguished by the label. ) Is calculated as a feature amount for each region, and the feature amount is output to the connection extending unit 76 together with the image data obtained by the region division.

In the image divided by the region dividing means 74, the connection expanding means 76
A region including at least one of the corners is set as a reference of the background region. If the four corner regions or the region adjacent to the region including at least one of the four corners satisfies any of the following conditions, the background region And The reference of the background region may be a region outside the ellipse smaller than the image with the center of the original image as the center of gravity.

1. 1. When the difference between the background area and the average luminance, the average Cb, and the average Cr is equal to or less than a predetermined threshold value separately determined A predetermined area Sma separately determined by the area (number of pixels) of the region
2. If it is more than x, or if it is less than Smin. When the average coordinate position is outside a circle or an elliptical region of a predetermined size separately determined based on the center of the image, the connection expanding unit 76 connects and expands the region satisfying the above conditions as a background region, Output to

The contour extracting means 84 extracts a contour line of the original image by using the original image to create a contour extracted image. In the contour extraction, the contour can be emphasized by a general differential filter such as a first derivative (gradient) and a second derivative (Laplacian), and finally the contour is extracted by a boundary line extracting filter called a Sobel filter. Can create a contour-extracted image with a thin line.

The separation means 78 divides the entire image into two parts, a background area and a main part area, by setting the area remaining unabsorbed in the background area by the connection expanding means 76 as the main part area.

The mask generation unit 80 superimposes the main region separated by the separation unit 78 and the contour extraction image obtained by the outline extraction unit 84 to adjust the outline of the main region. Then, an image mask overlapping with the main area is generated.

The blur processing means 82 performs a blur process on an area other than the above-mentioned mask. As the blurring process, a low-pass filter for noise reduction is suitable. As the simplest smoothing filter, the value of a predetermined pixel is replaced with a calculated value using a filter matrix for the values of 3 × 3 surrounding pixels. Processing can be used. For example,

[0050]

(Equation 1)

The following filter matrix can be used. Note that a median filter that has a noise reduction effect while maintaining sharpness and outputs an intermediate value (median) in a local region may be used as slight blurring processing.

The intensity of the blur can be adjusted by changing the value of the filter matrix element. If a filter matrix corresponding to several levels of the blur intensity is set, the blur intensity can be changed according to the user's specification. Can be. For example, the inside of the blur mask area may have a uniform blur intensity, or may change in accordance with the in-screen distance from the center of the main part area (the greater the distance, the greater the blur intensity).

It should be noted that the blur intensity of the blur process may be a constant blur intensity or a plurality of levels of blur intensity may be set. If it is possible to set several levels of blur intensity, a filter matrix corresponding to several levels of blur intensity is set, and among several levels of blur intensity,
The operator can change the blur intensity by inputting information indicating the change in the blur intensity to the operation unit 55. For example, it is possible to change the blur intensity by providing a blur intensity selection switch or the like on the operation unit 55.

Next, a description will be given of software for processing for blurring the operation in the present embodiment.

When the operator operates the operation unit 55 and inputs information for executing the blurring process, the blurring process is performed.

Next, the image data once stored in the RAM 52 is taken into the main controller 42, and the taken image data is automatically blurred. In addition,
The blurring process is performed according to the flowchart shown in FIG. Hereinafter, the blurring process will be described with reference to the flowchart of FIG. 4 and FIG.

FIG. 5A shows an original image in which a rose flower before the blurring process is used as a subject in the main area. When information for performing the blurring process is input to the operation unit 55, the area dividing unit 74 in step 100
The region division processing of the original image shown in FIG. 5A is performed, and a region division image as shown in FIG. 5B is created.

In step 102, when the area adjacent to the four corner areas of the image shown in FIG.
The background region is estimated by sequentially expanding and connecting the background region. Then, the separation unit 78 sets the region left unabsorbed as the background region by the connection expansion unit 76 as the main region, divides the image into two regions, the background region and the main region, and extracts the main region ( (FIG. 5 (C)).

In step 104, the contour extracting means 84 extracts a contour from the original image shown in FIG. 5A, and generates a contour extracted image shown in FIG. 5D.

In the following step 106, the main part area extracted by the separating means 78 and the contour extraction image obtained by the contour extracting means 84 are overlapped, and the contour closest to the contour of the main part area is selected from the contour extracted image. By doing so, the contour of the main area is adjusted. Then, a blur mask is generated using the main portion region whose contour has been adjusted as a mask (FIG. 5E).

In step 108, the blur processing is performed by the blur processing means 82 using the generated blur mask, and an image in which only the background region is blurred as shown in FIG. 5F can be obtained. In other words, only the main part, such as a picture taken with a 35 mm single-lens reflex camera with a priority given to the aperture and a shallower depth of field, is taken from an image in which the entire screen is focused in pan focus taken with a digital still camera by blurring You can create prints that are sharp and have a blurred background.

As described above, the operator simply inputs information indicating whether or not to perform the blurring process on the operation unit 55 without performing a complicated operation of designating the boundary between the main part area and the background area. , Can automatically perform the blurring process.

Next, another configuration for the above-described blurring process will be described with reference to FIG.

As shown in FIG. 6, similarly to the above-described blurring processing, the image processing apparatus includes a region dividing means 74, a link expanding means 76, a contour emphasizing means 86, a separating means 78, a mask generating means 80, and a blurring processing means 82. I have.

In the above-described blurring processing, the contour extracting means 84 is provided for adjusting the contour of the mask when the mask generating means generates the blurring mask, whereas the area dividing line for dividing the area is adjusted. An outline emphasis means 86 is provided. That is, the configuration is such that the outline of the original image is emphasized, and the area is divided for the image whose outline is enhanced. By doing so, the boundary line becomes clear, and the accuracy of the boundary line between the background region and the main region after the region division can be improved.

Next, the blur processing configured as described above will be described with reference to the flowcharts of FIGS. 6 and 7 and FIG.

FIG. 8A shows an original image in which a person before blurring processing is a subject in a main part area. In this flowchart, when an instruction to perform the blurring process is input to the operation unit 55, the outline emphasis unit 86 performs outline emphasis on the original image shown in FIG. An outline emphasized image is generated.

In step 202, based on the contour line whose contour has been emphasized by the contour emphasizing means 86, the area dividing means 7
4 is performed, and a region-divided image shown in FIG. 8C is created. By performing the region division processing on the image on which the contour enhancement has been performed, the accuracy of the boundary between the background region after the region division and the main subject region can be improved.

In step 204, the area divided image (FIG. 8)
Based on (C)), if the area adjacent to the four corner areas satisfies any of the above-described conditions, the area is set as a background area, and the background area is estimated by sequentially connecting and expanding the background areas. The main area is extracted by dividing the area that has not been absorbed as the background area into the main area and dividing the region-divided image into the background area and the main area (FIG. 8).
(D)).

In step 206, a blur mask is generated using the extracted main area as a mask (FIG. 8).
(E)).

In step 208, a blur process is performed using the generated blur mask, and an image in which only the background region is blurred as shown in FIG. 8F can be obtained.

In the above description, the mask is generated in the main area by connecting and expanding the background area and separating the background area from the image data. However, the main area other than the background area is connected and expanded to form the mask. A partial region may be generated, and a mask may be generated from the generated main region.

In the above description, the luminance signal Y and the color difference signal C
Although the blur processing is performed using the signal composed of b and Cr, the blur processing may be performed using the RGB color signals.

Further, in the present embodiment, the present invention has been described by taking the printer 10 as an example. However, the present invention is not limited to the printer 10, and may be applied to, for example, a digital still camera, The present invention may be applied to a digital lab system that processes and prints.

[0075]

As described above, according to the present invention, the main part and the background can be automatically separated from each other without the operator having to perform a complicated operation for specifying the boundary between the main part and the background.
There is an effect that it is possible to provide an image processing apparatus and an image processing method capable of performing a blurring process only on a background region.

[Brief description of the drawings]

FIG. 1 is a schematic diagram of a printer according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a control circuit of the printer according to the embodiment of the present invention.

FIG. 3 is a block diagram illustrating a configuration of a blurring process in special image processing.

FIG. 4 is a flowchart illustrating an example of the flow of a blurring process.

FIG. 5 is a diagram illustrating a process of an example of a blurring process;

FIG. 6 is a block diagram showing another configuration of the blurring process.

FIG. 7 is a flowchart illustrating another example of the blurring process.

FIG. 8 is a diagram showing a process of another example of the blurring process.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Printer 42 Main controller 74 Area division means 76 Connection expansion means 78 Separation means 80 Mask generation means 82 Blur processing means 84 Outline extraction means 86 Outline enhancement means

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI theme coat ゛ (Reference) H04N 5/262 H04N 1/40 101D 5C077 5/91 5/91 HF term (Reference) 2H110 BA16 CB27 CB31 CD05 5B057 BA02 BA30 CA01 CA08 CA12 CA16 CB01 CB08 CB12 CB16 CC03 CE04 CE06 CE16 CH09 CH18 DA08 DA17 DB02 DB06 DB09 DC04 DC16 DC36 5C023 AA06 AA07 AA31 BA03 BA13 CA08 DA04 DA08 EA02 EA10 EA14 5C05 LA03 A01 GA03 AA32 AA36 AA40 BA05 BA06 CA10 5C077 LL19 MP07 MP08 PP02 PP03 PP21 PP27 PP28 PP32 PP34 PP46 PP47 PP55 PP58 PP60 PP61 PP68 PQ08 PQ20 RR14 TT04 TT09

Claims (10)

[Claims] A calculating unit configured to divide an image into regions so that adjacent pixels having a data difference between pixels equal to or less than a threshold value are included in the same region, and calculate a feature amount for each region; Area estimating means for estimating a background area or a main part area from the divided areas based on the feature amount calculated by the above-described method; and selectively selecting only the background area estimated by the area estimating means from the entire image. An image processing apparatus comprising: a blur processing unit that performs a blur process or selectively performs a blur process only on an area excluding a main part area. 2. The image processing apparatus according to claim 1, wherein the blur processing unit includes: a separating unit configured to separate the entire image into a background region and a region excluding the background region, or a main unit region and a region excluding the main unit region; and a background region separated by the separating unit. A mask generating means for generating a mask in an area or a main part area excluding the above, and a processing means for performing a blurring process on an area other than the area of the mask generated by the mask generating means. Item 2. The image processing device according to Item 1. 3. The region estimating means sets at least one of the four corners of the image as a background region, and when the characteristic amount of the background region and the characteristic amount of an adjacent region are equal to or smaller than a predetermined threshold value. The image processing apparatus according to claim 1, wherein, in a certain case, the adjacent area is integrated as a background area. 4. The region estimating means, when the area of the divided region is equal to or larger than a predetermined first value or equal to or smaller than a predetermined second value. The image processing apparatus according to any one of claims 1 to 3, wherein is set as a background area. 5. The area estimating means sets an area in which the average coordinate position of the pixels of the area-divided area is located in an area outside a circle or an ellipse smaller than the image based on the center of the image as a background area. 5. The method according to claim 1, wherein:
The image processing apparatus according to claim 1. 6. The mask generation unit according to claim 2, wherein the mask generation unit adjusts the outline of the mask by extracting an outline in the image and overlapping the outline with the outline of the main region. The image processing apparatus according to claim 1. 7. The image processing apparatus according to claim 1, wherein the calculating unit emphasizes an outline in the image, and performs the area division based on the emphasized outline. The image processing apparatus according to any one of the preceding claims. 8. The image processing apparatus according to claim 1, further comprising: a blur processing strength designating unit for designating a strength of the blur processing. 9. The image processing apparatus according to claim 1, further comprising a switching unit that switches whether or not to perform the blurring processing. 10. An image is divided into regions so that adjacent pixels having a data difference between pixels equal to or less than a threshold value are included in the same region, and a feature amount of each of the divided regions is calculated. A background region or a main portion region is estimated from the divided regions based on the feature amount, and only the background region estimated from the entire image is selectively subjected to blur processing or a region excluding the main portion region. An image processing method characterized by selectively performing a blurring process only.